Comparisons Between Data Clustering Algorithms 

Osama Abu Abbas

Computer Science Department, Yarmouk University, Jordan

Abstract: Clustering is a division of data into groups of similar objects. Each group, called a cluster, consists of objects that are similar between themselves and dissimilar compared to objects of other groups. This paper is intended to study and compare different data clustering algorithms. The algorithms under investigation are: k-means algorithm, hierarchical clustering algorithm, self-organizing maps algorithm, and expectation maximization clustering algorithm. All these algorithms are compared according to the following factors: size of dataset, number of clusters, type of dataset and type of software used. Some conclusions that are extracted belong to the performance, quality, and accuracy of the clustering algorithms.  

Keywords: Clustering, k-means algorithm, hierarchical clustering algorithm, self-organizing maps algorithm, expectation maximization clustering algorithm. 

Received January 18, 2007; accepted May 2, 2007 

Incremental Transitivity Applied to Cluster Retrieval

Yaser Hasan¹, Muhammad Hassan², and Mick Ridley¹

¹Computing Department, University of Bradford, UK

²Computer Science Department, Zarqa Private University, Jordan

Abstract: Many problems have emerged while building accurate and efficient clusters of documents; such as the inherent problems of the similarity measure, and document logical view modeling. This research is an attempt to minimize the effect of these problems by using a new definition of transitive relevance between documents; i.e., adding more conditions on transitive relevance judgment through incrementing the relevance threshold by a constant value at each level of transitivity. Proving the relevance relation to be transitive, will make it an equivalence relation that can be used to build equivalence classes of relevant documents. The main contribution of this paper is to use this definition to partition a set of documents into disjoint subsets as equivalence classes (clusters). Another contribution is by using the incremental transitive relevance relation; the traditional vector space model can be made incrementally transitive.

A Knowledge-Based Advisory System for Software Quality Assurance  

Khalid Eldrandaly Information Systems Department,

 Zagazig University, Egypt 

Abstract: Software quality assurance is a planned and systematic approach to ensure that software processes and products confirms to the established standards, processes, and procedures. The goals of software quality assurance are to improve software quality by appropriately monitoring both software and the development process to ensure full compliance with the established standards and procedures. There are several models for software quality assurance, such as the ISO/IEC 90003, and the capability maturity model integration. However, the proper implementation of these models is often a difficult and a costly task for software companies especially small and medium ones.  This paper describes a prototype knowledge-based advisory system designed to play the role of a “virtual quality editor” to help individuals, organizations, and software companies who desire to implement these quality models. By identifying the gauging absence of prerequisites between the prerequisites and what actually exists in the present environment, this advisory system provides assessment results and suggestions to the companies so that successful implementation of these quality models can be achieved. Component object model technology is used in designing and integrating the different components of the prototype system to assure software interoperability between these components. The architecture, the development and the implementation of the prototype system are discussed in details. A typical example is also presented to demonstrate the application of the prototype system.  

Keywords: Expert systems, software quality assurance, ISO/IEC 90003, capability maturity model integration, component object mode.

 Received December 9, 2006; accepted April 28, 2007

Developing Approaches of Supply Chain Management Systems of Enterprises in Pakistan

  Danish Irfan^{1, 2}, Xu Xiaofei¹, and Deng Sheng Chun¹

¹School of Computer Science and Engineering, Harbin Institute of Technology, China

²Department of Computer Science, COMSATS Institute of Information Technology, Pakistan 

Abstract: This paper identifies the problems of pakistani enterprises concerning supply chain networks and orders solutions to improving its overall performance and competitiveness. Seldom, a non-trivial amount of research efforts has been made in emergent nations to improve and enhance supply chain management paradigm. Our survey comprised nine enterprises; among them four are successfully modelled and presented. The detailed supply chain structure and study is also made. The paper concludes by suggesting and adopting modelled approach to the corporations in their move towards implementing supply chain management strategies. 

Keywords: Supply chain, supply chain management, supply chain networks.

 Received October 14, 2006; accepted April 25, 2007

A SCOR Reference Model of the Supply Chain Management  System in an Enterprise

  Danish Irfan^{1, 2}, Xu Xiaofei¹, and Deng Sheng Chun¹

¹School of Computer Science and Engineering, Harbin Institute of Technology, China

²Department of Computer Science, COMSATS Institute of Information Technology, Pakistan 

Abstract: State-of-the-art supply chain management systems can be obtained from the supply chain operations reference-model for business applications, advancement and practices. Problematic management processes in standard process reference model form improves competitive advantage, communication, dimensions, management, control and alter to a specific purpose among the supply chain management processes. This helps organizations capture the "as-is" state of a process with the objective of achieving the desired "to-be" future state. This paper presents supply chain management efforts, key challenges and opportunities in pakistan’s industrial and organizational most successful sector and span of business ‘from seed to smoke’ by deploying the eminent supply chain operations reference-model.

Keywords: Supply chain management, supply chain operations reference-model, supply chain.

 Received October 2, 2006; accepted April 25, 2007

Toward a Multi-Temporal Approach for Satellite Image Interpretation  

Saheb Karim^1,2, Farah Riadh¹, Bassel Solaiman², and Ben-Ahmed Mohamed¹

¹Ecole Nationale des Sciences de l’Informatique, Campus Universitaire de Manouba, Tunis

²Laboratoire ITI, Technopôle Brest Iroise CS 83818, France

 Abstract: Multi-temporal remotely sensed images have proved to be of great interest for earth resource assessment and environment monitoring. Automatic or semi-automatic interpretation of these images becomes an important and complex task in computer vision for land use change detection. However, facing the complexity of such images, many classical image interpretation techniques become inefficient. In this paper, the proposed approach, based on multi-agent system and hierarchical blackboard architecture allows an intelligent, concise and flexible control of a multi-temporal scene interpretation. It proposes the combination of semantic network representing the generic description of the scene, and a state transition diagram, modeling the possible state transitions for each one of the classes of interest. This system produces a hierarchic description of the results as well as the structural context of the identified objects including the associated attributes. We illustrate the design and implementation of our system on a set of multi-temporal satellite images SPOT4 representing a center tunisian region for different dates in order to illustrate the potential of the proposed multi-temporal approach. 

Keywords: Multi-temporal remotely sensed images, interpretation, hierarchical blackboard architecture, multi-agent system. 

Received February 15, 2007; accepted April 24, 2007

A Source Tree Reliable Multicast Protocol for Ad-Hoc Networks

Tariq Al-Ahdal, Shamala Subramaniam, Mohamed Othman, and Zuriati Zukarnain

Department of Communication Technology and Networks, University Putra Malaysia, Malaysia. 

Abstract: Multicasting is an essential service for ad-hoc wireless networks. Many reliable multicast schemes were studied in order to reduce packet losses in the network. This paper describes our effort to build a source tree reliable multicast protocol for ad-hoc networks. Source tree reliable multicast protocol provides the delivery of an ordered contiguous sequence of data packets from one sender to many receivers in an ad-hoc network. It is designed to support applications based on bulk data transfer, like files, images and software packages. The core to its support of node mobility, and also what makes the protocol unique, is the dynamic selection of a sub set of 1-hop neighbors from the sender as its forward servers. The key idea behind the selection of this sub set 1-hop neighbors is to forward the retransmitted lost data packets that are needed by some receivers to achieve a higher throughput and to receive the acknowledgment packet from receivers to avoid the acknowledgment-implosion problem inherent in any reliable multicast scheme. Finally, simulation results show that the protocol has a high delivery ratio and low end-to-end delay compared to ReMHoc protocol. 

Keywords: Ad-hoc network, forward server, reliable multicast. 

Received January 10, 2007; accepted April 24, 2007

Modelling of Long Wavelength Detection of Objects Using Elman Network Modified Covariance Combination  

Lubna Badri and Mujahid AL-Azzo

Faculty of Engineering, Philadelphia University, Jordan

Abstract: The problem of spatially detection and imaging of closely separated buried objects is investigated. A high resolution modified covariance method is employed. A recurrent neural network is used as a preprocessing technique to decrease the effect of concealing media on the results. The in-line holography is applied to increase the signal to noise ratio.  Different concealing media and different values of signal to noise ratio are used to investigate the performance of such combination experimental results show that pre-processing the noisy data with recurrent neural network improves the performance. 

Keywords: Modified covariance method, recurrent neural network (RNN), in-line holography. 

Agent Based Bioinformatics Integration Using RETSINA 

Kulathuran Shunmuganathan¹, Kumar Deepika², and Kumaradhas Deeba²

¹Anna University, India.

²Noorul Islam College of Engineering, India

Keywords: Wrapper, system integration, agent technology, mediator, RETSINA. 

Multi-Segment Steganography Technique 

Integrating Internet into Traditional Education: A Practical Study of University Students’ Usage and Attitudes 

Jebreen Mohammed¹ and Jamal Al-Karaki²

¹Department of Curriculum and Instructions, The Hashemite University, Jordan

²Department of Computer Engineering, The Hashemite University, Jordan

Abstract: This paper describes a study conducted on the attitudes of students towards the integration of the web as a channel of communication and a study tool in traditional teaching at the Hashemite University of Jordan. In particular, the study aims at studying the effect of certain variables such as gender, major, computer experience, and the percentage of internet usage on the learning process.  A survey was carried out with a sample of 502 male and female of the university students.  A comprehensive and reliable questionnaire was designed and distributed to the students' sample. Drawing on evidence from lesson observations, teacher/student interviews and project reports, this paper also examines the university perspectives and strategies to incorporate use of internet resources and associated information and communication technology tools into humanities, social, and scientific studies. Amazingly, the results revealed a low percentage of internet usage in university education.  Also, the results showed that computer experience mainly affects how internet is used in the learning process. However, there were no significant differences in the results regarding the gender and major or even the bi- or tri-interactions between the variables under consideration.  The study also showed that email correspondence and internet search are the most common uses of internet in the university. Furthermore, our findings support the following assumptions: The Web cannot substitute entirely for face-to-face learning, but it can serve as a reasonable alternative when the latter is unavailable. The paper concludes with a list of recommendations.  

Keywords: Technology-based teaching, internet in education, world wide web, user attitude.  

Using Recommendation Systems in Course Management Systems to Recommend Learning Objects 

Jamil Itmazi¹ and Miguel Megías²

¹Department of Information Technology, Palestine Polytechnic University, Palestine

²Department of Computer Languages and Systems, University of Granada, Spain 

Abstract: Recommendation systems have been widely used in many internet activities and their importance is increasing due to the information overload problem arising from internet. This paper gives background and some example of the RS current usage domains, the different recommendation systems approaches. In addition, the paper discusses the recommendation systems ability using in the learning management systems / course management system to support students’ needs, as well as it discusses the suitability of every recommendation systems approach to recommend learning objects, finally it presents a design of a proposal structure of recommendation systems in learning management systems which could be recommend a suitable learning objects (e.g., courses) to students from a big list of those learning objects.  

Keywords: Recommendation systems, learning management systems, course management system, e-learning platform. 

Received December 25, 2006; accepted April 21, 2007

Simulation Study of a Heuristic Near-Maximum Ant-Based Dynamic Routing

 Tan Chin¹, Fouad Abbou², and Ewe Tat¹

¹Faculty of Information Technology, Multimedia University, Malaysia

²Alcatel Network Systems, Kuala Lumpur, Malaysia 

Abstract: A new ant-based routing is proposed for solving dynamic routing and wavelength assignment in mesh WDM network under the wavelength continuity constraint. The ant algorithm favors paths with maximum number of available wavelengths between two nodes, resulting in improved load balancing and less congested shortest path. The simulation results showed that the proposed ant-based routing algorithm is highly reliable in the sense that the number of ants used is predictable to achieve a steady performance in terms of blocking probability.  

Keywords: Ant colony optimization, ACO, dynamic routing, WDM.  

Trust Based Prioritization of Quality Attributes

Punam Bedi and Vibha Gaur

Department of Computer Science, University of Delhi, India

 Abstract:  Advances in network and distributed technology has revolutionized the advent of multi-agent systems. Nowadays, agent technology has been applied in revising and restructuring the complex software systems. Trust plays a key role in multi-agent systems as agents will be representing or acting on behalf of users or owners with different goals and motivations. Tasks are delegated to software agents that achieve their common goals based on interaction among themselves. Current trust models focus on the limited aspect of stakeholders trust in multi-agent systems; for example security issues of the entities involved in system while ignoring the other issues such as knowledgeability, usability, persistency, accessibility, performance, collaboration, realism and integrity.  This paper identifies the quality attributes of Multi-agent system, which foster the stakeholders trust in the system and hence establishes the multifaceted definition of trust in MAS. Building trust from the quality attributes would encourage the developer to induce these quality attributes in the MAS development life cycle and produce a system whose foundation will be the stakeholders trust. This paper also prioritizes the quality attributes of MAS in order of their impact on stakeholders’ trust. 

Keywords: Multi-Agent Systems (MAS), Intuitionistic Fuzzy Sets (IFS), trust, quality attribute. 

Received September 8, 2006; accepted June 16, 2007

Design of Linear Antenna Arrays for Side Lobe Reduction Using the Tabu Search Method

 Lotfi Merad, Fethi Bendimerad, and Sidi Meriah 

Faculty of Engineering, University of Abou-Bekr Belkaïd-Tlemcen, Algeria

Abstract: A design problem of uniform linear antenna arrays for maximal side lobe level reduction with the constraint of a fixed beam width is dealt with. This problem is modelled as a simple optimization problem. The method of tabu search is used to determine an optimum a set of weights of antenna element that provide a radiation pattern with maximal side lobe level reduction with the constraint of a fixed beam width. Numerical results show the effectiveness of the proposed method although it needs more CPU time and memory.

 Keywords: Linear antenna arrays, tabu search, side lobe level, optimization, radiation pattern. 

Received July 1, 2006; accepted November 21, 2006