Vol. 12 No. 8 AUG 2014

Vol. 12 No. 8 AUGUST 2014 International Journal of Computer Science and Information Security

Publication AUGUST 2014, Volume 12 No. 8 (Download Full Journal) (Archive) (Download 2)

Copyright © IJCSIS. This is an open access journal distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

1. Paper 31071429: Confidential Algorithm for Golden Cryptography Using Haar Wavelet (pp. 1-29)

Full Text: PDF

Marghny H. Mohamed & Yousef B. Mahdy, Computer Science, Faculty of Computer and Information Science, Assuit University

Wafaa Abd-Elwahed, Information Systems, Faculty of Computer and Information Science, Assuit University

Abstract — One of the most important consideration techniques when one want to solve the protecting of digital signal is the golden matrix. The golden matrices can be used for creation of a new kind of cryptography called the golden cryptography. Many research papers have proved that the method is very fast and simple for technical realization and can be used for cryptographic protection of digital signals. In this paper, we introduce a technique of encryption based on combination of haar wavelet and golden matrix. These combinations carry out after compression data by adaptive Huffman code to reduce data size and remove redundant data. This process will provide multisecurity services. In addition Message Authentication Code (MAC) technique can be used to provide authentication and the integrity of this scheme. The proposed scheme is accomplished through five stages, the compression data, key generation, encryption stage, the decryption stage and decompression at communication ends.

Keywords: Cryptography, Golden matrix, Adaptive Huffman Compression, Haar wavelet, Message Authentication Code(MAC)

2. Paper 31071434: A Secure Attribute-based Model to Foster Collaboration in Healthcare Systems (pp. 10-19)

Full Text: PDF

Sara Najam, RITM Laboratory-ESTC, ENSEM-Hassan II University, Casablanca, Morocco

Hajar Mousannif, LISI Laboratory-FSSM, Cadi Ayyad University, Marrakesh, Morocco

Mohamed Ouzzif, RITM Laboratory-ESTC, Hassan II University, Casablanca, Morocco

Abstract — In today´s rapidly-evolving globalized world, there is an undeniable trend towards establishing secure distributed collaborative work environments. In the prototypical example of health care institutions, critical research needs involve both effective collaboration modeling, and highly secured dynamic interactions insurance. In this paper, we introduce a new system design that enables both synchronous and asynchronous secure communication between different entities in a collaborative work environment. The proposed system provides a fine-grained attribute-based access control model to secure the collaboration in distributed systems, namely in Computer Supportive Cooperative Work (CSCW) systems. We opted for breast cancer diagnosis as a case study to apply our system design. Through a clear specification model of our system, we highlight the feasibility of achieving real-time and secure breast cancer diagnosis process, in which several medical organizations are engaged.

Keywords— Healthcare; CSCW systems; ABAC; Collaboration; Security

3. Paper 31071431: A two-stage architecture for stock price forecasting by combining SOM and fuzzy-SVM (pp. 20-25)

Full Text: PDF

Duc-Hien Nguyen, Hue University, Hue, VietNam

Manh-Thanh Le, Hue University, Hue, VietNam

Abstract — This paper proposed a model to predict the stock price based on combining Self-Organizing Map (SOM) and fuzzy – Support Vector Machines (f-SVM). Extraction of fuzzy rules from raw data based on the combining of statistical machine learning models is the base of this proposed approach. In the proposed model, SOM is used as a clustering algorithm to partition the whole input space into several disjoint regions. For each partition, a set of fuzzy rules is extracted based on a f-SVM combining model. Then fuzzy rules sets are used to predict the test data using fuzzy inference algorithms. The performance of the proposed approach is compared with other models using four data sets.

Keywords- Fuzzy rules; Support vector machine - SVM; Self-Organizing Map - SOM; Stock price forecasting; Data-driven model

4. Paper 31071430: Limitations of Current Security Measures to Address Information Leakage Attacks (pp. 26-32)

Full Text: PDF

Omar Hussein, Nermin Hamza, Hesham Hefny

Computer and Information Sciences Department, Institute of Statistical Studies and Research, Cairo University, Egypt

Abstract — Information leakage attacks represent a serious threat for their widespread and devastating effects. Their significance stems from the fact that they are committed by an organization’s authorized computer users, and/or processes executing on their behalf. The diverse avenues that could be exploited to carry out such attacks add another barrier towards addressing them. Based on literature review, this paper explores strengths of security measures intended to confront information leakage attacks, and focuses on pinpointing their respective limitations. It demonstrates that only few of them are capable of mitigating such attacks, whereas the rest suffer from conceptual and/or implementation-related limitations that render them vulnerable to circumvention. They are basically prone to high false positive and/or false negative rates, complex to apply, inflexible during execution, suffer from degraded performance, or require hardware modification. Most importantly, neither of them provides a remedy for new undetected malicious software, nor the ever increasing insider threat.

Index Terms—Information Security, Information Leakage, Security Measures, Security Limitations

5. Paper 31071428: Solving the Problem of the K Parameter in the KNN Classifier Using an Ensemble Learning Approach (pp. 33-39)

Full Text: PDF

Ahmad B. A. Hassanat (1), Mohammad Ali Abbadi (2), Ghada Awad Altarawneh (3)

(1, 2) IT Department, (3) Accounting department, Mu'tah University, Mu'tah – Karak, Jordan

Ahmad Ali Alhasanat, College of Business Administration & Economics, Al-Hussein Bin Talal University, Maan, Jordan

Abstract — This paper presents a new solution for choosing the K parameter in the k-nearest neighbor (KNN) algorithm, the solution depending on the idea of ensemble learning, in which a weak KNN classifier is used each time with a different K, starting from one to the square root of the size of the training set. The results of the weak classifiers are combined using the weighted sum rule. The proposed solution was tested and compared to other solutions using a group of experiments in real life problems. The experimental results show that the proposed classifier outperforms the traditional KNN classifier that uses a different number of neighbors, is competitive with other classifiers, and is a promising classifier with strong potential for a wide range of applications.

Keywords- KNN; supervised learning; machine learning; ensemble learning; nearest neighbor;

6. Paper 31071422: Proposing a New Hybrid Approach in Movie Recommender System (pp. 40-45)

Full Text: PDF

Monireh Amini, Department of Computer Engineering, Zanjan Branch, Islamic Azad University, Zanjan, Iran

Mahdi Nasiri, Computer Engineering Department, Iran University of Science and Technology, Tehran, Iran

Mahdi Afzali, Department of Computer Engineering, Zanjan Branch, Islamic Azad University, Zanjan, Iran

Abstract — Due to the unprecedented growth of information, goods and services, a lot of application programs have been created in recent years to help the selection of goods and services to customers. One of the most important application programs are recommender systems that used to things as proposed movies, books, web pages, and E-Business, etc. Most of recommender systems using Collaborative filtering (CF) and or content based filtering (CBF) to provide suggestion for users. In this paper a new approach examined for better assessing the interests of customers. With the understanding of customer behavior, appropriate offer will be provided to customers. In fact, by using a new hybrid approach, weakness of content based filtering and Collaborative filtering methods as much as possible to will be resolve. The results of this paper can be used to keep and attract customers in the institutions or stores that have fixed customers. In this paper, first we reviewing the recommender systems and investigating types of filtering. Then a new hybrid approach by using CF and CBF methods is presented in a movie recommender system. Results are evaluated on movielens valid data, that the results show improvement in the movie recommender system.

Keywords - Hybrid Recommender Systems; Collaborative filtering; Content-based filtering; Hybrid filtering; Spiking Neural Network (SNN); Naive Bayes; E-Business

7. Paper 31071420: Trellis Analysis of Transmission Burst Errors in Viterbi Decoding (pp. 46-53)

Full Text: PDF

Salehe I. Mrutu (1), Anael Sam (2) and Nerey H. Mvungi (3)

(1, 2) School of Computational and Communication Science and Engineering, Nelson Mandela Institution of Science and Technology, Arusha, Tanzania

(3) College of Information and Communication Technologies, University of Dar Es Salaam, Dar Es Salaam, Tanzania

Abstract — The Viterbi decoder is the most favorable solution to the problem of decoding codewords from a convolutional encoder. Viterbi decoder performs exceptionally well when a received codewords block contains single or multiple and scattered errors in a received codewords block. However, the formation of burst errors in data transmission due to high transmission speed and the widely varying error conditions of wireless media in fading channel creates decoding challenge for such conditions which result in unbearable amount of residual errors. By using Viterbi decoders’ trellis diagrams, this paper analyses the effects of burst errors to the decoder that lead to residual errors and proposes improvement to the encoding and decoding procedures of the existing (2, 1, 2) binary convolutional encoder. The improved version facilitate effectiveness in the decoder (Viterbi algorithm) in decoding burst errors and hence reduction of residual errors in a poor channel. The proposed enhancements improve the decoder’s operational performance by 75 percent. However, the proposed modification reduces the encoder’s data transmission rate from 1/2 to 1/6.

Keywords - Locked Convolutional encoder; Bust errors; Residual errors; Non Transmittable Codewords (NTCs); Viterbi Algorithm Decoding

8. Paper 31071436: Towards a Mobile-Based DSS for Smallholder Livestock Keepers: Tanzania as a Case Study (pp. 54-63)

Full Text: PDF

Bernard Mussa* & Zaipuna Yonah,

Computational and Communication Science and Engineering, The Nelson Mandela African Institution of Science and Technology, Arusha, Tanzania.

Charles Tarimo, College of Engineering and Technology, University of Dar Es Salaam, Dar Es Salaam, Tanzania.

Abstract -- Building a useful and responsive Decision Support System (DSS) requires a deep understanding of the pertinent application domain before starting the system design. In this paper we report about an attempt to develop a mobile-based DSS for smallholder livestock keepers with Arusha region as a case study. The objective of the reported study is to provide an information tool for decision making to the smallholder livestock keepers. The development process involved: 1) employing information gathering techniques to understand smallholder livestock keepers’ information needs 2) studying the current methods that are used for information flow among livestock stakeholders. (i.e. smallholder livestock keepers, extension officers and livestock researchers) 3) analysis of the current situation within Arusha: located in the northern parts of Tanzania in terms of mobile phones penetration, with prospects of leveraging the high mobile phone penetration rate for enhanced information sharing among the smallholder livestock keepers and 4) exploration of options for the platform/model to be used for information access and delivery. The outputs of the above four activities were used to inform the requirements elicitation, and design phases of the mobile-based DSS system development. In addition, the mentioned four activities were supplemented by an extensive literature review of related works on requirements engineering in DSS development. It is anticipated that once the system has been developed, it will be of help to livestock keepers, improving farm-level productivity and decision making process. Findings from the study indicate that majority of smallholder livestock keepers in the selected area possess mobile phones and are in need of access to specific information to support their livestock related decision making. However, information access platforms/models that are currently in place do not cater for a satisfactory solution to their needs. Analysis of various options for designing a DSS platform has shown that a model that considers the administrative, organizational structure, as well as roles of relevant stakeholders in the livestock information flow will be useful for the studied context. The proposed Role-based Information Decision Support (RIDS) Model will facilitate data querying, analysis and information delivery based on users’ information requirements for the design of the DSS’s data marts. This will, in turn, be the basis for implementing a system of information sharing and delivery mechanism that will improve the decision making process and livestock management for smallholder livestock keepers in the studied geographical environment.

Keywords: Decision Support System, Data Mart, Mobile phones, Smallholder livestock keepers.

9. Paper 31071425: Computational Algorithms Based on the Paninian System to Process Euphonic Conjunctions for Word Searches (pp. 64-76)

Full Text: PDF

(1) Rajitha V., Department of Computer Science, Meenakshi College for Women, Chennai, India

Research Scholar, Mother Teresa Women’s University, Kodaikanal, India

(2) Kasmir Raja S. V., Dean – Research, SRM University, Chennai, India

(3) Meenakshi Lakshmanan, Department of Computer Science, Meenakshi College for Women, Chennai, India

Abstract – Searching for words in Sanskrit E-text is a problem that is accompanied by complexities introduced by features of Sanskrit such as euphonic conjunctions or ‘sandhis’. A word could occur in an E-text in a transformed form owing to the operation of rules of sandhi. Simple word search would not yield these transformed forms of the word. Further, there is no search engine in the literature that can comprehensively search for words in Sanskrit E-texts taking euphonic conjunctions into account. This work presents an optimal binary representational schema for letters of the Sanskrit alphabet along with algorithms to efficiently process the sandhi rules of Sanskrit grammar. The work further presents an algorithm that uses the sandhi processing algorithm to perform a comprehensive word search on E-text.

Keywords – Sanskrit; euphonic conjunction; sandhi; linguistics; Panini; Sanskrit word search; E-text search.

10. Paper 31071411: Classification of Sleep Stages Using Neural Network Based on EEG and EOG signals (pp. 77-79)

Full Text: PDF

Shreya Garg, Vijay Khare

Jaypee Institute of Information Technology, Noida

Abstract — This paper introduces an algorithm for different sleep stages classification. The algorithm consists of wavelet packet transformation (WPT) which is applied to 30 seconds long epochs of EEG and EOG recordings to provide time-frequency information, a feature generator to quantify the information and reduce the data set size, and then artificial neural networks for doing optimal classification. This led to a classification method with efficiency of 90.41 percent.

Keywords-component; Neural Network (NN), Electroencephalograph (EEG), Electrooculorgraph (EOG), Polysomnography (PSG), Wavelet Packet Transform(WPT)

11. Paper 30061432: An Ultra Low Power and High Throughput FPGA Implementation of SHA-1 Hash Algorithm (pp. 80-86)

Full Text: PDF

Shahzad Khan Department of Computer Science Shaheed Benazir Bhutto University(SBBU) Khyber PukhtunKhwa, Pakistan

Zain-ul-Abideen Department of Communications System Engineering School of Electrical Engineering and Computer Sciences(SEECS), NUST, Islamabad, Pakistan

Shahzad Shahid Paracha Department of Communication System Engineering School of Electrical Engineering and Computer Science(SEECS), NUST, Islamabad, Pakistan

Abstract - In this paper, we present a low power and highly parallel SHA-1 architecture which is considered as extremely iterative in nature specifically suitable for power sensitive applications. That is achieved by first identifying non dependent operations among the consecutive iterations of the algorithm and then aligning them for their execution in a highly parallel way. Consequently, when iteration completes, some other iterations also get completed and only a few of their dependent operations are left. By using this approach we were able to perform up to four SHA-1 iterations simultaneously resulting an increase in its throughput approximately by four times. We also explain how our results critically effect in lowering down the power consumption of the design.

Keywords - Cryptography, Hash function, FPGA implementation

12. Paper 30061425: Various Solutions of Black Hole Attack in A mobile Ad Hoc Network (MANET) (pp. 87-92)

Full Text: PDF

Imad I. Saada, Majdi Z. Rashad, Sherihan Abuelenin

Department of Computer Science, Faculty of Computers and Information, Mansoura University, Egypt

Abstract - Mobile ad hoc network (MANET) is a kind of wireless network that has a number of nodes, these nodes are distributed and connected without dependency on any infrastructure. MANET security has been an important issue since many years, many researchers have concerned in the black hole threat which "announce itself that it has a route to the destination in all cases". There are many solutions have been proposed to encounter these threats, the problem is that the security threats still exist because it is not prevented or avoided completely, in addition the performance of MANET is adversely affected by these solutions, the objective is to find out to what degree it is possible to prevent this attack by these solutions without causing negative effect on efficiency of MANET, so this survey may facilitate developing or proposing more compact idea to encounter security threats. This paper discusses many important solutions that work to detect a black hole node by using different strategies. In this paper, a new strategy proposed but still under testing.

Keywords; MANET, Black hole, AODV, LIDBPP, Network security.

13. Paper 30061417: Tools and Techniques for Ontology Interoperability: A Survey (pp. 93-98)

Full Text: PDF

R. Lakshmi Tulasi, Professor& HOD, Department of IT, QISCET, Ongole, India

Dr M. Srinivasa Rao, Professor, Dean CIHL, SIT, JNTUH, Hyderabad, India

Abstract — The idea of the semantic web is to add machine process able information to web-based data in order to realize interoperability. Ontology is a shared conceptualization of knowledge representation of particular domain. These are used for the enhancement of semantic information explicitly. Ontologies play a prominent role in the concept of the semantic web to provide semantic information for assisting communication among heterogeneous information repositories. Ontology Interoperability provides the reusability of ontologies Different domain experts and ontology engineers create different ontologies for the same or similar domain depending on their data modeling requirements. These cause ontology heterogeneity and inconsistency problems. As increasing numbers of ontologies are developed by diverse communities, the demand for rapid ontology mapping is arising. For more better and precise results ontology mapping is the solution. As their use has increased, providing means of resolving semantic differences has also become very important. Papers on ontology interoperability report the results on different frameworks and this makes their comparison almost impossible. Therefore, the main focus of this paper will be on providing some basics of ontology interoperability and briefly introducing its different approaches. In this paper we survey the approaches that have been proposed for providing interoperability among domain ontologies and its related techniques and tools.

Keywords- Ontology Interoperability; Ontology Mapping; Ontology Alignment; Ontology Merging; Semantic heterogeneity; Semantic web;

14. Paper 31071491: Generic Lightweight Certificate Management Protocol (GLCMP) (pp. 99-105)

Full Text: PDF

Shahzad Khan Department of Computer Science Shaheed Benazir Bhutto University(SBBU) Khyber PukhtunKhwa, Pakistan

Muhammad Asif Department of Communications System Engineering School of Electrical Engineering and Computer Sciences(SEECS), NUST, Islamabad, Pakistan

Abstract - This paper describes a Generic Light Weight Certificate Management Protocol (GLCMP) for handling certificates on mobile devices. Theoretically, various security solutions are designed to protect the valuable information of mobile users. But, its power, memory and processing constraints, high response time and authentication latencies are the main challenges for the researcher to develop and integrate standard security mechanisms in it. It is observed that, most of mobile users are not technical enough to configure security parameters and even already developed libraries do not support extended security features like transparent handling of certificates, verification of identities, and distribution of certificates. In this paper, an innovative and comparatively efficient protocol is designed and implemented. It does not only overcome the shortcoming of the certificate handling in mobile devices but also provides some extended certificate related features like registration, authentication and trust delegation. The designed GLCMP is lightweight because all complex and computation-intensive operations, involved in creation of certificate request in PKCS#10 standard format, are offloaded to a proxy server. It also provides domain based secure registration and verification of the identities without exchanging any confidential information to the proxy servers and even no user’s credential is exchanged on network for authentication. After analyzing its performance, we noticed that authentication latency of GLCMP is 0.394 sec which is less than previously proposed protocols like NSI (4.7), PKI (5.01), and PKASSO (5.19 delegation time + 0.082 authentication times). We also formally verified our designed by using Z-Notation Modeling techniques and found that it is protected against man-in-the-middle, replay and impersonation and non-repudiation attacks.

15. Paper 31071421: Efficient RSA Variant for Resource Constrained Environment (pp. 106-112)

Full Text: PDF

Seema Verma, Computer Science Department, Thapar University, Patiala, India

Dr Deepak Garg, Computer Science Department, Thapar University, Patiala, India

Abstract—The work in this paper is concerned with the memory consumption as well as the performance of RSA cryptosystem so that the most popular public key algorithm can be used efficiently in the resource constrained environment also. For this purpose, RSA variant, RC RSA, is proposed which results in low computational cost and low memory consumption. RC RSA is the improvement over dual RSA small e (based on less memory consumption). Mathematically, as compared to Dual RSA, RC RSA results in the increase of decryption speed by a factor of 9 and in implementation roughly by a factor of 6. On the other hand the encryption speed becomes as low as in standard RSA. Besides the computational speed up, RC RSA is proved to be more secure than the Dual RSA scheme.

Keywords- cryptography; encryption; public key; security

16. Paper 31071417: Real Time Recommender System for Music Data (pp. 113-117)

Full Text: PDF

Mrs. Manjula Athani, Prof. Neelam Pathak, Prof. Asif Ullah Khan, Dr. Bhupesh Gour

CSE, T.I.T, RGPV Bhopal , India

Abstract — Recommender system is able to identifying the n-number of users preferences and adaptively recommend music tracks according to user preferences. we are extracting unique feature tempo of each music using Marsyas Tool. Then we are applying BLX- α crossover to a extracted feature of each music track. User favorite and user profiles are included. This system have been emerging as a powerful technique of ecommerce. The majority of existing recommender systems uses an overall rating value on items for evaluating user’s preference opinions. Because users might express their opinions based on some specific features of the item, recommender systems could produce recommendations that meet user needs. In this paper we presented a Real time recommender system for music data. Multiuser Real time recommender system combines the two methodologies, the content based filtering technique and the interactive genetic algorithm by providing optimized solution every time and which is based on user’s preferences We can also share the favorite songs to other user hence it give better result and better user system.

Keywords - Recommender system, Interactive Genetic algorithm, Content Based filtering BLX- α

17. Paper 31071405: ICT as a Tool For Improving Information Flow Among Livestock Stakeholders. A Case Study of Tanzania (pp. 118-128)

Full Text: PDF

Gladness Mwanga George, Nelson Mandela African Institution of Science and Technology, Arusha Tanzania

Fatma Simba, University of Dar es Salaam,

Zaipuna O.Yonah, Nelson Mandela African Institution of Science and Technology. Arusha Tanzania

Abstract - Traditionally, extension services are used as a means of conveying to rural areas knowledge derived from different studies and surveys. These extension services are an important way to help livestock farmers to improve their lives and production methods, which in turn leads to an increased and improved livestock and livestock products. Nowadays, extension services deliver far more beyond the traditional role. They involve helping farmers to form collaborative groups, solving marketing challenges, etc. This fact has been confirmed by the study reported in this paper. The main objective of the study was to evaluate the current recording system and information flow among livestock stakeholders, and how ICT it has been used as a tool to bridge the gap of information deficit. This paper presents an analysis of data collected from 15 wards from Arumeru district, Arusha region – Northern Tanzania including data from; district council, livestock farmers, extension officers and researchers from three institutions, Livestock Training Agency (LITA), National Artificial Insemination Centre (NAIC) and Nelson Mandela African Institution of science and Technology (NM-AIST). The results reveal that the current recording system is poor and there is a gap in the flow of information among the stakeholders of livestock farming. This gap has significantly contributed to the deterioration of access to extension services to livestock farmers. Along with other factors, this gap is also attributed to the researchers, who publish their research findings through various online journals. The gap signifies that it has been hard for extension officers to fulfil their roles due to a large number of farmers they serve and the challenge of traveling long distances to deliver such services to farmers. Based on the results of this analysis, the paper concludes with a set of identified approaches on how ICT can play a part in minimizing the gap that has been found for increasing the efficiency of the extension services in reaching livestock farmers.

Key words: ICT, information, Extension services, livestock stakeholders, record keeping.