Special Issue 2016, No. 1A
Print E-mail

Facial Recognition under Expression Variations

Mutasem Alsmadi

Department of Management of Information System, University of Dammam, Kingdom of Saudi Arabia

Abstract: Researchers in different fields such as image processing, neural sciences, computer programs and psychophysics have investigated number of problems related to facial recognition by machines and humans since 1975. Automatic recognition of the human emotions using facial expression is an important, but difficult problem. This study introduces a novel and automatic approach to analyze and recognize human facial expressions and emotions using a Metaheuristic Algorithm (MA), which hybridizes iterated local search and Genetic Algorithms with Back-Propagation algorithm (ILSGA-BP). Back-propagation algorithm (BP) was used to train and test the extracted features from the extracted right eye, left eye and mouth using radial curves and Cubic Bézier curves, MA was used to enhance and optimize the initial weights of the traditional BP. FEEDTUM facial expression database was used in this study for training and testing processes with seven different emotions namely; surprise, happiness, disgust, neutral, fear, sadness and anger. A comparison of the results obtained using the extracted features from the radial curves, Cubic Bézier curves and the combination of them experiments were conducted. The comparison shows the superiority of the combination of the radial curves and the Cubic Bézier curves with percentage ranges between 87% and 97% over the radial curves alone with a percentage ranges between 80% and 97% and over the Cubic Bézier curves with a percentage ranges between 83% and 97%. Moreover, based on the extracted features using the radial curves, Cubic Bézier curves and the combination of them, the experimental results show that the proposed ILSGA-BP algorithm outperformed the BP algorithm with overall accuracy 88%, 89% and 93.4% respectively, compared to 83%, 82% and 85% respectively using BP algorithm.

 Keywords: Face recognition, cubic bézier curves, radial curves, features extraction, MA, BP.

 

Received July 9, 2015; accepted October 18, 2015

Full Text

 


 
Print E-mail

Performance Comparison of Neuro-Fuzzy Cloud Intrusion Detection Systems

Sivakami Raja1 and Saravanan Ramaiah2

1Department of Information Technology, PSNA College of Engineering and Technology, India

2Department of Computer Science and Engineering, RVS Educational Trust's Group of Institutions, India

Abstract: Cloud computing is a subscription-based service where we can obtain networked storage space and computer resources. Since, access to cloud is through internet, data stored in clouds are vulnerable to attacks from external as well as internal intruders. In order to, preserve privacy of the data in cloud, several intrusion detection techniques, authentication methods and access control policies are being used. The common intrusion detection systems are predominantly incompetent to be deployed in cloud environments due to their openness and specific essence. In this paper, we compare soft computing approaches based on type-1, type-2 and interval type-2 fuzzy-neural systems to detect intrusions in a cloud environment. Using a standard benchmark data from a Cloud Intrusion Detection Dataset (CIDD) derived from DARPA Intrusion Detection Evaluation Group of MIT Lincoln Laboratory, experiments are conducted and the results are presented in terms of mean square error.

Keywords: Fuzzy neural networks, hybrid intelligent systems, intrusion detection, partitioning algorithms, pattern analysis.

Received September 13, 2015; accepted October 18, 2015

 

 
Print E-mail

Interactive Visual Search System Based on Machine Learning Algorithm

Anas Al-Fayoumi and Mohammad Hassan

Department of Computer Science, Zarqa University, Jordan

Abstract: This paper presents a tool that enables non-technical end-users to use free-form queries in exploring a large scale datasets with simple and interactive direct technique. The proposed approach is based on effective integration of different techniques, such as data mining, visualization and Human-Computer Interaction (HCI). The proposed model has been incorporated into a prototype developed as a web-based application using different programming languages and software tools. The system has been implemented based on a real dataset, whereas the obtained results indicate the efficiency of such approach.

Keywords: Visualization, feature extraction, data mining, visual data mining, machine learning.

 

Received September 17, 2015; accepted October 18, 2015

 

 
Print E-mail

Constraints Aware and User Friendly Exam Scheduling System

Mohammad Al-Haj Hassan1 and Osama Al-Haj Hassan2

1Computer Science Department, Zarqa University, Jordan

2Computer Science Department, Isra University, Jordan

 Abstract: Scheduling is a crucial task for schools, universities, an d industries. It is a vital task for any system containing utilization of resources to fulfill a certain criterion. Utilization of such resources usually includes several conflicting constraints that scheduling has to take into account. Exam scheduling is an essential key for schools and universities in order for exams periods to be smooth. In this paper, we present an exam scheduling system that employs graph coloring scheduling technique. We focus on two aspects: First, the constraints our system handles, second, the user friendly interface of the system.

 Keywords: Exam scheduling, user friendly, constraints, optimization, conflict, graph coloring.

 Received September 16, 2015; accepted October 18, 2015

 

 
Print E-mail

A Prototype for a Standard Arabic Sentiment Analysis Corpus

Mohammed Al-Kabi1, Mahmoud Al-Ayyoub2, Izzat Alsmadi3, and Heider Wahsheh4

1Computer Science Department, Zarqa University, Jordan

2Computer Science Department, Jordan University of Science and Technology, Jordan

3Computer Science Department, University of New Haven, USA

4Computer Science Department, King Khaled University, Saudi Arabia

Abstract: The researchers in the field of Arabic Sentiment Analysis (SA) need a relatively big standard corpus to conduct their studies. There are a number of existing datasets; however, they suffer from certain limitations such as the small number of reviews or topics they contain, the restriction to Modern Standard Arabic (MSA), etc., Moreover, most of them are in-house datasets that are not publicly available. Therefore, this study aims to establish a flexible and relatively big standard Arabic SA corpus that can be considered as a foundation to build larger Arabic corpora. In addition to MSA, this corpus contains reviews written in the five main Arabic dialects (Egyptian, Levantine, Arabian Peninsula, Mesopotamian, and Maghrebi group). Furthermore, this corpus has other five types of reviews (English, mixed MSA English, French, mixed MSA and Emoticons, and mixed Egyptian and Emoticons). This corpus is released for free to be used by researchers in this field, where it is characterized by its flexibility in allowing the users to add, remove, and revise its contents. The total number of topics and reviews of this initial version are 250 and 1,442, respectively. The collected topics are distributed equally among five domains (classes): Economy, Food-Life style, Religion, Sport, and Technology, where each domain has 50 topics. This corpus is built manually to ensure the highest quality to the researchers in this field.

Keywords: SA, opinion mining, making of Arabic corpus, arabic reference corpus, maktoob yahoo!.

 

Received September 17, 2015; accepted October 18, 2015

 

 
Print E-mail

Fingerprint Image Quality Fuzzy System

Abdelwahed Motwakel1 and Adnan Shaout2

1Collage of Post Graduate Studies, Sudan University of Science and Technology

2The Electrical and Computer Engineering Department, the University of Michigan-Dearborn

Abstract: In this paper, we present a novel technique to analysis fingerprint image quality using fuzzy logic. The quality of fingerprint image greatly affects the performance of minutiae extraction and the process of matching in fingerprint identification system. The system uses the extracted four features from a fingerprint image which are the Local Clarity Score (LCS), Global Clarity Score (GCS), Ridge_Valley Thickness Ratio (RVTR), and the contrast. The proposed fuzzy logic system uses mamdani fuzzy rule model which can analysis and determinate the fingerprint image type (oily, dry or neutral) based on the extracted feature values and fuzzy inference rules.

Keywords: Fingerprint image quality, LCS, GCS, RVTR, contrast, fuzzy inference system.

 

Received September 13, 2015; accepted October 18, 2015

 

 
Print E-mail

Algorithm for Answer Extraction based on Pattern Learning

Muthukrishnan Ramprasath1 and Shanmugasundaram Hariharan2

1Faculty in Computer Science and Engineering, Sathyabama University, India

2Department of Computer Science and Engineering, TRP Engineering College, India

Abstract: The rapid growth of information available on the internet has provoked the development of diverse tool for searching and browsing large document collections. Information Retrieval (IR) system act as a vital tool for identifying relevant document for user queries posted to search engine. Some special kind of IR system, such as: Google, yahoo and Bing which allow the system to retrieve the relevant information to user question form web. Question Answering System (QAS) play important role for identifying the correct answer to user question by relying on the many IR tools. In this paper, we propose a method for answer extraction based on pattern learning algorithm. Answer extraction component provide precise answer to user question. The proposed QA system uses the pattern learning algorithm which consists of following component such as question transformation, question and answer pattern generation, pattern learning, pattern based answer extraction and answer evaluation. The experiment has been conducted different type question on Textual Case-Based Reasoning (TREC) data sets. Our system used different ranking metrics in the experimental part to find the correct answer to user question. The experimental results were investigated and compare with different type of questions.

 Keywords: QAS, pattern learning, question transformation, answer extraction, TREC data set.

Received August 29, 2015; accepted October 18, 2015

 

 

 
Print E-mail

The Effect of Horizontal Database Table Partitioning on Query Performance

Salam Matalqa and Suleiman Mustafa

Department of Computer Information Systems, Yarmouk University, Jordan

Abstract: The need for achieving optimal performance for database applications is a primary objective for database designers and a primary requirement for database end users. Partitioning is one of the techniques used by designers to improve the performance of database access. The purpose of this study was to investigate the effect of horizontal table partitioning on query Response Time (RT) using three partitioning strategies: Zero partitioning, list partitioning and range partitioning. Three tables extracted from the Student Information System (SIS) at Yarmouk University in Jordan were used in this research. Variation in table size was used to determine when partitioning can have an impact (if any) on access performance. A set of 12 queries were run over a database of three different sizes. The results indicated that partitioning provided better RT than zero partitioning, on the other hand, range and list partitioning strategies showed little performance differences with the different database sizes.

Keywords: Table partitioning, horizontal partitioning, range partitioning, list partitioning, database performance.

 Received July 22, 2015; accepted October 18, 2015

 

 

 
Print E-mail

Integrated Replication-Checkpoint Fault Tolerance Approach of mobile agents “IRCFT”

 

Suzanne Sweiti1 and Amal Al Dweik2

1Deanship of Graduate Studies and Scientific Research, Palestine Polytechnic University, Palestine

2College of Information Technology and Computer Engineering,

 Palestine Polytechnic University, Palestine

                                                                       

Abstract: Mobile agents offer flexibility which is evident in distributed computing environments. However, agent systems are subject to failures that result from bad communication, breakdown of agent server, security attacks, lack of system resources, congestion in network, and situations of deadlock. If any of such things happen, mobile agents suffer loss or damage totally or partially while execution is being carried out. Reliability must be addressed by the mobile agent technology paradigm. This paper introduces a novel fault tolerance approach “IRCFT” to detect agent failures as well as to recover services in mobile agent systems. Our approach makes use of checkpointing and replication where different agents cooperate to detect agent failures. We described the design of our approach, and different failure scenarios and their corresponding recovery procedures are discussed. The proposed system is implemented over Agelt platform. The system improves the performance significantly.

 Keywords: Mobile agents, fault tolerance, reliability, checkpointing, replication.

Received September 15, 2015; accepted October 18, 2015

 

 

 

 
Print E-mail

Principal Component Regression with Artificial Neural Network to Improve Prediction of Electricity Demand

Noor Ismail and Syamnd Abdullah

Department of applied statistic, University of Malaya, Malaysia

 Abstract: Planning for electricity demand is a key factor for the success in the development of any countries. Such success can only be achieved if the demand for electricity is predicted correctly and accurately. This study introduces a new hybrid approach that combines Principle Component Regression (PCR) and Back-Propagation Neural Networks (BPNN) techniques in order to improve the accuracy of the electricity demand prediction rates. The study includes 13 factors that related to electricity demand, and data for these factors have been collected in Malaysia. The new combination (PCR-BPNN) starts to solve the problem of collinearity among the input dataset, and hence, the reliability of the results. The work focuses also on the errors that recoded at that output stage of the electricity prediction models due to changes in the patterns of the input dataset. The accuracy and reliability of the results have been improved through the new proposed model. Validations have been achieved for the proposed model through comparing the value of three performance indicators of the PCR-BPNN with the performance rates of three major prediction models. Results show the outperformance of the PCR-BPNN over the other types of the electricity prediction models.

 

Keywords: Electricity demand, accuracy and reliability, PCR, MLR, BPNN.

Received September 15, 2015; accepted October 18, 2015

 
Print E-mail

Word Sense Disambiguation for Arabic Text Categorization

 

Meryeme Hadni1, Said El Alaoui1, Abdelmonaime Lachkar2

1Department of Computer Science, Sidi Mohamed Ben Abdellah University, Morocco

2Department of Electrical and Computer Engineering, Sidi Mohamed Ben Abdellah University, Morocco

 

Abstract:  In this paper, we present two contributions for Arabic Word Sense Disambiguation. In the first one, we propose to use both two external resources Arabic WordNet (AWN) and WN based on term to term Machine Translation System (MTS). The second contribution consists of choosing the nearest concept for the ambiguous terms, based on more relationships with different concepts in the same local context. To evaluate the accuracy of our proposed method, several experiments have been conducted using Feature Selection methods; Chi-Square and CHIR, two machine learning techniques; the Naïve Bayesian (NB) and Support Vector Machine (SVM).The obtained results illustrate that using the proposed method increases greatly the performance of our Arabic Text Categorization System.

Keywords: WSD, arabic text categorization system, AWN, MTS.

Received September 1, 2015; accepted October 18, 2015

 

 

 
Print E-mail

Content-Based Image Retrieval Based on Integrating Region Segmentation and Colour Histogram

 

Duraisamy Yuvaraj1 and Shanmugasundaram Hariharan2

1Department of Computer Science and Engineering, M.I.E.T Engg College, India

2Department of Computer Science and Engineering, T.R.P Engg College, India

Abstract: Developments in multimedia technology, increasing number of image retrieval functions and capabilities has led to the rapid growth of CBIR techniques. Colour histogram could be compared in terms of speed and efficiency. We have presented a modified approach based on a composite colour image histogram. A major research perspective in CBIR emphasize on  matching similar objects based on shape, colour and texture using computer vision techniques in extracting image features. The colour histogram is perhaps the most popular one due to its simplicity. Image retrieval using colour histogram perhaps has both advantages and limitations. This paper presents some recommendations for improvements to CBIR system using unlabelled images. The experimental results presented using Matlab software significantly shows that region based histogram and colour histogram were effective as far as performance is concerned.

Keywords: Image analysis, CBIR, retinal imaging, gray scale and semantic description.

 

Received August 22, 2015; accepted October 18, 2015

Full Text

 

 
Print E-mail

Optimization of Position Finding Step of PCM-oMaRS Algorithm with Statistical Information

Ammar Balouch

Department of Computer Science, University of Rostock, Germany

Abstract: The PCM-oMaRS algorithm guarantees the maximal reduction steps of the computation of the exact median in distributed datasets and proved that we can compute the exact median effectively with reduction of blocking time and without needing the usage of recursive or iterative methods anymore. This algorithm provided more efficient execution not only in distributed datasets even in local datasets with enormous data. We cannot reduce the steps of PCM-oMaRS algorithm any more but we have found an idea to optimize one step of it. The most important step of this algorithm is the step in which the position of exact median will be determinate. For this step, we have development a strategy to achieve more efficiency in determination of position of exact median. Our aim in this paper to maximize the best cases of our algorithm and this was achieved through dividing the calculation of number of all value that smaller than or equal to temporary median in two groups: The first one contains only the values that smaller than the temporary median and the second group contains the values that equal to the temporary median. In this dividing we achieve other best cases of PCM-oMaRS algorithm and reducing the number of values that are required to compute the exact median. The complexity cost of this algorithm will be discussed more in this article. In addition some statistical information depending on our implementation tests of this algorithm will be given in this paper.

 Keywords: Median, parallel computation, algorithm, optimization, big data, evaluation, analysis, complexity costs

 Received June 11, 2015; accepted October 18, 2015

 

 
<
 
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/ccis2k.org/iajit/templates/rt_chromatophore/index.php on line 251 Warning: fsockopen(): unable to connect to oucha.net:80 (php_network_getaddresses: getaddrinfo failed: Name or service not known) in /hsphere/local/home/ccis2k/ccis2k.org/iajit/templates/rt_chromatophore/index.php on line 251 skterr