World of Computer Science and Information Technology Journal Authors Morteza Zahedi Ali Reza Manashty Authors Ghossoon. M. W. Al- Saadoon Hilal M.Y. Al-Bayatti Authors Elarbi Badidi Larbi Esmahi Authors Awodele O. Kuyoro S. O. Adejumobi A. K. Awe O., Makanju O Authors Ioannis Koskosas Maria-Mirela - Koskosa Authors Alaa H. AL-Hamami Mohammad A. AL-Hamami Soukaena H. Hashem Authors Morteza Zahedi Aboulfazl Sarkardei Authors Ahmed Khalid Izzeldin M. Osman Authors Vivian Ogochukwu - Nwaocha Inyiama H.C. Authors A. J. M. Abu Afza Dewan Md. Farid Chowdhury Mofizur - Rahman Authors Adewole Adetunji - Philip Akinwale Adio Taofiki Akintomide Ayo Bidemi Robust Sign Language Recognition System Using ToF Depth Cameras Abstract: Sign language recognition has been a difficult task, yet required  for many applications in real-time speed. Using RGB cameras for  recognition of sign languages is not very successful in practical situations  and accurate 3D imaging requires expensive and complex instruments. With introduction of Time-of-Flight (ToF) depth cameras in recent years, it has  become easier to scan the environment for accurate, yet fast depth images  of the objects without the need of any extra calibrating object. In this  paper, a robust system for sign language recognition using ToF depth  cameras is presented for converting the recorded signs to a standard and  portable XML sign language named SiGML for easy transfer and converting  to real-time 3D animated virtual characters. Feature extraction using  moments and classification using nearest neighbor classifier are used to  track hand gestures and significant result of 100% is achieved for the  proposed approach.  Keywords : sign language; Time-of-Flight camera; sign recognition; SigML; Moments; hand tracking; range cameras. A Comparison of Trojan Virus Behavior in Linux and Windows Operating Systems Abstract: Trojan virus attacks pose one of the most serious threats to  computer security. A Trojan horse is typically separated into two parts - a  server and a client. It is the client that is cleverly disguised as significant  software and positioned in peer-to-peer file sharing networks, or  unauthorized download websites. The most common means of infection is  through email attachments. The developer of the virus usually uses various  spamming techniques in order to distribute the virus to unsuspecting users.  Malware developers use chat software as another method to spread their  Trojan horse viruses such as Yahoo Messenger and Skype. The objective of this paper is to explore the network packet information  and detect the behavior of Trojan attacks to monitoring operating systems  such as Windows and Linux. This is accomplished by detecting and  analyzing the Trojan infected packet from a network segment -which passes through email attachment- before attacking a host computer. The results that have been obtained to detect information and to store  infected packets through monitoring when using the web browser also  compare the behaviors of Linux and Windows using the payload size after  implementing the Wireshark sniffer packet results. Conclusions of the  figures analysis from the packet captured data to analyze the control bits  and , check the behavior of the control bits, and the usability of the  operating systems Linux and Windows. Keywords : Trojan horse behavior; Internet Security; Segment of Network; Pcap- Packet CAPture; Payload. A Cloud-based Approach for Context Information Provisioning  Abstract: As a result of the phenomenal proliferation of modern mobile  Internet-enabled devices and the widespread utilization of wireless and  cellular data networks, mobile users are increasingly requiring services  tailored to their current context. High-level context information is typically  obtained from context services that aggregate raw context information  sensed by various sensors and mobile devices. Given the massive amount of sensed data, traditional context services are lacking the necessary  resources to store and process these data, as well as to disseminate high-  level context information to a variety of potential context consumers.   In this paper, we propose a novel framework for context information  provisioning, which relies on deploying context services on the cloud and  using context brokers to mediate between context consumers and context  services using a publish/subscribe model. Moreover, we describe a multi-  attributes decision algorithm for the selection of potential context services  that can fulfill context consumers' requests for context information. The  algorithm calculates the score of each context service, per context  information type, based on the quality-of-service (QoS) and quality-of-  context information (QoC) requirements expressed by the context  consumer. One of the benefits of the approach is that context providers can scale up  and down, in terms of cloud resources they use, depending on current  demand for context information. Besides, the selection algorithm allows  ranking context services by matching their QoS and QoC offers against the  QoS and QoC requirements of the context consumer. Keywords : mobile users; context-aware web services; context services; cloud services; quality-of-context; quality-of-service; service selection. Citadel E-Learning: A New Dimension to Learning System Abstract: E-learning has been an important policy for education planners  for many years in developed countries. This policy has been adopted by  education in some developing countries; it is therefore expedient to study  its emergence in the Nigerian education system. The birth of contemporary  technology shows that there is higher requirement for education even in  the work force. This has been an eye opener to importance of Education  which conveniently can be achieved through E-learning. This work presents  CITADEL E-learning approach to Nigeria institutions; its ubiquity, its  implementations, its flexibility, portability, ease of use and feature that are synonymous to the standard of education in Nigeria and how it can be  enhanced to improve learning for both educators and learners to help them  in their learning endeavour. Keywords : E-learning environment; ICT; Distance learning. Internet Banking Security Management through Trust Management Abstract: The aim of this research is to investigate information systems  security in the context of security risk management. In doing so, it adopts a social and organizational approach by investigating the role and  determinants of trust in the process of security goal setting with regard to  internet banking risks. The research seeks to demonstrate the important  role of trust in the risk management context from a goal setting point of  view through a case study approach within three financial institutions in  Greece. The determinants of trust are also explored and discussed as well  as the different goal setting procedures within different information system  groups. Ultimately, this research provides a discussion of an interpretive  research approach with the study of trust and goal setting in the risk  management context and its grounding within an interpretive epistemology. Keywords : trust; goal setting; security management; internet banking;  interpretive epistemology. A proposed Modified Data Encryption Standard algorithm by Using Fusing Data Technique Abstract: Data Encryption Standard (DES) is a block cipher that encrypts  data in 64-bit blocks. A 64-bit block of plaintext goes in one end of the  algorithm and a 64-bit block of cipher text comes out of the other end.  Blowfish is a block cipher that encrypts data in 8-byte blocks .Blowfish  consists of two parts: a key-expansion part and a data-encryption part. Key  expansion converts a variable-length key of at most 56 bytes (448 bits) into  several subkey arrays totaling 4168 bytes. Blowfish has 16 rounds, such as  DES.  In this research the fusion philosophy will be used to fuse DES's with  blowfish and Genetic Algorithms by taking the strong points in all of these  techniques to create a proposed Fused DES-Blowfish algorithm. The  proposed algorithm is presented as a modified DES depending on the  advantage in key generation complexity in blowfish and advantage of  optimization in Genetic Algorithm to give the optimal solution.  The  solution will be the depended tool for creation of the strong keys.  Keywords : Fusing; Blowfish; Genetic Algorithm; Strong keys; and Data  Encryption Standard.   Using MI Method for Feature Weighting to Improve Text Classification Performance Abstract: In text classification, feature weighting is a main step of  preprocessing. Commonly used feature weighting methods only consider  the distribution of a feature in the documents and do not consider the class  information for feature weighting. Mutual Information (MI) method which  represents the dependency of a feature in the regarding class, has been  previously used for feature selection. The aim of this paper is to show that  the use of MI method for feature weighting increases the performance of  text classification, in terms of average recall and average precision. While  K-nearest neighbor classifier is employed for classification, the average  recall is increased about 18% and average precision is increased about 10%.  It is shown that the results for average precision and average recall become  91.7% and 89.29% respectively. Keywords : text classification; mutual information; MI; feature weighting;  Hamshahri; K-nearest neighbor. A  Multi-Phase Feature Selection Approach for the Detection of SPAM Abstract:  In the past few years the Naïve Bayesian (NB) classifier has been  trained automatically to detect spam (unsolicited bulk e-mail). The paper  introduces a simple feature selection algorithm to construct a feature  vector on which the classifier will be built. We conduct an experiment on  SpamAssassin public email corpus to measure the performance of the NB  classifier built on the feature vector constructed by the introduced  algorithm against the feature vector constructed by the Mutual Information  algorithm which is widely used in the literature. The effect of the stop-list  and the phrases-list on the classifier performance was also investigated.  The results of the experiment show that the introduced algorithm  outperforms the Mutual Information algorithm. Keywords : component; detection; feature selection; Naïve Bayesian  classifiers. Precluding Emerging Threats from Cyberspace: An Autonomic Administrative Approach Abstract: Information Technology and Network Security Managers face  several challenges in securing their organization's network due to the  increased sophistication of attacks. Besides, the number of attacks and  vulnerabilities are rising due to the inability of the existing intrusion  detection and prevention system to detect and prevent novel attacks.  Hence, intrusion detection systems which were previously adequate to  wedge the evolving attacks in cyberspace have become ineffective in  impeding these attacks. Consequently, intrusion detection and prevention  systems are required to actually prevent attacks before they cause harm. A  major consideration of this work is to present an architecture that provides  protection through the self-healing and self-protecting properties of the  autonomic computing. The proposed system which operates by means of  autonomous agents is based on risk assessment. The application of risk  analysis and assessment reduces the number of false-positive alarms.  Furthermore, the system autonomous features enables it to automatically  diagnose, detect and respond to disruptions, actively adapt to changing  environments, monitor and tune resources, as well as anticipate and  provide protection against imminent threats. Keywords : Agent; Autonomic Computing; Computer system; Intrusion;  Intrusion detection and prevention; Network; Threats.  A Hybrid Classifier using Boosting, Clustering, and Naïve Bayesian Classifier Abstract: a new classifier based on boosting, clustering, and naïve Bayesian  classifier is introduced in this paper, which considers the misclassification  error produced by each training example and update the weights of training  examples in training dataset associated to the probability of each attribute  of that example. The proposed classifier clusters the training examples  based on the similarity of attribute values and then generates the  probability set for each cluster using naïve Bayesian classifier. Boosting  trains a series of classifiers for a number of rounds that emphasis to the  misclassification rate in each round. The proposed classifier addresses the  problem of classifying the large data set and it has been successfully tested  on a number of benchmark problems from the UCI repository, which  achieved high classification rate.    Keywords :  clustering; naïve Bayesian classifier; boosting; hybrid classifier. Artificial Neural Network Model for Forecasting Foreign Exchange Rate Abstract: The present statistical models used for forecasting cannot  effectively handle uncertainty and instability nature of foreign exchange  data. In this work, an artificial neural network foreign exchange rate  forecasting model (AFERFM) was designed for foreign exchange rate  forecasting to correct some of these problems. The design was divided into  two phases, namely: training and forecasting. In the training phase, back  propagation algorithm was used to train the foreign exchange rates and  learn how to approximate input. Sigmoid Activation Function (SAF) was  used to transform the input into a standard range [0, 1]. The learning  weights were randomly assigned in the range [-0.1, 0.1] to obtain the  output consistent with the training. SAF was depicted using a hyperbolic  tangent in order to increase the learning rate and make learning efficient.  Feed forward Network was used to improve the efficiency of the back  propagation. Multilayer Perceptron Network was designed for forecasting.  The datasets from oanda website were used as input in the back  propagation for the evaluation and forecasting of foreign exchange rates.  The design was implemented using matlab7.6 and visual studio because of  their supports for implementing forecasting system. The system was tested  using mean square error and standard deviation with learning rate of 0.10,  an input layer, 3 hidden layers and an output layer. The best known related  work, Hidden Markov foreign exchange rate forecasting model (HFERFM)  showed an accuracy of 69.9% as against 81.2% accuracy of AFERFM. This  shows that the new approach provided an improved technique for carrying  out foreign exchange rate forecasting. Keywords :  Artificial Neural Network; Back propagation Algorithm; Hidden  Markov Model; Baum- Weld Algorithm; Sigmoid Activation Function and  Foreign Exchange Rate.