Tuesday, 31 May 2011 03:02
Effect of Weight Assignment in Data Fusion Based Information Retrieval
Krishnan Batri1, Veerasamy Murugesh2 and Nagammapudur Gopalan3
1Department of Computer Engineering, European
University of Lefke, Turkey
2Department of Microsoft Information Technology, Keimyung University,
Republic of Korea
3Department of Computer Applications, National Institute of
Technology, India
Abstract: Variation in performances
of an Information Retrieval system, which merges results from a number of
retrieval schemes possessing equal and unequal weights, is studied in this
paper. Weight of the retrieval schemes for a particular document is derived
from the relevance scores of that corresponding document. Since, the relevance
scores are varying from document to document and corpus to corpus, the method
proposed is dynamic. A number of weight calculation methods, which are using
the error value for computation purpose, are discussed in this paper. The
effectiveness of the weight calculation is tested over three benchmark test collections
viz., ADI, CISI and MED. It has been identified that the methods discussed in
this paper retrieve articles effectively and they are independent of history or
any training data.
Keywords: Data fusion, information retrieval, relevance score, similarity measure, and weight assignment.
Received September 16, 2008; accepted August 3, 2009
Published in
vol 8, no. 3, july 2011
Wednesday, 09 June 2010 19:00
Prediction of Boiler Output Variables Through
Prediction of Boiler Output Variables Through
the PLS Linear Regression Technique
Abdelmalek Kouadri1, Mimoun Zelmat1, and AlHussein Albarbar2
1Applied Control Laboratory, University of Boumerdes, Algeria
2Department of Engineering and Technology, Manchester Metropolitan University, UK
1Applied Control Laboratory, University of Boumerdes, Algeria
2Department of Engineering and Technology, Manchester Metropolitan University, UK
Abstract: In this work, we propose to use the linear regression partial least square method to predict the output variables of the RA1G boiler. This method consists in finding the regression of an output block regarding an input block. These two blocks represent the outputs and inputs of the process. A criterion of cross validation, based on the calculation of the predicted residual sum of squares, is used to select the components of the model in the partial least square regression. The obtained results illustrate the effectiveness of this method for prediction purposes.
Keywords:Partial least square, principal component analysis, principal component regression, covariance, predicted residual sum of squares.
Received November 26, 2008; accepted May 17, 2009
Published in
vol 8, no. 3, july 2011
Wednesday, 09 June 2010 19:00
Zest Based Visualization for
Zest Based Visualization for
Requirements Negotiation
Mohammed Alharthi and Siti Salim
Faculty of Computer Science and Information Technology, University of Malaya, Malaysia
Faculty of Computer Science and Information Technology, University of Malaya, Malaysia
Abstract: Win-Win is one of the most common models which help systems stakeholders resolve conflicts during system Requirements Negotiation (RN). Solving conflicts, however, is barely dependent on visualizing the negotiation itself. The Zest algorithm was invented to visualize a standard e-mail discussion and demonstrate a concise overview of the discussion to facilitate a more productive one. This research sought to use this algorithm in visualizing the RN process in groupware. We have formulated a conceptual diagram on which to base the design and development process of the groupware for implementing the Zest algorithm. Our implementation of the algorithm has later been enhanced to produce more types of visualizations in relation to RN.
Keywords: Win-Win model, Requirements negotiation, visualization, Zest algorithm, groupware.
Received August 27, 2009; accepted January 3, 2010
Published in
vol 8, no. 3, july 2011
Tuesday, 09 February 2010 19:00
Cognitive Filtering of Textual Information Agents Based Implementation
Cognitive Filtering of Textual Information Agents Based Implementation
Omar Nouali1, Nadia Nouali-Taboudjemat1, and Bernard Toursel2
1Theory and computer engineering, DTISI, CERIST, Algeria
2University of Sciences and Technologies of Lille, France
1Theory and computer engineering, DTISI, CERIST, Algeria
2University of Sciences and Technologies of Lille, France
Abstract: The study presented in this paper has multiple objectives. The first objective is to automate the information filtering process by taking into account the relative importance of information and resources needed for its treatment. The second one is to demonstrate the applicability and contribution of an agents based implementation to automatic information filtering. The third one is to show how learning can improve the effectiveness of filtering and that automatic learning is necessary in the design of automatic information filtering systems. We propose an open, dynamic and evolving solution that offers to the filtering process the opportunity to learn, exploit the learned knowledge and adapt itself to the application environment. We have adopted agents to improve the response time compared to a sequential algorithmic solution. To validate our filtering approach, we led a set of experiments to evaluate performances of the techniques and tools we have developed.
Keywords: Information filtering, machine learning, linguistic agents, and filtering criteria.
Received March 25, 2009; accepted August 4, 2009
Published in
vol 8, no. 3, july 2011
Tuesday, 09 February 2010 19:00
A Novel Mobile Crawler System Based
A Novel Mobile Crawler System Based
on Filtering off Non-Modified Pages for
Reducing Load on the Network
Rajender Nath1 and Satinder Bal2
1Department of Computer Science and Applications, Kurukshetra University, India
2Department of Computer Science and Applications, Vaish College of Engineering, India
1Department of Computer Science and Applications, Kurukshetra University, India
2Department of Computer Science and Applications, Vaish College of Engineering, India
Abstract: The studies in the literature show that about 40% of the current Internet traffic and bandwidth consumption is due to web crawlers that retrieve pages for indexing by the different search engines. This traffic and bandwidth consumption will increase in future due to the exponential growth of the web. This paper addresses the problem of bandwidth consumption by introducing an efficient indexing system based on mobile crawlers. The proposed system employs mobile agents to crawl the pages. These mobile agent based crawlers retrieve the pages, process them, compare their data to filter out pages that are not modified after last crawl, and then compress them before sending them to the search engine for indexing. The experimental results of the proposed system are very encouraging.
Keywords: Search engine, mobile crawler, network resources, internet traffic.
Received March 17, 2009; accepted August 3, 2009
Published in
vol 8, no. 3, july 2011
Tuesday, 09 February 2010 19:00
Self-organization and Topology's Control
Self-organization and Topology's Control
for Mobile Ad-hoc Networks
Maâmar Kef1, Leila Chergui1, and Mohamed Benmohammed2
1Department of Computer Sciences, University Larbi Ben Mhidi, Algeria
2Department of Computer Sciences, University Mentouri, Algeria
1Department of Computer Sciences, University Larbi Ben Mhidi, Algeria
2Department of Computer Sciences, University Mentouri, Algeria
Abstract: The study and the implementation of routing algorithms to ensure the connexion of Ad-hoc networks as often defined (every summit may reach any other) is a hard task to achieve. The environment is dynamic and evolves through time; the topology of network may change frequently. Two distinct approaches tried to apprehend this problem. The first one is based on a flat vision of network, and the second one on an auto-organisation structure. This paper deals with the study of auto-organisation solution for Ad-hoc networks, and the benefit effects brought by this last to the routing-process.
Keywords: Auto-organization, topology's control, routing-process, and ad-hoc networks.
Received April 12, 2008; accepted August 4, 2008
Published in
vol 8, no. 3, july 2011
Tuesday, 09 February 2010 19:00
A Flexible Design of Network Devices Using Reconfigurable Content Addressable Memory
A Flexible Design of Network Devices Using Reconfigurable Content Addressable Memory
Qutaiba Ali
Computer Engineering Department, University of Mosul, Iraq
Computer Engineering Department, University of Mosul, Iraq
Abstract: The content addressable memory is a memory unit that uses content matching instead of addresses. content addressable memory are used in different networking, telecommunications and storage applications because of their parallel, fast search capabilities. This paper presents a new method (called array method) for designing Reconfigurable Content Addressable Memory (RCAM). The behavior of the new method was described using VHDL and implemented using FPGA technique. Then, the performance of the method was compared to other traditional content addressable memory design methods. The proposed RCAM is configured and used as the main part of different network security devices and units (ethernet address filtering unit, WLAN MAC address filtering unit ,firewall on chip, QoS packet classification unit, routing table search unit and network intrusion detection system search unit). The successful implementations of the suggested RCAM prove its high flexibility to be used in different network applications.
Keywords: Content addressable memory, network security devices, FPGA, throughput.
Received August 18, 2008; accepted November 5, 2009
Published in
vol 8, no. 3, july 2011
Tuesday, 09 February 2010 19:00
The Chained-Cubic Tree Interconnection Network
The Chained-Cubic Tree Interconnection Network
Malak Abdullah1, Emad Abuelrub2, and Basel Mahafzah3
1Department of Computer Science, Jordan University of Science and Technology, Jordan
2Department of Computer Science, Zarqa Private University, Jordan
3Department of Computer Science, The University of Jordan, Jordan
1Department of Computer Science, Jordan University of Science and Technology, Jordan
2Department of Computer Science, Zarqa Private University, Jordan
3Department of Computer Science, The University of Jordan, Jordan
Abstract: The core of a parallel processing system is the interconnection network by which the system’s processors are linked. Due to the great role played by the interconnection network’s topology in improving the parallel processing system’s performance, various topologies have been proposed in the literature. This paper proposes a new interconnection network topology, referred to as the chained-cubic tree, in which chains of hypercubes are arranged in a tree structure. The major topological properties of the proposed topology have been investigated, including its diameter, degree, connectivity, bisection width, size, cost, and hamiltonicity. A comparative study is then conducted between the proposed Chained-Cubic Tree (CCT) and other interconnection networks’ topologies, including tree and hypercube in order to evaluate the rank occupied by CCT among other well-known topologies in terms of various performance and cost metrics. The concluding results proved that the CCT topology overcomes the shortcomings of its progenitors, tree and hypercube, while keeping most of its appealing properties.
Keywords: Chained-cubic tree, hypercube, tree, interconnection network topology.
Received October 15, 2009; accepted November 5, 2009
Published in
vol 8, no. 3, july 2011
Tuesday, 09 February 2010 19:00
State of Component Models Usage: Justifying the Need for a Component Model Selection Framework
State of Component Models Usage: Justifying the Need for a Component Model Selection Framework
Hazleen Aris1 and Siti Salim2
1College of Information Technology, Universiti Tenaga Nasional, Malaysia
2Faculty of Computer Science and Information Technology, University of Malaya, Malaysia
1College of Information Technology, Universiti Tenaga Nasional, Malaysia
2Faculty of Computer Science and Information Technology, University of Malaya, Malaysia
Abstract: The presence of a large number of component models has caused some difficulties in selecting suitable component models to be used, either for research purpose or for software development purpose. Lack of framework or standard that can be used to guide the process of selecting suitable component models is believed to be one of the reasons that have caused the difficulties. Therefore, in this article, the need for a component model selection framework is justified. The selection framework can be applied by the software developers to help them determine suitable component models to be used in their software development projects. Possible contributions of the framework to the research and software development industry communities are also identified. These are achieved by examining the current state of component models usage in both research and software development industry communities, which are obtained from a number of related resources found from exhaustive literature search.
Keywords: Component model usage, component model selection, selection framework, component model survey, component model review, and component model state.
Received July 3, 2009; accepted November 5, 2009
Published in
vol 8, no. 3, july 2011
Saturday, 14 February 2009 19:00
Detection and Compensation of Undesirable Discontinuities within the Farsi/Arabic Subwords
Detection and Compensation of Undesirable Discontinuities within the Farsi/Arabic Subwords
Majid Ziaratban and Karim Faez
Electrical Engineering Department, Amirkabir University of Technology, Iran
Electrical Engineering Department, Amirkabir University of Technology, Iran
Abstract: In this paper, an unexplored subject in the domains of Farsi/Arabic handwritten word preprocessing is introduced. Subwords play a vital role in many applications such as cheque amount recognition, text recognition, lexicon reduction and subword-based word recognition. Correcting the faults occurred in subwords will improve the overall performance of these applications. A subword is a connected-component in the main body of a word. The occurrence of a discontinuity in a subword, divides the subword into two isolated parts. These parts are detected as two incorrect subwords. In our algorithm, before correcting these faults, the baseline of each subword is corrected using the proposed baseline correction method. Then, to limit the exploration area in matching stage, the dots are removed. Undesirable discontinuities in subwords are detected by using a template matching algorithm. Disconnected parts of a subword are joined together by using three different methods. Experiments show that the cubic polynomial-based compensation method causes the best results and 2.87 % improvement in the subword recognition rate.
Keywords: Detection, compensation, Farsi/Arabic subword, and cubic polynomial curve fitting.
Received May 12, 2009; accepted November 5, 2009
Published in
vol 8, no. 3, july 2011
Saturday, 14 February 2009 19:00
On the Genus of Pancake Network
On the Genus of Pancake Network
Quan Nguyen1 and Saïd Bettayeb2
1University of Texas at Dallas, Richardson, USA
2University of Houston at Clear Lake, Houston, USA
1University of Texas at Dallas, Richardson, USA
2University of Houston at Clear Lake, Houston, USA
Abstract: Both the pancake graph and star graph are Cayley graphs and are especially attractive for parallel processing. They both have sublogarithmic diameter, and are fairly sparse compared to hypercubes. In this paper, we focus on another important property, namely the genus. The genus of a graph is the minimum number of handles needed for drawing the graph on the plane without edges crossing. We will investigate the upper bound and lower bound for the genus of pancake graph and compare these values with the genus of the star graph as well as that of the hypercube.
Keywords: Genus, binary hypercube, permutation, pancake network, cayley graph, and prefix reversal.
Received April 28, 2009; accepted November 5, 2009
Published in
vol 8, no. 3, july 2011
Saturday, 14 February 2009 19:00
DWT-Based Audio Watermarking
DWT-Based Audio Watermarking
Ali Al-Haj1, Ahmad Mohammad1, and Lama Bata2
1College of Electrical Engineering, Princess Sumaya University, Jordan
2Faculty of Engineering and Technology, University of Jordan, Jordan
1College of Electrical Engineering, Princess Sumaya University, Jordan
2Faculty of Engineering and Technology, University of Jordan, Jordan
Abstract: TMany effective watermarking algorithms have been proposed and implemented for digital images and digital video, however, few algorithms have been proposed for audio watermarking. This is due to the fact that, the human audio system is far more complex and sensitive than the human visual system. In this paper, we describe an imperceptible and robust audio watermarking algorithm based on the discrete wavelet transform. Performance of the algorithm has been evaluated extensively, and simulation results are presented to demonstrate the imperceptibility and robustness of the proposed algorithm.
Keywords: Multimedia, copyright protection, audio watermarking, imperceptibility, robustness, transform-domain watermarking, and discrete wavelet transform.
Received September 16, 2009; accepted January 3, 2010
Published in
vol 8, no. 3, july 2011
Saturday, 14 February 2009 19:00
A Hierarchical K-NN Classifier for Textual Data
A Hierarchical K-NN Classifier for Textual Data
Rehab Duwairi1 and Rania Al-Zubaidi2
1Jordan University of Science and Technology, Jordan
2Jordan University of Science and Technology, Jordan
1Jordan University of Science and Technology, Jordan
2Jordan University of Science and Technology, Jordan
Abstract: This paper presents a classifier that is based on a modified version of the well known K-Nearest Neighbors classifier (K-NN). The original K-NN classifier was adjusted to work with category representatives rather than training documents. Each category was represented by one document that was constructed by consulting all of its training documents and then applying feature selection so that only important terms remain. By this, when classifying a new document, it is required to be compared with category representatives and these are usually substantially fewer than training documents. This modified K-NN was experimented with in a hierarchical setting, i.e., when categories are represented as a hierarchy. Also, a new document similarity measure was proposed. It focuses on co-occurring or matching terms between a document and a category when calculating the similarity. This measure produces classification accuracy compared to the one obtained if the cosine, Jaccard or Dice similarity measures were used; yet it requires a much less time. The TrechTC-100 hierarchical dataset was used to evaluate the proposed classifier.
Keywords: Text categorization, hierarchical classifiers, K-NN, similarity measures, category representatives.
Received October 23, 2008; accepted August 3, 2009
Published in
vol 8, no. 3, july 2011
Saturday, 14 February 2009 19:00
Investigation of Golay Code (8, 12, 24) Structure
Investigation of Golay Code (8, 12, 24) Structure
in Improving Search Techniques
Eyas El-Qawasmeh1, Maytham Safar2, and Talal Kanan3
1College of Computer and Information Sciences, King Saud University, Saudi Arabia
2Department of Computer Science, Jordan University of Science and Technology, Jordan
3Department of Computer Engineering, Kuwait University, Kuwait
1College of Computer and Information Sciences, King Saud University, Saudi Arabia
2Department of Computer Science, Jordan University of Science and Technology, Jordan
3Department of Computer Engineering, Kuwait University, Kuwait
Abstract: This paper presents a new technique for hash searching that is designed for approximate matching problem of multi-attribute objects. The suggested technique can be used for improving the search operation when the multi-attribute objects are partially distorted or when the searching criterion is not specified properly. The suggested approach is based upon reversing the conventional scheme of Golay code (8, 12, 24), which maps 24-bit vectors into 12-bit message words. In this technique, a multi-dimensional space is used to represent objects, where each object is given by a 24-bit vector. The closeness of the objects is determined by partitioning a 24-dimensional cube. In addition, the possibility of 1-bit distortions is considered through bit modifications of the 24-bit vectors. Thus, we proposed a hash table of 4096 entries that is fault-tolerant in the sense that the index is the same for any two different 24-bit vectors that share the same sphere. This allows organizing a direct retrieval of a neighborhood of 24-bit vectors with two or possibly more mismatches. The simulation experiments measured the retrieval capabilities of the proposed system and the expected hash distribution.
Keywords: Golay code, information retrieval, distortion, and hashing.
Received December 25, 2008; accepted August 3, 2009
Published in
vol 8, no. 3, july 2011
Saturday, 14 February 2009 19:00
Distributed Efficient Multi Hop Clustering
Distributed Efficient Multi Hop Clustering
Protocol for Mobile Sensor Networks
Shahzad Ali and Sajjad Madani
Department of Computer Science, COMSATS Institute of Information Technology, Pakistan
Department of Computer Science, COMSATS Institute of Information Technology, Pakistan
Abstract: This paper presents a Distributed Efficient Multi hop Clustering (DEMC) protocol for mobile wireless sensor networks. An overwhelming majority of current research on sensor networks considers static networks only, while we consider mobile environment. DEMC is distributed, works well with mobile nodes, and has a recovery mechanism that is used to reduce the packet loss during inter cluster communication. The recovery mechanism also improves the connectivity between cluster heads during inter cluster communication. On average, each node sends less than one message during clustering, and does not rely on periodic hello messages. As a result reducing number of transmissions leads to energy efficiency. Simulation results show that DEMC is energy efficient, incurs less packet loss, increases packet delivery ratio, and exhibits robustness against moderate to high mobility of nodes.
Keywords: Distributed routing protocol, mobile wireless sensor networks, mobility, clustering.
Received June 17, 2009; accepted November 5, 2009
Published in
vol 8, no. 3, july 2011