HCI Bibliography Home | HCI Conferences | MUSIC Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
MUSIC Tables of Contents: 13

MUSIC 2013: Mobile, Ubiquitous, and Intelligent Computing 2013-09-04

Fullname:MUSIC 2013: Mobile, Ubiquitous, and Intelligent Computing
Editors:James J. (Jong Hyuk) Park; Hojjat Adeli; Namje Park; Isaac Woungang
Location:Gwangju, Korea
Dates:2013-Sep-04 to 2013-Sep-06
Publisher:Springer Berlin Heidelberg
Series:Lecture Notes in Electrical Engineering 274
Standard No:DOI: 10.1007/978-3-642-40675-1 hcibib: MUSIC13; ISBN: 978-3-642-40674-4 (print), 978-3-642-40675-1 (online)
Papers:95
Pages:651
Links:Online Proceedings | Conference Website
  1. Active Media Technologies (AMT)
  2. Computational Awareness for Telecommunication/Energy-Efficient Systems
  3. Human Centric Computing and HCI
  4. Portable and Smart Devices
  5. Security and Monitoring of WSN
  6. Security, Privacy, Authentication, Trust for IoT
  7. Ubiquitous Context Awareness
  8. WSN Applications and Technologies
  9. Data-Intensive Intelligence and Knowledge
  10. Data Intensive Computing and Applications
  11. Multimedia Cloud Computing and Its Applications
  12. Mobile Computing
  13. Ubiquitous Computing
  14. Intelligent Computing
  15. Intelligent and Mobile Services
  16. 3D Converged IT and Optical Communications
  17. Frontier Computing -- Theory, Technologies and Applications

Active Media Technologies (AMT)

A Novel Ranking Technique Based on Page Queries BIBAKFull-Text 1-5
  Gwangbum Pyun; Unil Yun
Keyword-based information retrieval finds webpages with queries composed of keywords to provide users with needed information. However, since the keywords are only a part of the necessary information, it may be hard to search intended results from the keyword-based methods. Furthermore, users should make efforts to select proper keywords many times in general because they cannot know which keyword is effective in obtaining meaningful information they really want. In this paper, we propose a novel algorithm, called PQ_Rank, which can find intended webpages more exactly than the existing keyword-based ones. To rank webpages more effectively, it considers not only keywords but also all of the words included in webpages, named page queries. Experimental results show that PQ_Rank outperforms PageRank, a famous algorithm used by Google, in terms of MAP, average recall, and NDCG.
Keywords: Information retrieval; Page query; Grouping webpages; Ranking technique
Ranking Book Reviews Based on User Discussion BIBAKFull-Text 7-11
  Heungmo Ryang; Unil Yun
Ranking algorithm is one of the most important issues in information retrieval researches. It can be divided into two types according to purpose, general and specific purpose algorithms. Although the general purpose algorithms can be effective for retrieving relevant documents on the internet, it is difficult to find meaningful information of specific targets such as blogs or twitter since the algorithms do not consider unique characteristics of the targets. Recently, ranking algorithms have been proposed for searching useful book reviews by reflecting unique features of book reviews. In this paper, we propose a novel algorithm, RUD (Ranking based on User Discussion), for ranking book reviews to improve performance based on user discussion. For performance evaluation, we conduct precision and recall tests. The experimental results show that the proposed algorithm outperforms previous algorithms.
Keywords: Book review; Information retrieval; Ranking; User discussion
The Blog Ranking Algorithm Using Analysis of Both Blog Influence and Characteristics of Blog Posts BIBAKFull-Text 13-17
  Jiwon Kim; Unil Yun
In recent years, while amounts of the information in the blogosphere increase rapidly, the problems of information quality have come up. Discovering a good quality data is the important issue in blog space with overflowed information. In this paper, we present WCT algorithm for efficient blog ranking. This method performs a ranking process using both interconnection of blogs and structural weights for content in blog. In the performance analysis, we discuss the comparison between our algorithm and the previous algorithm for blog ranking. The result shows that our proposal has the high performance than other blog retrieval method.
Keywords: Blog ranking; Information retrieval; Blog structure
Frequent Graph Mining Based on Multiple Minimum Support Constraints BIBAKFull-Text 19-23
  Gangin Lee; Unil Yun
Frequent graph mining allows us to find useful frequent sub-graph patterns from large and complicated graph databases. In a lot of real world applications, graph patterns with relatively low supports can be used as meaningful information. However, previous methods based on a single minimum support threshold have trouble finding them. That is, they cause "rare item problem", which means that useful sub-graphs with low supports cannot be extracted when a minimum support threshold is high, while an enormous number of patterns have to be mined to obtain these useful ones when the value is low. To overcome this problem, we propose a novel algorithm, FGM-MMS (Frequent Graph Mining based on Multiple Minimum Support constraints). After that, we demonstrate that the suggested algorithm outperforms a state-of-the-art graph mining algorithm through comprehensive performance experiments.
Keywords: Frequent graph mining; Multiple minimum supports; Sub graph
Design of Automatic Paper Identification System with QR Code for Digital Forensics BIBAKFull-Text 25-30
  Ha-Kyung Jennifer Lee; Young-Mi Yun; Kee-Hyung Yoon; Dong-Sub Cho
As the printing technology is developing, the use of digital documents which is printed with a variety of printing papers is increasing. And the increase in criminal forgery of digital documents has been the cause of many social problems. As a result, the development of digital forensic techniques to collect and analyze the evidence of the crime of forgery of digital documents is increasingly important. In this paper, we propose an automatic paper identification technique with QR code that is formed based on the extracted features from each paper's microscope image.
Keywords: Printing Paper; Paper Id; Identification System; QR code; Digital Forensics; Digital Document

Computational Awareness for Telecommunication/Energy-Efficient Systems

Processing Continuous Range Queries with Non-spatial Selections BIBAKFull-Text 31-38
  HaRim Jung; Seongkyu Kim; Joon-Min Gil; Ung-Mo Kim
In this paper, we explore the problem of scalable evaluation of Continuous Range Queries (CRQs) with non-spatial selections, each of which continually retrieves the moving objects that (i) are currently within a specified spatial query region and (ii) satisfy specified non-spatial selections. We propose a new query indexing structure, called the Bit-vector Query Region tree (BQR-tree), which enables the server to cooperate with moving objects for evaluation of CRQs with non-spatial selections. Through simulations, we verify the efficiency of the BQR-tree.
Keywords: Continuous range queries; moving objects; index structures
DSPI: An Efficient Index for Processing Range Queries on Wireless Broadcast Stream BIBAKFull-Text 39-46
  Kwanho In; Seongkyu Kim; Ung-Mo Kim
This paper addresses the problem of processing range queries on wireless broadcast streams. In order to support range queries efficiently, we propose a novel index called Distributed Space-Partitioning Index (DSPI). DSPI consists of hierarchical grids that provide mobile clients with the global view as well as the local view of the broadcast data. The algorithm for processing range queries based on DSPI is also proposed. Simulation experiments demonstrate DSPI is superior to the existing index schemes.
Keywords: Continuous range queries; moving objects; index structures
End-to-End High Speed Forward Error Correction Using Graphics Processing Units BIBAKFull-Text 47-53
  Md Shohidul Islam; Jong-Myon Kim
Forward error correction (FEC) is an efficient error recovery mechanism for wireless networks in which erroneous packet is corrected in the destination node. More importantly, real-time and high-speed wireless networks require fast error recovery to ensure quality of service (QoS). Since graphics processing units (GPUs) offer massively parallel computing platform, we propose a GPU-based parallel error control mechanism using extended Hamming code supporting single-bit as well as multiple-bit error correction. We compare the performance of the proposed GPU-based approach with the equivalent sequential algorithm that runs on the traditional CPU for error strength, t, such that 1 ≤ t ≤ 7. Experimental results demonstrate that the proposed GPU-based approach outperforms the sequential approach in terms of execution time. Moreover, the proposed parallel implementation yields significant reduction in computational complexity from O(n³) of the sequential algorithm to O(n) of the GPU-based approach, leading to tremendous speedup gain.
Keywords: Real-time wireless communication; multiple bit error FEC; extended Hamming code; GPU

Human Centric Computing and HCI

DirectSpace: A Collaborative Framework for Supporting Group Workspaces over Wi-Fi Direct BIBAKFull-Text 55-61
  Jong-Eun Park; Jongmoon Park; Myung-Joon Lee
Wi-Fi Direct is a new feature in the 4.0 version of the Android operating system, allowing devices to connect directly to each other via Wi-Fi without an intermediate access point. Based on the Wi-Fi Direct feature, in this paper, we present a collaborative framework named DirectSpace that supports various collaborative situations to resolve the problems of traditional application paradigm such as client-server model. For this, we design collaborative services to support various environments, and develop collaborative protocols for providing workspaces over Wi-Fi Direct. On the top of DirectSpace, we also present a collaborative application for utilizing the group workspaces.
Keywords: Collaborative framework; Wi-Fi Direct; DirectSpace; workspace
Specification of Communication Based Train Control System Using AADL BIBAKFull-Text 63-68
  Lichen Zhang; Bingqing Xu
The development of railway cyber physical systems is a challenging process. In this paper we present our current effort to extend AADL to include new features for separation of concerns of railway cyber physical systems, we extend AADL in spatial aspect, dynamic continuous aspect, physical world modeling aspect. Finally, we illustrate the proposed method via an example of specification of communication based train control system.
Keywords: AADL Railroad system; Train control system; component; OSATE
An Agent Modeling for Overcoming the Heterogeneity in the IoT with Design Patterns BIBAKFull-Text 69-74
  Euihyun Jung; IlKwon Cho; Sun Moo Kang
The Internet of Things (IoT) has been considered as a core infrastructure that provides the connectivity to anyone, anywhere, anytime and especially anything. Due to this advantage, the IoT is expected to change the whole society and to enrich people's everyday life, but there are a lot of technical issues in realizing the IoT. Among them, the heterogeneity is an urgent and essential issue that cannot be easily resolved. In this paper, we described an agent modeling that can hide the heterogeneity of devices using the Strategy, Dependency Injection, and Reflection design patterns. The designed agent was implemented as the agent system named iSilo and various devices were developed and bound to the agents in the iSilo. Several experiments were conducted in Korea and Japan and these evaluations showed the proposed modeling could be a novel solution to overcome the heterogeneity in the IoT.
Keywords: IoT; Agent; Design Patterns
BK-means Algorithm with Minimal Performance Degradation Caused by Improper Initial Centroid BIBAKFull-Text 75-80
  Hoon Jo; Soon-cheol Park
K-means algorithm has the performance degradation problem due to improper initial centroids. In order to solve the problem, we suggest BK-means (Balanced K-means) algorithm to cluster documents. This algorithm uses the value, α, to adjust each cluster weight which is first defined in this paper. We compared the algorithm to the general K-means algorithms on Reutor-21578. The experimental results show about 11% higher performance than that of the general K-means algorithm with the balanced F Measure (BFM).
Keywords: Clustering; Information Retrieval; K-means; BK-means; Outlier

Portable and Smart Devices

Detect Spatial and Temporal Gait Parameters by Dual Accelerometers BIBAKFull-Text 81-85
  Wann-Yun Shieh; An-Peng Liu; Tyng-Tyng Guu
The process of walking or running is called the "gait". In clinical research, gait detection can be used to investigate the features of normal or abnormal gait for demonstrating a change from treatment or from disease progression. In the past, many optical-based gait detection approaches have been proposed. In these approaches, we have to paste many reflective markers on the subject's limbs and use multiple cameras from different directions to take the images of walking. They can provide high accuracy measurements for gait detection, but they also need very expensive optical equipment. Also, the experiments are restricted to the laboratory environment, which means that the collection of gait data will be limited in a short distance or a short time interval. In this paper we will propose a portable design, which uses dual accelerometers pasted on a subject's left and right waist to do the gait detection at any time, any place. Particularly, we will apply the wireless communication to develop a gateway, as well as its App on the smart phone, to collect sensing data. The data collected from the sensors can be uploaded to the remote cloud for many telemedicine applications.
Keywords: Gait detection; accelerometers; telemedicine applications
Implementation of Load Management Application System in Energy Management Service BIBAKFull-Text 87-92
  Taekyeong Kang; Hyungkyu Lee; Dong-Hwan Park; Hyo-Chan Bang; Namje Park
As the Smart grid is intelligent power grid, combining information Technology to the existing power grid. Electricity suppliers and consumers exchange real-time information to two-way and is a next-generation power grid to optimize energy efficiency. This paper suggests the implementation of load management application system in energy management service environment.
Keywords: Energy Management; Load Management; Smart Socket; Energy
Toward a Mobile Application for Social Sharing Context BIBAKFull-Text 93-98
  Meng-Yen Hsieh; Ching-Hung Yeh; Yin-Te Tsai; Kuan-Ching Li
Due to wireless and sensing technologies powerful in smartphone, a number of smartphone applications, a.k.a APPs. have combined social sharing mechanisms. This paper defines social sharing contexts on a social framework suitable to APPs. A tourism APP based on the sharing mechanisms is implemented to include social behavior for data sharing, while smartphone have supported various wireless technologies. Besides, smartphone as a hand-held device is hold by users so that various gestures with user hands are adaptive to handle the key features of sharing data between smartphones.
Keywords: social sharing; peer-to-peer; social network media
A Research Based on the Effect of Smart Phone Use on Consumption Life of Teenagers in a Smart Era BIBAKFull-Text 99-104
  Jeonghan Son; Keon Uk Kim; Yeon-gyeong Seo; Wonyeong Oh; Seowon Choi; Ana Kang
An advent of the smart era, including a smart phone, tablet, PC, online community, etc, brings a groundbreaking change in everyday life. Internet connection is possible at any time. Also, smart era, including smart home, smart building, smart city, etc, is beyond a simple technology, and brings an enormous change in a way of communication in daily life and even economic life. This smart era also brings a big change in the life of teenagers. This research points out that the economic value consumed in return for the convenience by teenagers in the smart era is increased, and their consumption becomes unwise and unreasonable. This research finds out the problem of teenagers' consumption activity in the smart era and proposes a solution to improve the problem.
Keywords: Smartphone; consumption; teenager

Security and Monitoring of WSN

Quality-Workload Tradeoff in Pig Activity Monitoring Application BIBAKFull-Text 105-110
  Haelyeon Kim; Yeonwoo Chung; Sungju Lee; Yongwha Chung; Daihee Park
Generally, there is a tradeoff between quality and computational workload required to obtain that quality. In this paper, we focus on practical issues in implementing a pig activity monitoring system. We first propose a method for evaluating the quality-workload tradeoff in the activity monitoring application. Then, we derive the cost-effective solution within the acceptable range of quality for the activity monitoring application. Based on the experiments with the video monitoring data obtained from a pig farm, our method can derive the cost-effective resolution size and frame rate without degrading the accuracy significantly.
Keywords: Activity Monitoring; Quality; Accuracy; Workload; Tradeoff

Security, Privacy, Authentication, Trust for IoT

Applying Different Cryptographic Algorithms for Mobile Cloud Computing BIBAKFull-Text 111-116
  Sung-Min Jung; Nam-Uk Kim; Seung-Hyun Lee; Dong-Young Lee; Tai-Myoung Chung
Cloud computing is new paradigm to use computing resources that are delivered as services over a network. It optimizes the usage of IT resources such as CPU, storage, and network. Many services related of cloud computing are popular to end users and it is becoming more important these days. There are many smart phones, smart pads and other mobile devices and end users can access to cloud computing environment through these mobile devices. Thus, they can use powerful computing resources on their physical devices. This environment indicates mobile cloud computing in this paper. There are two devices such as a physical device in real world and a virtual device in cloud computing environment. Service providers should use strong cryptographic algorithms to guarantee secure communication between a physical device and a virtual device. However, the strong cryptographic algorithms waste time to process each tasks and it causes network congestion. The network congestion occurs when a physical device is processing too many data packets. Also, it cannot be guaranteed its network quality of service. We need to consider the network quality of service to avoid this congestion. We should try to reduce the execution time to guarantee quality of service. We propose suitable method that the cryptographic algorithms with different key lengths at various environment.
Keywords: Cloud computing; QoS management; Cryptographic algorithm
Intrusion-Tolerant Jini Service Architecture for Ensuring Survivability of U-Services Based on WSN BIBAKFull-Text 117-124
  Sung-Ki Kim; Jae-Yeong Choi; Byung-Gyu Kim; Byoung-Joon Min
U-Service environment based on WSN (Wireless Sensor Network) is poor in reliability of connection and has a high probability that the intrusion and the system failure may occur. Therefore, it is very important to ensure that the legitimate users make use of trustable services without discontinuance or obstacle of the services they are enjoying despite the presence of failures and intrusions. In this paper, we propose an intrusion-tolerant Jini service architecture integrating security and survivability mechanisms in order to provide end users with Jini services having a persistent state in wireless sensor networks. The proposed architecture is able to protect a Jini system not only from faults such as network partitioning or server crash, but also from attacks exploiting flaws. It is designed to provide performance enough to show a low response latency so as to support seamless service usage. Through the experiment on a test-bed, we have confirmed that the architecture is able to provide high security and availability at the level that the degradation of services quality is ignorable.
Keywords: Intrusion-tolerance; Jini; Apache River; Jgroup/ARM; Security
Creation Mechanism for Access Group Based on User Privacy Policy-Based Protection BIBAKFull-Text 125-130
  Taekyeong Kang; Hyungkyu Lee; Dong-Hwan Park; Hyo-Chan Bang; Namje Park
Smart grid is the next-generation intelligent power network which optimizes energy efficiency through the mutual real time exchange of information between power supplier and consumer through the integration of existing power network and the information technology (IT). However, the smart grid environment can have problems involved with personal privacy invasion. This paper suggests the creation mechanism of privacy policy-based protection system.
Keywords: Smart Grid Security; Privacy Policy; Privacy Protection; Access control; Privacy Exchange Format

Ubiquitous Context Awareness

Specification of Train Control Systems Using Formal Methods BIBAKFull-Text 131-136
  Bingqing Xu; Lichen Zhang
Just as what the public pursue, we need a much safer railway system with a higher level of automation in control. To achieve this goal, the author aims to specify the Train Control System by formal methods which can specify the communication of various processes in the system clearly. This paper applies Timed-CSP which concerns the time-delay to the specification of the control flow and communication among flows in Train Control System, and specifies the state and data change by Object-Z. By Timed-CSP and Object-Z, the specification of a simplified Train Control System especially the time constraints is more concrete.
Keywords: Timed-CSP; Object-Z; specification; control and sensor
Formal Descriptions of Cyber Physical Systems Using Clock Theory BIBAKFull-Text 137-142
  Bingqing Xu; Lichen Zhang
Cyber Physical Systems are in charge of the control of physical processes characterized by their own dynamics. This control must comply with timing constraints -- sometimes stringent ones -- imposed by the Cyber Physical Systems. It is crucial to address these timing issues as early as possible in the development process to detect inconsistencies in the requirements or in the constraints and to capture changes in the system. This paper aims to apply the clock theory to the specification of Cyber Physical Systems. To illustrate the concept we develop a well-known case study: the Steam Boiler Control System.
Keywords: Cyber Physical Systems; continuous; discrete; clock theory; time analysis
An Intelligent Dynamic Context-Aware System Using Fuzzy Semantic Language BIBAKFull-Text 143-149
  Daehyun Kang; Jongsoo Sohn; Kyunglag Kwon; Bok-Gyu Joo; In-Jeong Chung
The prevalence of smart devices and the wireless Internet environment have enabled users to exploit environmental sensor data in a variety of fields. This has engendered various research issues in the development of context-awareness technology. In this paper, we propose a novel method where semantic web technology and the fuzzy concept are used to perform tasks that express and infer the user's dynamic context, in distributed heterogeneous computing environments. The proposed method expresses environmental information using numerical values, and converts them into fuzzy OWL. Then, we make inferences based on the user context, using FiRE, a fuzzy inference engine. The suggested method allows us to describe user context information in heterogeneous environments. Because we use fuzzy concepts to represent contextual information, we can easily express its degree or status.
Keywords: Context-aware computing; Fuzzy; Knowledge Representation; Inference; Fuzzy Web Ontology Language (OWL)

WSN Applications and Technologies

Efficient Data Monitoring in Sensor Networks Using Spatial Correlation BIBAKFull-Text 151-156
  Jun-Ki Min
In order to reduce r the energy consumption of sensors, we present an approximate data gathering technique, called CMOS, based on the Kalman filter. The goal of CMOS is to efficiently obtain the sensor readings within a certain error bound. In our approach, spatially close sensors are grouped as a cluster. Since a cluster header generates approximate readings of member nodes, a user query can be answered efficiently using the cluster headers. Our simulation results with synthetic data demonstrate the efficiency and accuracy of our proposed technique.
Keywords: sensor network; data monitoring; Kalman filter
Power-Time Tradeoff of Parallel Execution on Multi-core Platforms BIBAKFull-Text 157-163
  Sungju Lee; Heegon Kim; Yongwha Chung
It is anticipated that high-performance handheld multi-core devices will be used as WSN processing nodes in the near future. Reducing execution time by deploying parallel applications on multi-core platforms comes at the cost of increasing power consumption compared to using fewer cores. This paper focuses on such tradeoff between power consumption and execution time and subsequently achieves maximal energy saving when executing applications in parallel. Based on the experiments on a multi-core platform, we can verify that parallel execution with frequency scaling is an effective approach at the application level in order to reduce energy consumption.
Keywords: multi-core platform; energy saving; parallel application
Effective Object Identification through RFID Reader Power Control BIBAFull-Text 165-171
  Shung Han Cho; Sangjin Hong; Nammee Moon
In this paper, we present a RFID reader scheduling strategy for an effective identification by varying coverage through power control. A RFID reader is incorporated into a surveillance system which tracks objects with visual sensors. A required separation between objects is defined to avoid group identification with multiple objects. In order to reduce the power consumption of a RFID reader, the power cost is simply modeled with object positions to determine the activation time and the range of a RFID reader. The power cost model also considers the effect of added power consumption establishing group identification. A RFID reader is activated when the estimated power cost for the current sampling time is smaller than the estimated power cost for the next sampling time. The simulation results demonstrate that the proposed method reduces the power consumption with the effective object identification.
Market-Based Resource Allocation for Energy-Efficient Execution of Multiple Concurrent Applications in Wireless Sensor Networks BIBAKFull-Text 173-178
  Mo Haghighi
Many engineering disciplines have become reliant on WSNs in order to detect and track certain events of interest by monitoring various variables, through a number of specially distributed wireless sensors. Due to resource constraints of sensor hardware, traditional WSN applications involved exchanging an excessive amount of data, usually in an offline mode, between sensor nodes and a central unit, in order to apply computational analysis on the captured data. New sensor devices however, are equipped with more powerful resources and capable of running multiple concurrent processing, and applying computational data analysis can be implemented online and often in a distributed fashion. In this paper we will investigate the application of market-based algorithms for energy management, tasks allocation and resource coordination in WSNs with multiple concurrent applications. We will also propose a number of algorithms for calculating costs and utilities for multi-paradigm application requirements.
Keywords: Market-based; Auction-based; WSN; Sensor; Utility; Sensomax; Concurrency
Clustering Objects in Heterogeneous Information Network Using Fuzzy C-Mean BIBAFull-Text 179-184
  Muhammad Shoaib; Wang-Cheol Song
In this paper we have proposed a fuzzy c-mean based clustering algorithm for categorization of different types of objects present in a heterogeneous information network. We have addressed a particular scenario in this paper when exact structure of objects and their relationships with other objects is either hidden or not known. We have performed the experiments on an agriculture information network and our results depicts that combining automatic extraction of structure of an information network with information objects can improve the quality of clustering.

Data-Intensive Intelligence and Knowledge

Better Induction Models for Classification of Forest Cover BIBAKFull-Text 185-189
  Hyontai Sug
The data set of forest cover types based on cartographic data consists of very large data set of 581,102 instances. So, decision tree-based data mining methods that need relatively less computing resources could be used for better classification models. Random forests consisting of multitude of special decision trees are known to be a good data mining tool, and a technique based on grid search of random forests was investigated to find very accurate classifier. Experiments showed that a classifier of high accuracy could be found for the data set of forest cover types.
Keywords: Data mining; forest cover type; decision trees; random forests
Application for Temporal Analysis of Scientific Technology Information BIBAKFull-Text 191-195
  Myunggwon Hwang; Do-Heon Jeong; Jinhyung Kim; Jangwon Gim; Sa-kwang Song; Sajjad Mazhar; Hanmin Jung; Shuo Xu; Lijun Zhu
In recent, business intelligence becomes one of important issues due to various analyses on technology trends. Especially, understanding the relations and influences between technologies is core property for the high-performed analysis. To do this, a few works have utilized ontologies constructed automatically but still have many errors and it causes difficulty while interpreting technology trends. Therefore this paper introduces an application which visualizes relationships and influences between technologies according to time series. Our application provides clues for intuitive observations of relationship change between technologies.
Keywords: Technology information; temporal analysis; business intelligence; co-occurrence
ROI Extraction in Dermatosis Images Using a Method of Chan-Vese Segmentation Based on Saliency Detection BIBAKFull-Text 197-203
  Zehan Wang; Lijun Zhu; Jiandong Qi
Extraction of ROI (Region-Of-Interest) in dermatosis images can be used in content-based image retrieval (CBIR). Image segmentation takes an important part in it. And the performance of the segmentation algorithm directly influences the efficiency of the ROI extraction results. In this paper, a method of Chan-Vese segmentation based on saliency detection to extract the ROI of the dermatosis images is proposed. Firstly the spectral residual approach (SR) [11] is used to get the saliency map of the dermatosis images. Secondly threshold segmentation is used to get the initial ROI images. Finally the Chan-Vese model is used to segment the initial ROI images to get the final ROI images, which can ensure the active contours evolve close to the object and remove the redundant information from the complex background. The experiment results show that the proposed method has the better performance than only using Chan-Vese method.
Keywords: saliency detection; Chan-Vese model; dermatosis images; ROI extraction
The Study on Semantic Self-sufficiency in Factual Knowledge Extraction BIBAFull-Text 205-209
  Yunliang Zhang
In this paper, semantic self-sufficiency of elements in factual knowledge extraction is introduced to meet the real knowledge service demands. The characters of semantic self-sufficiency are analyzed, and based on which the processing strategies are proposed. Though there are still some difficulties to cope with, semantic self-sufficiency is a tool to clearly define the elements of factual knowledge and can improve the results of automatic factual knowledge extraction.
XML-Based Document Retrieval in Chinese Diseases Question Answering System BIBAKFull-Text 211-217
  Haodong Zhang; Lijun Zhu; Shuo Xu; Weifeng Li
A Chinese Diseases Question Answering System (Hestia QA) is being developed by ISTIC. As a part of Hestia QA, a XML-based document retrieval and similarity calculation model is established here. The texts which describe diseases in Chinese are indexed and wrapped in XML tags. The query is compared with related tags in XML document and the similarity is calculated with a deformed cosine similarity algorithm. The Chinese terms semantic similarity calculation algorithm is used to get the similarity of two terms in the system. The result shows that our model works well. The Chinese disease XML datasets will be analyzed in different granularity levels or dimensions. The corpus of diseases in Chinese will be established after the automatic XML annotation software is completed in the next step.
Keywords: ML; Chinese Diseases QA; Chinese Terms Similarity; Cosine Similarity
Mathematical Document Retrieval Model Using Structural Information of Equations in Pseudo-documents BIBAKFull-Text 219-223
  Yeongkil Song; Junsoo Shin; Harksoo Kim
Math-aware search engines are required to effectively retrieve mathematical documents including various equations. In this paper, we propose a mathematical document retrieval system by which users can retrieve documents using any combination of keywords and equations. The proposed system indexes equations and their surrounding keywords from mathematical documents. Then, it searches and ranks mathematical documents using a language model modified for the heterogeneous indexing units (i.e., mixtures of equations and keywords). In the experiments, the proposed system performed well, especially for high ranks.
Keywords: Mathematical Document Retrieval; Heterogeneous Indexing Term; Pseudo-document
Lexical Feature Extraction Method for Classification of Erroneous Online Customer Reviews Based on Pattern Matching BIBAKFull-Text 225-229
  Maengsik Choi; Junsoo Shin; Harksoo Kim
In morpheme-based languages such as Korean and Japanese, spacing and spelling errors that frequently occur in online documents make it difficult to reliably extract informative lexical clues for sentiment analysis. To overcome this problem, we propose a simple, reliable lexical feature extraction method for sentiment classification systems; this method targets online customer reviews in Korean, which include numerous spacing and spelling errors. The proposed method performs longest-matching between input sentences and two kinds of patterns (spacing-unit patterns and phoneme patterns) that are automatically constructed from a large POS tagged corpus. Thereafter, the method returns content words associated with the longest matched patterns. In the experiments on sentiment classification, the proposed method outperformed previous lexical feature extraction methods, which are based on conventional morphological analyzers.
Keywords: Lexical feature extraction; spacing and spelling errors; spacing-unit pattern; phoneme pattern
Unified Concept Space and Mapping Discovery Algorithm for Heterogeneous Knowledge Systems BIBAKFull-Text 231-237
  Lijun Zhu; Chen Shi; Jianfeng Guo
For heterogeneous scientific and technical knowledge systems (HSTKSs), the computer-aided concept-mapping discovery becomes very difficult among the HSTKSs. First, this paper puts forward and establishes a public concept model oriented to the HSTKS used for the standardized description of scientific and technological knowledge concepts. Then, in the mapping discovery algorithm, the algorithms to discover the relations of inheritance, "is a characteristic of", "is a part of", relevance and other partial ordering relations between heterogeneous concepts are put forward and designed through mapping transfer. Finally, the empirical results show that, in public concept space, the mapping discovery algorithm put forward and designed by this paper, is feasible and have certain practical significance.
Keywords: Knowledge System; Concept Space; Mapping Discovery; Mapping Transfer
Author-Topic over Time (AToT): A Dynamic Users' Interest Model BIBAKFull-Text 239-245
  Shuo Xu; Qingwei Shi; Xiaodong Qiao; Lijun Zhu; Hanmin Jung; Seungwoo Lee; Sung-Pil Choi
One of the key problems in upgrading information services towards knowledge services is to automatically mine latent topics, users' interests and their evolution patterns from large-scale S&T literatures. Most of current methods are devoted to either discover static latent topics and users' interests, or to analyze topic evolution only from intra-features of documents, namely text content without considering directly extra-features of documents such as authors. To overcome this problem, a dynamic users' interest model for documents using authors and topics with timestamps is proposed, named as Author-Topic over Time (AToT) model, and collapsed Gibbs sampling method is utilized for inferring model parameters. This model is not only able to discover latent topics and users' interests, but also to mine their changing patterns over time. Finally, the extensive experimental results on NIPS dataset with 1,740 papers indicate that our AToT model is feasible and efficient.
Keywords: Author-Topic (AT) Model; Topic over Time (ToT) Model; Author-Topic over Time (AToT) Model; Dynamic Users' Interest Model; Collapsed Gibbs Sampling
Scalable RDF Path Query Processing Based on Runtime Class Path Lookup Scheme BIBAKFull-Text 247-251
  Sung-Jae Jung; Dong-min Seo; Seungwoo Lee; Hanmin Jung
With the rapidly growing amount of information represented in RDF format, efficient querying RDF graph has become a fundamental challenge. There have been several relationship finding services based on querying RDF database to discover relationships between two objects of interest. Conventional relationship-finding service requires computationally expensive graph search operations which involve multiple self joins. It becomes even more challenging when the graph data is large and diverse. In this paper we propose an algorithm which uses RDF schema information for efficient RDF path query processing. By utilizing the pre-calculated class path expressions, the graph search space is significantly reduced. Compared with the conventional BFS algorithm, the proposed algorithm (bidirectional BFS combined with class path lookup approach) achieves performance improvement by 3 orders of magnitude. Additionally, the proposed algorithm is scalable, because it operates based on B-Tree index when it accesses to triple repository and pre-calculated class path information. Thus, the proposed algorithm is expected to return graph search results within a reasonable response time on even much larger RDF graph.
Keywords: RDF schema; path expression; SQL based graph search; RDF path query; class path pre-calculation; bidirectional Breadth First Search
Risk Aversion Parameter Estimation for First-Price Auction with Nonparametric Method BIBAKFull-Text 253-260
  Xin An; Jiancheng Chen; Yuan Zhang
More and more clues show that the bidders tend to risk averse. However, traditional nonparametric approach is only applicable for the case of risk neutrality. This study proposes a generalized nonparametric structural estimation procedure for the first-price auctions. To evaluate the performance, extensive Monte Carlo simulation experiments are conducted for ten different values of risk aversion parameter including the risk neutrality case in multiple classic scenes. Though there are no unique estimators for risk aversion parameter, four (weighted) combinations of all estimators are obtained, and some guidance is also given for real-world applications. Finally, empirical results on USFS bidding dataset show that the our nonparametric method can capture bidders' risk aversion to some extend.
Keywords: Risk Aversion; First-Price Auction; Keyword Auction; Monte Carlo; Nonparametric Method
Diverse Heterogeneous Information Source-Based Researcher Evaluation Model for Research Performance Measurement BIBAKFull-Text 261-266
  Jinhyung Kim; Myunggwon Hwang; Do-Heon Jeong; Sa-kwang Song; Jangwon Gim; Hanmin Jung; Shuo Xu; Lijun Zhu
Analysis, prediction, and recommendation of information about experts are very important tasks for future research planning and strategy establishment. However, it takes much time and efforts even for precise analysis of experts because we need to analyze huge and diverse heterogeneous information. There are several application and tools for supporting analysis about researchers, but they provides fragmentary analysis result based on simple evaluation criteria. Therefore, in this paper, we suggest new researcher evaluation model based on diverse performance evaluation features, named RSW model. By using RSW model, we can analyze and compare researchers in various perspectives. In addition, we can ranked researchers and recommend outstanding collaborator in a specified research field.
Keywords: Researcher Performance Evaluation; Evaluation Features; Researcher Evaluation Model; Performance Measurement
Fast Big Textual Data Parsing in Distributed and Parallel Computing Environment BIBAKFull-Text 267-271
  Jung-Ho Um; Chang-Hoo Jeong; Sung-Pil Choi; Seungwoo Lee; Hanmin Jung
Currently, tremendous numbers of scientific and technical articles are being published due to the rapid development of the scientific and technical fields. Also, systems are being proposed which can give useful information to users by extracting information from scientific and technical articles. For such systems, we need to be able to extract information from a massive number of documents very fast and reliably. However, legacy parsers, such as Stanford, Enju and so on, cannot consider a large number of documents because such parsers analyze wide context range of the sentence for their parsing, and so those parsers require a lot of time to run. Therefore, in this paper, we report on the development of a parser which is based on MapReduce, a distributed and parallel programming model. Our parser has achieved about nineteen times better performance than that of one of the-state-of-the-art legacy parsers.
Keywords: distributed and parallel computing; big textual data; parsing; MapReduce
K-Base: Platform to Build the Knowledge Base for an Intelligent Service BIBAKFull-Text 273-277
  Sungho Shin; Jung-Ho Um; Sung-Pil Choi; Hanmin Jung; Shuo Xu; Lijun Zhu
Recently, there is an increasing interest in effectively using big data. It is also thought that the machine learning methods are crucial to effectively extract knowledge from big text data when they are coupled with big data technologies such as MapReduce and Hadoop. For tasks such as the knowledge extraction from huge amount of texts and the reasoning, it produces better results to simultaneously apply a machine learning method and big data technologies to the system. In this research, we propose a system using a machine learning method and big data technologies, and compare it with the existing system in terms of velocity and accuracy. The proposed system is expected to faster and more accurately build the knowledge base than the existing system.
Keywords: Distributed and parallel computing; Knowledge base; Machine learning; Knowledge extraction; Reasoning
A Novel Anomaly Detection System Based on HFR-MLR Method BIBAKFull-Text 279-286
  Eunhye Kim; Sehun Kim
Reducing the data space and then classifying anomalies based on the reduced feature space is vital to real-time intrusion detection. In this study, a novel framework is developed for logistic regression-based anomaly detection and hierarchical feature reduction (HFR) to preprocess network traffic data before detection model training. The proposed dimensionality reduction algorithm optimally excludes the redundancy of features by considering the similarity of feature responses through a clustering analysis based on the feature space reduced by factor analysis, thus helping to rank the importance of input features (essential, secondary and insignificant) with low time complexity. Classification of anomalies over the reduced feature space is based on a multinomial logistic regression (MLR) model to detect multi-category attacks as an outcome with the goal of reinforcing detection efficiency. The proposed system not only achieves a significant detection performance, but also enables fast detection of multi-category attacks.
Keywords: Anomaly detection; Dimensionality reduction; Hierarchical clustering; Multinomial logistic regression
Knowledge Discovery and Integration: A Case Study of Housing Planning Support System BIBAKFull-Text 287-291
  Junyoung Choi; Daesung Lee; Hanmin Jung
Different information related to knowledge should be collected and filtered by system in order to discover and integrate knowledge in multiple knowledge representation environment. Especially, various information exist in multiple forms and in separate spaces, such as geographic information, housing statistics, and policy statements, in housing planning domain. Housing policy support system is needed to make a proper spatial decision for supply housing when housing demand is occurred. Thus, this paper proposes a conceptual design of system for knowledge discovery and integration in housing planning support. This system uses GIS based Housing Demand and Supply Mapping Model (HDSMM) to support policy decision using housing statistics and geographic information and to be composed of multi-dimension analysis, policy monitoring functionalities.
Keywords: knowledge discovery; knowledge integration; housing supply; housing demand; GIS

Data Intensive Computing and Applications

Performance Analysis of MapReduce-Based Distributed Systems for Iterative Data Processing Applications BIBAKFull-Text 293-299
  Min Yoon; Hyeong-il Kim; Dong Hoon Choi; Heeseung Jo; Jae-woo Chang
Recently, research on big data has been actively made because big data are generated in various scientific applications, such as biology and astronomy. Therefore, distributed data processing techniques have been studied to manage the big data in large number servers. Meanwhile, some scientific applications like genome data analysis require loop control in analyzing big data using a MapReduce framework. In this paper, we first describe the existing MapReduce-based distributed systems which support iterative data processing. In addition, we do the performance analysis of the existing distributed systems in terms of execution time for various scientific applications which require iterative data processing. Finally, based on the performance analysis, we discuss some requirements for a new MapReduce-based distributed system which supports iterative data processing efficiently.
Keywords: Big data; MapReduce-based distributed systems; iterative data processing
A Semi-clustering Scheme for Large-Scale Graph Analysis on Hadoop BIBAKFull-Text 301-306
  Seungtae Hong; Youngsung Shin; Dong Hoon Choi; Heeseung Jo; Jae-woo Chang
With the evolution of IT technologies, large-scale graph data have lately become a growing interest. As a result, there are a lot of research results in large-scale graph analysis on Hadoop. The graph analysis based on Hadoop provides parallel programming models with data partitioning and contains iterative phases of MapReduce jobs. Therefore, the effectiveness of data partitioning depends on how the data partitioning maintains data locality in each node of cluster. In this paper, we propose a semi-clustering scheme for large-scale graph analysis such as PageRank algorithm on Hadoop and show that the proposed scheme is effective. With experiment results, PageRank computation with the semi-clustering improves the performance.
Keywords: large-scale graph analysis; semi-clustering; Hadoop; PageRank
Multi-stream Parallel String Matching on Kepler Architecture BIBAKFull-Text 307-313
  Nhat-Phuong Tran; Myungho Lee; Sugwon Hong; Dong Hoon Choi
Aho-Corasick (AC) algorithm is a commonly used string matching algorithm. It performs multiple patterns matching for computer and network security, bioinformatics, among many other applications. These applications impose high computational requirements, thus efficient parallelization of the AC algorithm is crucial. In this paper, we present a multi-stream based parallelization approach for the string matching using the AC algorithm on the latest Nvidia Kepler architecture. Our approach efficiently utilizes the HyperQ feature of the Kepler GPU so that multiple streams generated from a number of OpenMP threads running on the host multicore processor can be efficiently executed on a large number of fine-grain processing cores. Experimental results show that our approach delivers up to 420Gbps throughput performance on Nvidia Tesla K20 GPU.
Keywords: string matching; Kepler GPU; multi-stream; HyperQ; multithreading
Microscopic Bit-Level Wear-Leveling for NAND Flash Memory BIBAFull-Text 315-320
  Yong Song; Woomin Hwang; Ki-Woong Park; Kyu Ho Park
By microscopically observing widely used data files, we identified the considerable room for life time improvement in NAND flash memory, which is due to the discovery of a non-uniformity in bit-level data patterns. In an attempt to exploit the discovery, we propose a novel bit-level wear-leveling scheme. Instead of considering only the view of page-level or block-level, we incorporate the non-uniformity in data encoding patterns into wear-leveling scheme. Because of its orthogonality to the existing block-level wear-leveling approaches, our solution can be adopted over the existing solutions without considerable overhead and extend NAND flash's life span up to 36% in case of SLC.
HASV: Hadoop-Based NGS Analyzer for Predicting Genomic Structure Variations BIBAFull-Text 321-327
  Gunhwan Ko; Jongcheol Yoon; Kyongseok Park
The NGS technology produces large scale biologic data sets much cheaper and faster than the previous methods. As it is almost impossible to store or analyze such large scale NGS data with a traditional method on a commodity server, many problems arise. Hadoop is an alternative to this requirement. We aim to address the issues involved in the large scale data analysis on the cloud in bioinformatics. Accordingly, we propose analysis service for predicting genome structural variations associated with diseases by using Hadoop. The result of this study reveals that the system proposed in this study efficiently predicts genomic variations from large scale data sets.

Multimedia Cloud Computing and Its Applications

Provisioning On-Demand HLA/RTI Simulation Environment on Cloud for Distributed-Parallel Computer Simulations BIBAKFull-Text 329-334
  In-Yong Jung; Byong-John Han; Chang-Sung Jeong
In distributed parallel computer simulation, there are various simulation platforms supporting HLA on the Grid or Cloud, but almost of them shows lack of various aspects because of limitation of HLA. In this paper, we present an architecture of Cloud Distributed-Parallel Simulation Platform for HLA (CDSPH) supporting distributed parallel computer simulation on the cloud. It offers on-demand resource provisioning for scalable simulation, self-organization of adaptive simulation environment, and enhanced security by isolation between federation executions. Besides, it support web-based user interface to support easy access and simulation management.
Keywords: Cloud Computing; PaaS; Distributed-parallel simulation; HLA; RTI
Fast Shear Skew Warp Volume Rendering Using GPGPU for Cloud 3D Visualization BIBAKFull-Text 335-339
  Chang-Woo Cho; Ki-Hyun Kim; Ki-Young Choi; Chang-Sung Jeong
We present a method for fast volume rendering using GPGPU for Cloud 3D Visualization. Each of the threads is processed in parallel on the GPU. Our algorithm is to use the computational power of graphics processors to speed up the rendering process of the Shear-Warp algorithm. Our GPU-based method provides real-time frame rates and outperforms the GPU-based implementation. Our experimental results show that our algorithm is much faster than the CPU-based Shear-Warp volume visualization in terms of rendering speed and image quality. It can be launched on GPU computing clusters provided by cloud infrastructures such as Amazon EC2.
Keywords: Shear warp volume rendering; Cloud 3D Visualization; Parallel Processing
A Vision-Based Robust Hovering Control System for UAV BIBAKFull-Text 341-345
  Tyan Vladimir; Dongwoon Jeon; Doo-Hyun Kim
This paper introduces an algorithm for real-time line detection and tracking utilizing the Graphic Processing Units (GPUs) for UAV's vision-based hovering control system. We concentrate that there are many of lines where UAV can fly, and extract meaningful line to grasp of vehicle's attitude. We implemented image processing techniques on GPUs for real-time performance because detection and tracking of lines need huge computational resources. Experiments show affordable frame throughput that our approach is feasible in real flight.
Keywords: CUDA; Line Detection; Line Tracking; Hough Transform; Hovering System
Finding Relationships between Human Affects and Colors Using SVD and pLSA BIBAKFull-Text 347-351
  Umid Akhmedjanov; Eunjeong Ko; Yunhee Shin; Eun Yi Kim
In this paper, a new method is presented to automatically find relationships between human affects and colors. For this, the probabilistic latent semantic model analysis (pLSA) and singular value decomposition (SVD) is applied. The proposed method is composed of three modules: feature extraction, feature transform and pLSA training. We first segment the image using mean-shift clustering, then extract color compositions by analyzing the colors from one region and its adjacent regions. Next, for the occurrence matrix, the SVD and pLSA are used. Using SVD, the occurrence matrix is decomposed into rank and null space matrix, where the null space is discarded and only the space corresponding to the singular values is used for further processing. For the reconstructed matrix, the pLSA is applied to obtain the correlation between affective classes and color compositions. To assess the effectiveness of the proposed system, it was applied to index the images using human affects. Then the results showed the effectiveness of the proposed method.
Keywords: Affective Mapping; Probabilistic Affective Model; Singular Value Decomposition; Probabilistic latent semantic analysis
Home Appliance Control and Monitoring System Model Based on Cloud Computing Technology BIBAKFull-Text 353-357
  Yun Cui; Myoungjin Kim; Seung-woo Kum; Jong-jin Jung; Tae-Beom Lim; Hanku Lee; Okkyung Choi
With the development of intelligent home appliance technology, real-time home appliance status information is now generated in large quantities. New technology is necessary in order to process the large amount of status information that is generated every day. An innovative technology that has recently been used to process large amounts of data is cloud computing. Therefore, in this paper, we propose a system model to control and monitor home appliances using home network and cloud computing technologies in a smart home environment. UPnP technology is used to extract status information from home appliances. Cloud computing technology analyzes and processes the information and also provides virtualization services to users. In the proposed method, the gateway collects and stores home appliance information using home network technologies and sends the information to the cloud server for storage and management.
Keywords: cloud computing technology; UPnP; virtualization services; smart home
Load Distribution Method for Ensuring QoS of Social Media Streaming Services in Cloud Environment BIBAKFull-Text 359-363
  Seung Ho Han; Myoungjin Kim; Yun Cui; SeungHyun Seo; Yi Gu; Hanku Lee
As various types of smart devices have recently appeared, SNS (Social Networking Services) have been expanded. Thus, the demand for social media streaming is on the rise. In the previous study, a media conversion system for ensuring QoS (Quality of service) of media streaming was presented. The presented system implemented a distributed streaming environment with multiple servers in order to perform reliable streaming of converted media. The method of distributing streaming job is crucial in implementing a distributed environment. Thus, the presented system established distributed streaming servers that applied RR (Round Robin) and LC (Least Connection) algorithms. However, since systems that applied RR and LC do not consider CPU utilization rate and network transmission traffic, they have limitations on reducing the burdens of servers. This study will present a SRC (Streaming Resource-based Connection) scheduling algorithm for ensuring QoS in the distributed streaming environment. The focus of this SRC algorithm considering CPU utilization rate and transmission traffic of servers is resolving the limitations of existing algorithms. As a performance evaluation, utilization rate of different systems that each applied SRC, RR and LC will be compared.
Keywords: cloud computing; Social Media Streaming; Load Distribution; QoS
A Robust Cloud-Based Service Architecture for Multimedia Streaming Using Hadoop BIBAKFull-Text 365-370
  Myoungjin Kim; Seung Ho Han; Jong-jin Jung; Hanku Lee; Okkyung Choi
Delivering scalable rich multimedia applications and services on the Internet requires sophisticated technologies for transcoding, distributing, and streaming content. Although cloud computing provides an infrastructure for such technologies, the specific challenges of task management, load balancing, and fault tolerance remain. To address these issues, we propose a cloud-based distributed multimedia streaming service, or CloudDMSS. The system is designed to run on all major cloud computing services, and is highly adapted to the structure and policies of Hadoop, which give it additional capabilities for transcoding, task distribution, load balancing, content replication and distribution.
Keywords: Streaming Service; Mobile Media Service; Cloud Computing; Media Transcoding
Video Image Based Hyper Live Spatial Data Construction BIBAKFull-Text 371-376
  Yongwon Cho; Muwook Pyeon; Daesung Kim; Sujung Moon; Illwoong Jang
Recently, Spatial information technology closer to reality three-dimensional space, a variety of information services and Web-based content, to provide information services through the space. CCTV video and camera video based multi-dimensional image data in order to build real-time spatial information-based and user-built CCTV video was up-loaded to the online video you should use them. In order to use the video upload large amounts of processing is required, spatial information can be presented as an alternative to building, large amount of data as a way to effectively use big data and cloud computing.
Keywords: Spatial Data; Hyper Live Map; service; Video Image; CCTV
A Peer-to-Peer Based Job Distribution Model Using Dynamic Network Structure Transformation BIBAKFull-Text 377-383
  Seungha Lee; Yangwoo Kim; Woongsup Kim
Typically, many systems in organizations suffer from limited computer resources, while there are a huge number of under-utilized computers available which are able to contribute to continue reliable services when a system faces operational overloads. This paper proposes a Peer-to-Peer (P2P) based distributed job distribution model for job allocation and aggregation using Hub Peers. To this end, we first provide a peer-to-peer job distribution model using periodically collected peer information, and then proposed a peer structure transformation algorithm that composes P2P network topology dynamically for using under-utilized computing resources efficiently. Finally, we prove the benefits of our approach by comparing our proposed approach to other works.
Keywords: Peer-to-Peer system; distributed processing; job management; load sharing; tree transformation
Distributed 2D Contents Stylization for Low-End Devices BIBAKFull-Text 385-389
  Mingyu Lim; Yunjin Lee
As a variety of computing devices have been developed and the Internet helps them to provide various content services in ubiquitous computing environments, users want higher quality of such services. Existing approaches focus on 3D content rendering by a remote server in order to solve the limitation of low-end devices. In this paper, we propose a distributed rendering mechanism for 2D content using multiple servers. Since large 2D image stylization also requires high computation overhead to render an image, a low-end client partition it into several image pieces. Each piece is sent to a different server, which then performs rendering. A client merges the rendered pieces to one output image again. The proposed method enables large images to be rendered by collaboration of multiple servers with reasonable processing and communication cost.
Keywords: Content stylization; Distributed rendering; Multiple servers; Lowend clients
Authority Delegation for Safe Social Media Services in Mobile NFC Environment BIBAKFull-Text 391-395
  Jinsung Choi; Okkyung Choi; Yun Cui; Myoungjin Kim; Hanku Lee; Kangseok Kim; Hongjin Yeh
With the rapid development of NFC (Near Field Communications) technologies, NFC-enabled mobile devices are replacing the existing RFID such as mobile payment service, access control system of door locks, and ticketing service. Your service access authority through authenticating the mobile device can be delegated to any person temporarily. But when a person wants to share one's authority to others, it would be considered prevention for abuse of authority. For example, when parents give the payment right to their children, it can be used indiscriminately. And it can be abused when the authority is transferred to a third party without checking authentication code. So transferred tickets or copied access authority of door locks can occur. In this paper, for safe authority delegation, we will contain the authorized user's identity and check authorization in mobile device whether it contains suitable delegation information or not.
Keywords: Near Field Communication; Authority; Delegation; Smartphone; User Identification

Mobile Computing

Introspection-Based Periodicity Awareness Model for Intermittently Connected Mobile Networks BIBAKFull-Text 397-403
  Okan Turkes; Hans Scholten; Paul Havinga
Recently, context awareness in Intermittently Connected Mobile Networks (ICMNs) has gained popularity in order to discover social similarities among mobile entities. Nevertheless, most of the contextual methods depend on network knowledge obtained with unrealistic scenarios. Mobile entities should have a self-knowledge determination in order to estimate their activity routines in a group of communities. This paper presents a periodicity awareness model which relies on introspective spatiotemporal observations. In this model, hourly, daily, and weekly locations of mobile entities are being tracked to predict future trajectories and periodicities within a targeted time period. Realistic simulations are utilized to analyze the predictions in weekly observation sets. The results show that a reasonable accuracy with an increasing level of determination can be obtained which does not require global network knowledge. In this regard, the presented model can give insights for any type of ICMN objectives.
Keywords: Intermittently-connected mobile networks; social networks; context-awareness; periodicity awareness model; spatiotemporal correlations
Collaborative Recommendation of Mobile Apps: A Swarm Intelligence Method BIBAKFull-Text 405-412
  Xiao Xia; Xiaodong Wang; Xingming Zhou; Tao Zhu
The explosive growth of mobile apps has given rise to the challenge of finding out interesting apps for users. Recommender systems are employed to meet this challenge. However, as the lack of user and app data, the development of recommender systems for mobile apps is still at a slow pace. Therefore, we propose a system-level collaboration approach to facilitate the development of new systems by making a better use of the data from existing systems. To this end, we model the recommendation generation as an optimization problem and propose a new set-based particle swarm optimization method to solve it. We further develop three systems to evaluate our approach and algorithm. Evaluations based on real data have verified their performances on both the effectiveness and the efficiency.
Keywords: Mobile app; recommender system; collaboration; set based PSO
Enhanced Implementation of Max* Operator for Turbo Decoding BIBAKFull-Text 413-419
  Dongpei Liu; Hengzhu Liu; Li Zhou
Max* operator is the kernel operation in MAP decoding. An intuitive approximation to the correction term of max* operator is presented. The binary-tree based architecture for multi-variable max* calculation is also suggested. The proposed max* operator provides a good trade off between hardware overhead and logic delay, and can be easily realized in parallel. Simulations on (37,21) turbo code demonstrate that the BER performance of proposed scheme is almost near the optimal Log-MAP algorithm and significantly superior to the Max-Log-MAP algorithm. The proposed enhanced implementation of max* operator has potential applications in turbo decoder.
Keywords: Max* operator; Correction function; Turbo decoding; Log-MAP algorithm; Max-Log-MAP algorithm

Ubiquitous Computing

A Context Description Language for Medical Information Systems BIBAKFull-Text 421-432
  Kurt Englmeier; John Atkinson; Josiane Mothe; Fionn Murtagh; Javier Pereira
Contextualized delivery of information is one of the many strengths of ubiquitous computing. It makes information actionable and helps us to better understand our situations. In the realm of healthcare, contextual information provides a terse but precise picture of the patient's health situation. The patient context can have many facets, ranging from nutrition context over health heritage context to the context of symptoms, just to name a few. Setting up the correct health condition context of a patient favors better and faster recognition of the patient's actual health situation.
   Context-awareness in medical monitoring mainly concentrates on gathering numerical facts depicting special aspects of a person's health condition. In this paper we want to broaden the focus on the textual dimension in context development, by considering semantic annotation in designing context-awareness. We describe an approach for a context description language (CDL) that supports the uniform presentation of textual facts in medical reports and automatic reasoning on these facts. Term clusters in medical reports represent in a unique way symptoms and findings that set up the health context reflected in this particular report. These clusters manifest potential health condition contexts where a patient can be viewed in. A reasoning engine operates on these context presentations and selects those that match best the patient's health situation. Locating the right context supports the physician in faster getting a first picture of the probable health situation of a new patient to be examined. We present experiments with a CDL applied on reports related to respiratory problems.
Keywords: Context-awareness; context design and development; semantic annotation; domain-specific language; information mining; natural language interaction; medical reports
Eccentricity-Based Data Gathering and Diameter-Based Data Forwarding in 3D Wireless Sensor Networks BIBAKFull-Text 433-439
  A. S. M. Sanwar Hosen; Gi-hwan Cho
This paper proposes an efficient method of constructing three dimensional (3D) wireless sensor networks (WSNs) with aiming to minimization of the overall routing cost. It tries to divide the network into subspaces, and elects a routing centroid node in the eccentricity region from any node in a subspace in terms of minimizing communication cost of that space. The node in an eccentricity region is naturally close to the distance of the radius of that subspace. As a result, the centroid node can forward the gathered data to the node on the diameter. The minimization of the path cost in data gathering and forwarding towards the sink is an efficient approach to design a cost effective 3D WSNs.
Keywords: Wireless Sensor Network; 3D space; Routing Cost; Eccentricity; Diameter; Radius
Weighted Mining Frequent Itemsets Using FP-Tree Based on RFM for Personalized u-Commerce Recommendation System BIBAKFull-Text 441-450
  Young Sung Cho; Song Chul Moon
This paper proposes a new weighted mining frequent itemsets using FP-tree based on RFM for personalized u-commerce recommendation system under ubiquitous computing. Existing recommendation system using association rules still does not only reflect exact attributes of item but also has the problem, such as delay of processing speed from a cause of frequent scanning a large data, scalability and accuracy. In this paper, to solve these problems, it is necessary for us to make RFM (Recency, Frequency, Monetary) score of item and to extract the most frequently purchased data from the whole data in order to improve the accuracy of recommendation, to consider frequently changing the weighted patterns by emphasizing the important items with high purchasability according to the threshold for creative the weighted mining frequent itemsets using FP-tree without occurrence of candidate set. To verify improved performance, we make experiments with dataset collected in a cosmetic internet shopping mall.
Keywords: Association Rules; RFM; Weighted Mining Frequent Itemsets using FP-tree
The System of Stress Estimation for the Exposed Gas Pipeline Using the Wireless Tilt Sensor BIBAKFull-Text 451-456
  Jeong Seok Oh; Hyo Jung Bng; Si-Hyung Lim
Gas pipelines are exposed to the danger especially in bridges, roads and subway construction areas. It can cause leakage accidents from the stress and vibration changes and it can threaten human's life. To avoid that, the gas pipelines should be monitored continuously. The system of stress estimation using MEMS (Micro electro mechanical system_ wireless tilt sensor has been developed and has been evaluated by a lob test bench.
Keywords: MEMS; gas pipeline; tilt sensor; stress; wireless communication
The Architecture Design of Semantic Based Open USN Service Platform Model BIBAKFull-Text 457-462
  Hyungkyu Lee; Namje Park; Hyo-Chan Bang
Open USN service is USN service to enable enhanced and flexible service interoperability and provisioning based on the use of standards interfaces. In this paper is to define an open USN service framework, and provide reference architecture of open USN service framework. The use of standard interfaces of open USN service framework will ensure USN service reusability, portability across several USN services, as well as accessibility and interoperability by USN application providers and/or developers. This paper will contribute to the development and activation of new variety USN service by deploying sensor node constructed on variety field or sensor network to share and utilize at different service field.
Keywords: Open USN Service; Semantic; Platform Model; USN; WSN
Use-Cases and Service Modeling Analysis of Open Ubiquitous Sensor Network Platform in Semantic Environment BIBAKFull-Text 463-468
  Taegyeong Kang; Namje Park; Hyungkyu Lee; Hyo-Chan Bang
The ubiquitous sensor network is a well-known keyword in information and communication technology area and many standards development organizations are developing standards for USN and other similar technologies. However, ubiquitous sensor network services are not widely spread yet because current ubiquitous sensor network services require user or application developer to have knowledge of sensors and sensor networks for using ubiquitous sensor network services or developing Ubiquitous Sensor Network applications. In this paper describes Use-cases and Service Modeling Analysis specific to the support of open ubiquitous sensor network service framework.
Keywords: Open USN Service; Semantic; Platform Model; USN; WSN
A Neural Network Based Simple Weak Learner for Improving Generalization Ability for AdaBoost BIBAKFull-Text 469-474
  Jongjin Won; Moonhyun Kim
The performance of ensemble, including AdaBoost, is determined by accuracy and generalization ability. However, the currently available AdaBoost's weak learners mostly show high accuracy but rather low generalization ability. In this paper, we introduce three requirements that weak learners must satisfy in order to improve generalization ability of AdaBoost. Then, we propose w-delta learning rule based neural network (NN) as a weak learner that satisfies those requirements. Through experiments, we show that the proposed method improves generalization ability while maintaining the high accuracy.
Keywords: AdaBoost; Generalization ability; W-delta learning rule

Intelligent Computing

Serial Dictatorial Rule-Based Games for Camera Selection BIBAKFull-Text 475-481
  Gowun Jeong; Yong-Ho Seo; Sang-Soo Yeo; Hyun S. Yang
A wireless, battery-powered, stationary camera sensor network optimizes trade-off between extending its lifetime and enhancing its sensing accuracy by activating only a desirable camera subset for given targets in a timely fashion. This paper models this selection problem in a cooperative bargaining game based on the serial dictatorial rule, where cooperative sensors sequentially decide their mode between "sleep" and "active" in descending order of their bargaining power. Simulated resource overheads and the concerned performances, network lifetime and sensing accuracy are given as well.
Keywords: sensor scheduling; coverage; cooperative bargaining game; serial dictatorial rule
Security Analysis on a Group Key Transfer Protocol Based on Secret Sharing BIBAKFull-Text 483-488
  Mijin Kim; Namje Park; Dongho Won
Group key exchange protocols are cryptographic algorithms that describe how a group of parties can communicate with their common secret key over insecure public networks. In 2013, Olimid proposed an improved group key transfer protocol based on secret sharing, and claimed that he eliminated the flaws in Sun et al.'s group key transfer protocol. However, our analysis shows that the protocol is still vulnerable to outsider and insider attacks and does not provide known key security. In this paper, we show a detailed analysis of flaws in the protocol.
Keywords: key exchange protocol; group key transfer; secret sharing; attack; confidentiality
Analysis of Cyber Attacks and Security Intelligence BIBAKFull-Text 489-494
  Youngsoo Kim; Ikkyun Kim; Namje Park
A cyber attack is deliberate exploitation of computer systems, technology-dependent enterprises and networks. Cyber attacks use malicious code to alter computer code, logic or data, resulting in disruptive consequences that can compromise data and lead to cybercrimes, such as information and identity theft. Cyber attack is also known as a computer network attack (CNA). Cyber attacks occurred targeting banks and broadcasting companies in South Korea on March 20. The malware involved in these attacks brought down multiple websites and interrupted bank transactions by overwriting the Master Boot Record (MBR) and all the logical drives on the infected servers rendering them unusable. It was reported that 32,000 computers had been damaged and the exact amount of the financial damage has not yet been calculated. More serious is that we are likely to have greater damages in case of occurring additional attacks, since exact analysis of cause is not done yet. APT(Advanced Persistent Threat), which is becoming a big issue due to this attack, is not a brand new way of attacking, but a kind of keyword standing for a trend of recent cyber attacks. In this paper, we show some examples and features of recent cyber attacks and describe phases of them. Finally, we conclude that only the concept of security intelligence can defend these cyber threats.
Keywords: Cyber Attacks; Security Intelligence; MBR; APT; Threat
Protection Profile for PoS (Point of Sale) System BIBAKFull-Text 495-500
  Hyun-Jung Lee; Youngsook Lee; Dongho Won
A PoS system immediately obtains the data related to the sale at the time and place of purchase. It provides an initial interface for the credit card transaction to happen. Due to its dealing with sensitive data such as credit card information, many relevant organizations have been trying to suggest security standards. However, there still is no PoS system that guarantees security, which results in a lot of hacked PoS systems in different countries. This paper intends to draw out security functional requirements for a PoS system based on the CC, which can be used as a reference for its security evaluation.
Keywords: Protection Profile; CC (Common Criteria); PoS(Point of Sale); Security Requirement; Vulnerability

Intelligent and Mobile Services

A Probabilistic Timing Constraint Modeling and Functional Validation Approach to Dynamic Service Composition for LBS BIBAKFull-Text 501-508
  Weimin Li; Xiaohua Zhao; Jiulei Jiang; Xiaokang Zhou; Qun Jin
Location Based Services (LBS) is a kind of real-time service with uncertain factors, and its modeling and validation is essential. In this paper, with the introduction of probability, we propose a Color Probability-TCPN (CP-TCPN) by using the tokens with specific colors as the research objects and redefining several relative parameters. We use CP-TCPN to realize modeling and functional verification of the dynamic services composition for LBS. Simulation result is presented to illustrate the application of CP-TCPN in the modeling and analyzing of the real-time system with uncertain factors.
Keywords: CP-TCPN; Probability; Colors; Service composition; Modeling; Functional validation
An Implementation of Augmented Reality and Location Awareness Services in Mobile Devices BIBAFull-Text 509-514
  Pei-Jung Lin; Sheng-Chang Chen; Yi-Hsung Li; Meng-Syue Wu; Shih-Yue Chen
The popularization of smartphones and advances in location-based technology have led to the creation of many applications and diverse mobile cloud technologies. Augmented reality (AR), which integrates virtual reality with the real world, is one of the mobile service technologies that have been receiving considerable attention in recent years. This study focuses on AR technology, which in conjunction with point of interest (POI) information
   established in a cloud database, enables users to instantly obtain services with the camera lenses of their mobile devices. The developed system allows users to quickly share AR images and information with others in their social
   networks from their current locations. This work includes social communities, photographing, radar detection, and GPS positioning that facilitate various human-machine interactions and information searches.
Development of STEAM Education Program Centering on Non-traditional Energy BIBAKFull-Text 515-520
  Yilip Kim; Jeongyeun Kim; Namje Park; Hyungkyu Lee
The purpose of this paper is as follows. First, it is to develop a STEAM education program centering on non-traditional energy (gas hydrate and shale gas) targeting high school students. Second, it is to develop teaching-learning materials that can be used in the education program. To achieve the purpose of this paper, we investigated the contents of the subject using newspaper articles, academic books, and academic journals and also analyzed the curriculum and textbooks of related subjects revised in Korea's 2009 to determine the learning capability of the students by age.
Keywords: STEAM; Non-traditional Energy; Elementary School; Teaching Method
Scalable Key Management for Dynamic Group in Multi-cast Communication BIBAKFull-Text 521-527
  Fikadu B. Degefa; Dongho Won
To have secure multicast group communication, group key management plays an essential role to guarantee data security. Because communication bandwidth, storage memory, and computational power are limited resources, most group key management schemes for scalable secure multicast communications have focused on reducing the number of update messages, number of stored keys, and computational load. Here, also we propose efficient scheme in such a way that solves these problems.
Keywords: Key management; Dynamic group; Multi-cast
Result of Implementing STEAM Program and Analysis of Effectiveness for Smart Grid's Education BIBAKFull-Text 529-534
  Jeongyeun Kim; Yilip Kim; Namje Park
STEAM education allows students to improve themselves in cognitive and affective domains in math.science. Visualizing and storytelling science as described above is a good approach towards learning science in an easy and exciting way. Aware of such, our paper plans to advocate convergence curriculum to nurture students' creativity, problem-solving skills and ultimately support them to become a creative talent built on convergence. This paper intends to investigate the effect of STEAM education program using smart grid on elementary school students. For this, this paper implemented STEAM education program using smart grid for ten students in the 4th grade in Jeju Special Self-Governing Province.
Keywords: STEAM; Smart Grid; effectiveness; Science; Technology
Security Enhanced Unlinkable Authentication Scheme with Anonymity for Global Mobility Networks BIBAKFull-Text 535-540
  Youngseok Chung; Seokjin Choi; Youngsook Lee; Dongho Won
Recently, Chung, Lee, and Won [1] proposed an improved authentication scheme with anonymity which remedies security faults showed by Youn, Park, and Lim [2]. Their improved scheme guarantees anonymity, but does not provide unlinkability. In their scheme, it is possible for attackers to know particular sessions, that have already been occurred several times, are originated by one same user. In this paper, we propose an unlinkable authentication scheme with anonymity by modifying Chung et al.'s scheme. Our scheme provides not only anonymity and security as the previous scheme does, but also unlikability against malicious mobile users. Since proposed scheme still uses only low-cost functions, it is suitable for mobility networks.
Keywords: anonymity; linkability; authentication; mobility network

3D Converged IT and Optical Communications

A Feature-Based Small Target Detection System BIBAKFull-Text 541-548
  Jong-Ho Kim; Young-Su Park; Sang-Ho Ahn; Sang-Kyoon Kim
Existing small target detection systems generally use the difference image between a predicted background image and an original image. This method has two disadvantages. First, to predict the background image, the size of the structural element has to be carefully selected considering the size of small targets. Second, because of blurring, clutter such as clouds can occur around the edge of the background. To deal with these problems we propose a new feature-based detection system. The proposed method selects candidate pixels with Harris corner detector and then, again selects pixels that have a higher intensity than a threshold among the candidates. After labeling the selected candidates in order to obtain the number of pixels they have, the system decides which is a small target. In an experiment, our proposed method gave better results than the existing methods.
Keywords: Harris corner detector; New White Top-Hat; Labeling; Histogram

Frontier Computing -- Theory, Technologies and Applications

A Small Target Detection System Based on Morphology and Modified Gaussian Distance Function BIBAKFull-Text 549-556
  Jong-Ho Kim; Jun-Jae Park; Sang-Ho Ahn; Sang-Kyoon Kim
We propose a new small target detection system that detects small target candidates based on morphology operations and detects actual targets using a modified Gaussian distance function. To reduce clutter on the edges of clouds, a median filter is applied as preprocessing. Two kinds of images are calculated with closing and opening morphological operators, respectively. In the morphology operations, various sizes of structure elements are used to consider the sizes of targets and candidate targets are extracted from difference images between the two images in the closing and opening operations. With a modified Gaussian distance function, small targets are detected from the candidate targets. The proposed method is less sensitive to clutters than existing methods, and has a detection rate of 98%.
Keywords: IR Image; Small Target; Gaussian Distance Function; Top-Hat; NWTH
Using Hardware Acceleration to Improve the Security of Wi-Fi Client Devices BIBAKFull-Text 557-562
  Jed Kao-Tung Chang; Chen Liu
As mobile devices prevail, communications security has become a critical and popular topic in recent years. For example, when a mobile device accesses Wi-Fi, the data communicated with the Wi-Fi access point may be encrypted to provide extra security. However, on a mobile device with limited energy budget, data encryption/ decryption often imposes high pressure on the battery life. In this paper, we review how a recently introduced Extremely Heterogeneous Architecture (EHA) can potentially be used to improve performance and energy efficiency of data encryption/decryption on mobile devices.
Keywords: Wi-Fi; Extremely Heterogeneous Architecture (EHA); cryptography; hardware acceleration; performance analysis; hotspot function
An Anonymous Communication Scheme with Non-reputation for Vehicular Ad Hoc Networks BIBAKFull-Text 563-568
  Ching-Hung Yeh; Meng-Yen Hsieh; Kuan-Ching Li
Vehicular Ad hoc Network (VANET) is a kind of open wireless communication network and uses 802.11p protocol to interconnect vehicles and provides numerous services. Although it brings many of convenient applications but unlike traditionally wired networks are protected by several defenses such as firewalls and gateways, it may face a variety of security challenges such as security and privacy. The users need anonymous mechanisms to enable unlink ability but impartial third party requests non-reputation for accidents or certain events. In brief, the privacy preserving and non-reputation are contradictory. To overcome these flaws, an anonymous communication scheme with non-reputation for vehicular ad hoc networks is proposed. Our proposed scheme not only accomplishes anonymously communication between vehicle-to-vehicle and vehicle-to-roadside infrastructure for protecting privacy, but also achieves non-reputation function for identifying vehicle.
Keywords: vehicular ad hoc network; security; privacy; non-reputation
A Mobility Management Scheme for Internet of Things BIBAKFull-Text 569-575
  Yuan-Kai Hsiao; Yen-Wen Lin
In this paper, a new mobility management scheme is proposed for IoT (Internet of Things) environment. Usually, the mobile nodes are resource-limited. Specially, the mobile nodes may move together with human or vehicles in IoT. To provide ubiquitous IoT services, a network-based mobility management scheme supporting global mobility and group mobility is proposed in this paper. The performance analysis shows that the proposed scheme improves handover delay and control overhead in the context of IoT.
Keywords: IoT; Network Based; Global Mobility; Group Mobility
An Overlay Network Based on Arrangement Graph with Fault Tolerance BIBAKFull-Text 577-583
  Ssu-Hsuan Lu; Kuan-Ching Li; Kuan-Chou Lai; Yeh-Ching Chung
As people change the habit of using the Internet, network technology has become matured. Unlike client-server, peer-to-peer (P2P) technology increases the convenience of people's daily life. The routing efficiency of P2P system without centralized server always is an important issue. This paper proposes a virtual peer mechanism of P2P overlay network based on the arrangement graph to make exiting physical peers be agent peers for vacant peers. Each vacant peer is managed by a physical peer who often is its neighbor, and the vacant peer is called virtual peer. Physical peers and virtual peers make the arrangement graph full, and make the number of routing hops can be limited within the diameter of the arrangement graph. From experimental results, this system can keep routing efficiency no matter the number of peers and do not increase system overhead.
Keywords: Peer-to-Peer; overlay network; arrangement graph; fault tolerance
Event Detection in Wireless Sensor Networks: Survey and Challenges BIBAKFull-Text 585-590
  Aziz Nasridinov; Sun-Young Ihm; Young-Sik Jeong; Young-Ho Park
In typical wireless sensor networks (WSNs), sensor nodes have limited resources such as battery power, computing capability and memory. Creating an event detection method comprising with those resource limitations is not an easy task and this sets several challenges. In this paper, we first describe challenges in event detection in WSNs. Then, we investigate the previous studies that have been done for solving those challenges.
Keywords: Event detection; wireless sensor networks; survey
Accelerating Adaptive Forward Error Correction Using Graphics Processing Units BIBAKFull-Text 591-597
  Md Shohidul Islam; Jong-Myon Kim
The demand of error free high-speed, real-time wireless communication is mounting day by day. Adaptive forward error correction (AFEC) is one of the error control mechanisms in which corrupted packets are automatically corrected at the destination end. Graphics processing units (GPUs) offer highly parallel computing platform, and we propose a GPU based AFEC approach for fast error recovery in this paper. We develop a massively parallel AFEC algorithm using the GPU and accomplish performance comparison with an equivalent serial algorithm that runs on the traditional CPU. Experimental results demonstrate that the proposed GPU based AFEC approach enormously outperforms the sequential approach yielding significant reduction in execution time while improving buffer utilization. In addition, the proposed GPU based approach achieves the average speedup of 74X over the sequential algorithm using the CPU while reducing the computational complexity from O(n³) of the sequential algorithm to O(n) by using the single instruction multiple data (SIMD) based GPU.
Keywords: High-speed real-time wireless communication; packet corruption; AFEC; Hamming code; GPU
High-Performance Sound Engine of Guitar on Optimal Many-Core Processors BIBAKFull-Text 599-607
  Myeongsu Kang; Cheol-Hong Kim; Jong-Myon Kim
This paper presents design space exploration of optimal many-core processors for physics-based sound synthesis of an acoustic guitar by quantitatively evaluating the impact of the sample-per-processing element (SPE) ratio, which is the amount of sample data directly mapped to a processing element (PE). This paper evaluates system performance in terms of execution time, area and energy efficiencies for high-performance sound engine of the guitar as the SPE ratio is varied. Experimental results indicate that the SPE ratio in the range of 2,756 (or PEs=24) to 11,025 (or PEs=96) provides the most efficient operation for synthesizing guitar sounds with 6-note polyphony sampled at 44.1 kHz.
Keywords: Area efficiency; design space exploration; energy efficiency; many core processors; physics-based sound synthesis; sample-per-processing element
Community Identification in Multiple Relationship Social Networks BIBAKFull-Text 609-614
  Ting-An Hsieh; Kuan-Ching Li; Kuo-Chan Huang; Kuo-Hsun Hsu; Ching-Hsien Hsu; Kuan-Chou Lai
In a social network, individuals often simultaneously belong to multiple social communities; therefore, the detection of relationships among individuals is very important. However, most of community detection methods only apply a single relationship in dynamic social networks with multi-relationships among individuals. Therefore, this study proposes a CNET Hierarchical Division Algorithm (CHDA) to detect communities efficiently. Experimental results show that the proposed CHDA could detect communities with more precise recognition, regarding their characterization.
Keywords: community detection; dynamic social network; hierarchical division; recognition
An Improved ACO by Neighborhood Strategy for Color Image Segmentation BIBAKFull-Text 615-620
  Shih-Pang Tseng; Ming-Chao Chiang; Chu-Sing Yang
This paper presents an efficient method for speeding up ant colony optimization (ACO) in solving the color image segmentation problem. The proposed method is inspired by the heuristics of image segmentation to reduce the computation time. To evaluate the performance of the proposed method, we applied the method on well-known test images. Our experimental results shows that the proposed method can significantly reduce the computation time about 19% to 45%.
Keywords: Color image segmentation; clustering; ant colony optimization
A Novel Spiral Optimization for Clustering BIBAKFull-Text 621-628
  Chun-Wei Tsai; Bo-Chi Huang; Ming-Chao Chiang
Because most traditional search methods are unable to satisfy the current needs of data mining, finding a high performance search method for data mining has gradually become a critical issue. The spiral optimization (SO) is a promising search algorithm designed to emulate the natural phenomena, such as swirl and low pressure, to find the solutions of optimization problems within an acceptable computation time. In this paper, a novel SO is presented to solve the clustering problem. Unlike the original SO, which rotates the points around the elitist center iteratively, the proposed algorithm, called distributed spiral optimization (dSO), splits the population into several subpopulations so as to increase the diversity of search to further improve the clustering result. The k-means and oscillation methods are also used to enhance the efficacy of dSO. To evaluate the performance of the proposed algorithm, we apply it to the clustering problem and compare the results it found with those of the spiral optimization and genetic k-means algorithm. The results show that the proposed algorithm is quite promising.
Keywords: Metaheuristic; spiral optimization; clustering
Recent Development of Metaheuristics for Clustering BIBAKFull-Text 629-636
  Chun-Wei Tsai; Wei-Cheng Huang; Ming-Chao Chiang
Metaheuristics have been successfully applied to quite a lot of services, systems, and products frequently found in our daily life. Until now, none of the metaheuristics ever proposed are perfect for all the optimization problems; rather, each algorithm has its pros and cons. Although several high-performance metaheuristics exist, there is still plenty of room to improve the final result they produce and the computation time they take. Since 2001, quite a few number of novel metaheuristics have been developed to provide a better way for solving the optimization problems. A brief review for eight of these novel metaheuristics is given in this paper. To evaluate the performance of these algorithms, we apply them to a well-known combinatorial optimization problem, data clustering, and the results are analyzed and discussed.
Keywords: Metaheuristics; clustering; combinatorial optimization problem
The Originality of a Leader for Cooperative Learning BIBAKFull-Text 637-642
  Po-Jen Chuang; Chu-Sing Yang
A team work begins with cooperative learning, the responsibility of associators, abilities of individuals, and team progress are rewarded for helping and sharing to each other. The leader leads the group comprehending tasks, directing discussion, and progressing in studying. For better cooperation, we explode the previous research of the grouping strategy -- pairing strategy [1] which enhances the learning and testing results of students based on the relationship of social network. Basing on the society relationships, this method provides members of the groups to learn from or mimic learners who have good relationships with them. This paper discusses about supports, including parents, teachers, and members of the group, for a leader leading the group to a higher achievement. The specific factors are also discussed, that the parents' support is the biggest among the others.
Keywords: Social Learning; Cooperative Learning; Grouping Strategy; Learning Strategy; Learning Achievement; Regression Line
A Hybrid Ant-Bee Colony Optimization for Solving Traveling Salesman Problem with Competitive Agents BIBAKFull-Text 643-648
  Abba Suganda Girsang; Chun-Wei Tsai; Chu-Sing Yang
This paper presents a new method called hybrid ant bee colony optimization (HABCO) for solving traveling salesman problem which combines ant colony system (ACS), bee colony optimization (BCO) and ELU-Ants. The agents, called ant-bees, are grouped into three types, scout, follower, recruiter at each stages as BCO algorithm. However, constructing tours such as choosing nodes, and updating pheromone are built by ACS method. To evaluate the performance of the proposed algorithm, HABCO is performed on several benchmark datasets and compared to ACS and BCO. The experimental results show that HABCO achieves the better solution, either with or without 2opt.
Keywords: Hybrid; Ant Colony System; Bee Colony System; Traveling Salesman Problem