September 2013, No. 5
Print E-mail

Efficient High Dimension Data Clustering using Constraint-Partitioning K-Means Algorithm

Aloysius George
Grian Technologies Private Limited, Research and Development Company, India


With the ever-increasing size of data, clustering of large dimensional databases poses a demanding task that should satisfy both the requirements of the computation efficiency and result quality. In order to achieve both tasks, clustering of feature space rather than the original data space has received importance among the data mining researchers. Accordingly, we performed data clustering of high dimension dataset using Constraint-Partitioning K-Means (COP-KMEANS) clustering algorithm which did not fit properly to cluster high dimensional data sets in terms of effectiveness and efficiency, because of the intrinsic sparse of high dimensional data and resulted in producing indefinite and inaccurate clusters. Hence, we carry out two steps for clustering high dimension dataset. Initially, we perform dimensionality reduction on the high dimension dataset using Principal Component Analysis (PCA) as a preprocessing step to data clustering. Later, we integrate the COP-KMEANS clustering algorithm to the dimension reduced dataset to produce good and accurate clusters. The performance of the approach is evaluated with high dimensional datasets such as Parkinson’s dataset and Ionosphere dataset. The experimental results showed that the proposed approach is very effective in producing accurate and precise clusters.

Keywords: Clustering, dimensionality reduction, PCA, COP-KMEANS algorithm, clustering accuracy, parkinson's dataset, ionosphere dataset.
  Received July 14, 2011; accepted December 30, 2011; published online August 5, 2012
Print E-mail

3D Model Retrieval Based on 3D Fractional Fourier Transform

Liu YuJie1, Bao Feng1, Li ZongMin1, and Li Hua2
1School of Computer Science Communication Engineering, China University of Petroleum, China
2National Research Center for Intelligent Computing Systems, Institute of Computing Technology, Chinese Academy of Sciences, China


In this paper, a new tool that is fractional Fourier Transform is introduced to 3D model retrieval. And we propose a 3D model descriptor based on 3D factional Fourier transform. Fractional Fourier transform is a general format of Fourier transform, and add a variables that is order. Our approach is based on volume. The first step of the approach is that voxelize these 3D models. A coarse voxelization is regarded as the input for the 3D discrete factional Fourier transform (3DDFRFT). A set of (complex) coefficient is obtained by 3DDFRFT in each order. The absolute values of coefficients are considered as components of the feature vector in each order. We also can integrate these feature vectors into the mixed feature vector, which is named as Multi-Order fractional Fourier Feature Vector (MOFFFV). We finally present our results and compare our method to 3D descriptor based on 3D Fourier Transform on the Princeton Shape Benchmark database.

Keywords: 3D discrete factional fourier transform, 3D model, feature extraction, retrieval.
Received September 23, 2010; accepted May 24, 2011
Print E-mail

Parrondo’s Paradox Based Strategies in The Serious Game of RTGS Using Forest Fire Model

Saiful Bukhori1, Mochamad Hariadi2, I Ketut Eddy Purnama3, and Mauridhi Hery Purnomo4
1Program Study of Information System, Jember University, Indonesia
2,3,4Electrical Engineering Department, Sepuluh Nopember Institute of Technology, Indonesia


This research proposed parrondo’s paradox strategies in the serious game of Real Time Gross Settlement (RTGS) using forest fire model, which develop the existence of the parrondo paradox and applied in serious game of RTGS system as switching in the settlement process. The settlement process, that our proposed at this paper, is managed by clearing house. The mechanism at clearing house is a transmitter client sends a message of transaction through transmitter bank, that having canal at clearing house, then continue to receiver client through receiver bank by using forest fire model. When settlement process done by one transmitter bank (Process A), the probability of increase Net Worth (NW) is p. When settlement process done by more than one transmitter bank (Process B), we have introduced the probabilities of a self-transition in each state, that is, if the capital is a multiple of three we have a probability r1 of remaining in the same state, whereas if the capital is not a multiple of three then the probability is r2. We will turn to the random alternation of process A and B with probability γ. This will be named as process AB. Examination result of process A change in net worth trend to decrease, process B trend to decrease and process AB that switches randomly between process A and process B trend to increase net worth. Simulation of parrondo’s paradox based strategies in the serious game RTGS using star logo by randomize process A and process B so distribution net worth lot  in the bank that has wealth in intermediate level, total money and total loan trend to rise, total saving loan trend to rise but total wallets trend to decrease.

Keywords: Parrondo paradox, RTGS, forest fire model, net worth.
Received November 23, 2010; accepted May 24, 2011
Print E-mail

Multi-View Gait Based Human Identification System with Covariate Analysis

Hu Ng, Wooi-Haw Tan, and Junaidi Abdullah
 Faculty of Computing and Informatics, Multimedia University, Malaysia


This paper presents a multi-view gait based human identification system. The system is able to perform well under different walking trajectories and various covariate factors such as apparel, loan carrying and speed of walking. Our approach first applies perspective correction to adjust silhouettes from an oblique view to side-view plane. Joint positions of hip, knees and ankles are then detected based on human body proportion. Next, static and dynamic gait features are extracted and smoothed by the Gaussian filter to mitigate the effect of outliers. Feature normalization and selection are subsequently applied before the classification process. The performance of the proposed system was evaluated on SOTON Covariate Database and SOTON Oblique Database from University of Southampton. It achieved 92.1% correct classification rates for both databases.

Keywords: Gait recognition, biometrics, human identification, covariate factors and classification
Received June 7, 2012; accepted March 13, 2013;
Print E-mail

Optical CDMA: Performance of Spectral-Amplitude Coding with new Direct Recovery Scheme using Vector Combinatorial (VC) Code

Hassan Yousif Ahmed1, Mohammed Elmaleeh2, Hilal Adnan Fadhil3, and Syed Aljunid3
1Department of Electrical Engineering, Salam Bin Abdulaziz University, Kingdom of Saudi Arabia
 2Faculty of Engineering, University of Gezira, Sudan
3School of Computer and Communication Engineering, Universiti Malaysia Perlies, Malaysia


In this paper, a new code structure with ideal in-phase cross-correlation for the Spectral-Amplitude-Coding Optical Code-Division Multiple-Access (SAC-CDMA) system called Vectors Combinatorial (VC) code is proposed. VC code is constructed using Euclidian vectors and a simple algebraic way for any positive integer number based on the relationship between the number of users N and the weight W.    We have also studied the performance of OCDMA systems using a newly proposed Direct Recovery Scheme (DRS) under various link parameters. The impact of the detection techniques and data rate effects on the Multi-User Interference (MUI) is reported using a commercial optical systems simulator, Virtual Instrument Photonic (VPITM). The VC code is compared mathematically with other codes which use similar techniques. We analyzed and optimized the data rate, fiber length, and channel spacing in order to reduce the BER effect. A comparison between Complementary and DRS techniques for theoretical and simulation results taken from VPITM is demonstrated.  It is verified that, for a high data rate (higher than 2.5 Gb/s), even if dispersion compensated devices are not deployed, the BER can be significantly improved when the VC code with desired parameters are selected using DRS technique. Also it is found that as the channel spacing width goes from very narrow to wide, the BER decreases and best performance occurs at a spacing bandwidth between 0.8 and 1 nm. In addition, we have shown that the proposed new DRS technique utilizing VC code significantly improves the performance compared with the conventional SAC Complementary subtraction technique.

Received February 1, 2011; accepted July 28, 2011
Print E-mail

A Corpus based Approach to Find Similar Keywords for Search Engine Marketing

Muazzam Siddiqui1, Mohammad Fayoumi2, and Nidal Yusuf3
1Faculty of Computing and Information Technology, King Abdulaziz University, KSA
2Faculty of Computers and Information Systems, Umm Al-Qura University, KSA
3Faculty of Information Technology, Al Isra University, Jordan


Automatic thesaurus generation is used by search engines for query expansion. The same concept is used by search engine marketing companies to suggest keyword terms to their clients to improve the client’s ratings for different search engines. This paper presents and evaluates a corpus based method to find similar terms. The corpus is generated by scraping websites in different categories. A feature selection method is developed that rewards category specific terms and penalizes terms shared by two or more categories. The similarity measure is decomposed into three distinct components, namely contextual, functional and lexical similarities. The contextual similarity measure finds terms that are found in the same context. Functional similarity finds terms on co-occurrence basis while the lexically similar terms share one or more words. An overall similarity measure combines the evidence from these three measures.

Keywords: Information retrieval, text mining, term similarity, search engine marketing.
Received July 3, 2011; accepted May 22, 2012
Print E-mail

The Tracking Algorithm for Maneuvering Target Based on Adaptive Kalman Filter

Zheng Tang, Chao Sun, Zongwei Liu
School of Marine Technology, Northwestern Polytechnical University, China


The application of kalman filter in tracking the maneuver target is not available as it is used in tracking the target of uniform motion. Therefore, a improved method for tracking a maneuver target is proposed. In the proposed approach, the maneuver detector provides the estimate of time instant at which a target starts to maneuver, when a target maneuver is determined, the kalman filter model will be adjusted with varied target motion state. The maneuver, modeled as an acceleration, is estimated recurslvely. Finally, the performance of the proposed approach is shown to be superior to kalman filter by simulation.

Keywords: Adaptive kalman filter, maneuver target tracking, maneuver detector, state estimation.
Received June 29, 2011; accepted December 30, 2011
Print E-mail

Gender Classification in Speech Recognition using Fuzzy Logic and Neural Network

Kunjithapatham Meena1, Kulumani Subramaniam2, and Muthusamy Gomathy3
1Vice Chancellor, Bharathidhasan University, Principal and Director, India
2Department of Computer Application, Shrimathi Indira Gandhi College, India
3Department of Computer Science, Shrimathi Indira Gandhi College, India


Nowadays classification of gender is one of the most important processes in speech processing. Usually gender classification is based on considering pitch as feature. The pitch value of female is higher than the male. In most of the recent research works gender classification process is performed using the abovementioned condition. In some cases the pitch value of male is higher and also pitch of some female is low, in that case this classification does not produce the exact required result. By considering the aforementioned problem we have here proposed a new method for gender classification method which considers three features. The new method uses fuzzy logic and neural network to identify the gender of the speaker. To train fuzzy logic and neural network, training dataset is generated by using the above three features. Then mean value is calculated for the obtained result from fuzzy logic and neural network. By using this threshold value, the proposed method identifies the speaker belongs to which gender. The implementation result shows the performance of the proposed technique in gender classification.

Keywords: Gender classification, fuzzy logic, neural network, energy entropy, short time energy, zero crossing rate.
Received July 16, 2011; accepted December 30, 2011
Print E-mail

Soccer Event Detection via Collaborative Multimodal Feature Analysis and Candidate Ranking

Alfian Abdul Halin1, Mandava Rajeswari2, and Mohammad Abbasnejad3
1Faculty of Computer Science and Information Technology, Universiti Putra Malaysia, Malaysia
2School of Computer Sciences, Universiti Sains Malaysia, Malaysia
3College of Engineering & Computer Science, Australian National University, Australia


This paper presents a framework for soccer event detection through collaborative analysis of the textual, visual and aural modalities. The basic notion is to decompose a match video into smaller segments until ultimately the desired eventful segment is identified. Simple features are considered namely the minute-by-minute reports from sports websites (i.e. text), the semantic shot classes of far and closeup-views (i.e. visual), and the low-level features of pitch and log-energy (i.e. audio). The framework demonstrates that despite considering simple features, and by averting the use of labeled training examples, event detection can be achieved at very high accuracy. Experiments conducted on ~30-hours of soccer video show very promising results for the detection of goals, penalties, yellow cards and red cards.

Keywords: Soccer event detection, sports video analysis, semantic gap, webcasting text.
Received August 20, 2011; accepted December 30, 2011
Print E-mail

Efficient Scheduling Strategy for Task Graphs in Heterogeneous Computing Environment

Saima Ijaz1, Ehsan Ullah Munir1, Waqas Anwar2, and Wasif Nasir1
1Department of Computer Science, COMSATS Institute of Information Technology, Wah Cantt Pakistan
2Department of Computer Science, COMSATS Institute of Information Technology, Abbottabad Pakistan


Today’s multi-computer systems are heterogeneous in nature, i.e., the machines they are composed of, have varying processing capabilities and are interconnected through high speed networks, thus, making them suitable for performing diverse set of computing-intensive applications. In order to exploit the high performance of such a distributed system, efficient mapping of the tasks on available machines is necessary. This is an active research topic and different strategies have been adopted in literature for the mapping problem. A novel approach has been introduced in the paper for the efficient mapping of the DAG-based applications. The approach that takes into account the lower and upper bounds for the start time of the tasks. The algorithm is based on list scheduling approach and has been compared with the well known list scheduling algorithms existing in the literature. The comparison results for the randomly synthesized graphs as well as the graphs from the real world elucidate that the proposed algorithm significantly outperforms the existing ones on the basis of different cost and performance metrics.

Keywords: Directed acyclic graphs, task scheduling, task prioritization, makespan.
Received August 9, 2011; accepted December 30, 2011
Print E-mail

RIO: Rhetorical Structure Theory Based Indexing Technique for Image Objects

Syed Khaldoon Khurshid and Muhammad Shoaib
Department of Computer Science and Engineering, University of Engineering and Technology, Pakistan


Efficient and relevant retrieval of any type of the data, especially multimedia objects, is based on indexing technique. Due to complexity of multimedia data, existing indexing techniques for multimedia objects suffer from irrelevant retrieval. Rhetorical Structure Theory (RST) has already been successfully implemented for indexing text documents, which has reduced irrelevancy. The focus of this research paper is to propose an indexing technique for image objects using RST. An indexing technique is proposed for image objects using image relation framework for RST. To elaborate the functionality of proposed indexing technique, a case study is presented for image objects. Further research can be carried out for indexing technique of other multimedia objects.

Keywords: indexing techniques, RST, image indexing.
Received September 14, 2011; accepted December 30, 2011
Print E-mail

A Key-Policy Attribute-Based Broadcast Encryption

Jin Sun1,2, Yupu Hu1, and Leyou Zhang1
1Department of Application Mathematics, Xi’an University of Technology, China
2Key Lab of Computer Network and Information Security, Xidian University,


According to the broadcast encryption scheme with wide applications in the real world without considering its security and efficiency in the model simultaneously an “unbounded”, Key-Policy Attribute-Based Broadcast Encryption scheme(KP-ABBE) was proposed by combining with waters dual system encryption, attribute-based encryption and broadcast encryption system. Based on the standard model, the scheme can achieve constant-size public parameters, the public parameters do not impose additional limitations on the functionality of the systems (unbounded) and either a small universe size or a bound on the size of attribute sets avoid to fixed at setup. The scheme is proved by using the dual system encryption argument and the four static assumptions which do not depend on the number of queries the attacker makes. The analysis results show that the scheme of this paper is selective secure. China

Keywords: Attribute-based encryption, broadcast encryption, dual system, KP-ABBE, provably secure.
Received March 21, 2011; accepted June 13, 2013;
Print E-mail

Collaborative and Integrated Network and Systems Management: Management Using Grid Technologies

Mohammad Hassan
Faculty of Information Technology, Al-Ahliyya Amman University, Jordan


Current Internet trends are moving towards decentralization of computation, storage, and resources. Supporting network management for such a vast and a highly complex system has become a challenging issue. A management platform has to sufficiently support decentralization, collaboration, and integration. Grid technologies have the potential to serve as management architecture due to the support of the above features.  In this paper, we developed a collaborative network management architecture leveraging the key features of grid technology. Benefiting from this integration, we were able to show that multiple management tasks can be integrated and completed in parallel. This assures the management scalability and efficiency. We also showed that the management information at different networking domains can freely consume the computational resources provided through the grid interface while being executed. Grid interface has guaranteed scalability and reliability for the network management tasks. We have simulated the system prototype and closely studied its efficiency.

Keywords: Grid, network and systems management, integrated management, virtual organizations.
Received September 12, 2011; accepted December 30, 2011
Copyright 2006-2009 Zarqa Private University. All rights reserved.
Print ISSN: 1683-3198.
Warning: fsockopen(): php_network_getaddresses: getaddrinfo failed: Name or service not known in /hsphere/local/home/ccis2k/ on line 251 Warning: fsockopen(): unable to connect to (php_network_getaddresses: getaddrinfo failed: Name or service not known) in /hsphere/local/home/ccis2k/ on line 251 skterr