ABOUT US
JOURNALS
PUBLICATION ETHICS
PEER REVIEW
 
       
Asian Network for Scientific Information is a leading service provider to the publishers of Science, Technology and Medicine (STM) in Asia. Currently Asian Network for Scientific Information is serving more than 37 peer-reviewed journals covering a wide range of academic disciplines to foster communication among scientists, researchers, students and professionals - enabling them to work more efficiently and intelligently, thereby advancing knowledge and learning.

Journal of Artificial Intelligence
eISSN: 2077-2173
pISSN: 1994-5450

Editor-in-Chief:  Santosh Kumar Nanda
Current Issue
Archive
Guide for Authors
Editorial Board
Article Processing Charges
 
Research Article
Towards Enhancing Non-Cooperative Iris Recognition using Improved Segmentation Methodology for Noisy Images
A. Alice Nithya and C. Lakshmi
Background and Objective: Iris recognition is one of the popular winning biometric frameworks, giving promising outcomes in the identity authentication and access control systems. In this study, an efficient, fast and robust segmentation methodology suitable for non-cooperative and noisy iris images is proposed. Materials and Methods: This proposed methodology considers both shape and spatial feature properties of iris images taken from both the visible spectrum and near infrared spectrum. Circular hough transform is applied to the input image and iris outer boundary is identified. A minimum rectangular bounding box, MRB is defined using the obtained radius and center coordinates. High intensity valued, specular reflections and low intensity valued, pupil region, eyelids and eyelashes are identified using iterative thresholding and removed to reduce processing time. Scale invariant feature transform (SIFT) is directly applied on the segmented iris ROI, without performing normalization stage and system accuracy is tested. Results: By narrowing down the searching space to 65 times, this methodology provides robustness to noise as well as ensures faster segmentation of 0.34, 0.35 and 0.29 sec for CASIA V1.0, V3.0-interval and UBIRIS V1.0 datasets, respectively. Conclusions: The results obtained using improved segmentation methodology performs with improved recognition accuracy and reduced computational time and mislocalization count.
Research Article
Combining the Previous Measure of Evidence to Educational Entrance Examination
Andino Maseleno, Miftachul Huda, Maragustam Siregar, Roslee Ahmad, Aminudin Hehsan, Zulkiflee Haron, Mohd Nasir Ripin, Siti Suhaila Ihwani and Kamarul Azmi Jasmi
Background and Objective: Educational entrance examination refers to the extent in selecting the student to enroll through admission into educational institution. It has an entire procedure administered to achieve from primary to higher education. However, not many researches were conducted using mathematical theory of evidence. This study aims to investigate the examination process about the admission into educational institutions using mathematical theory of evidence. Materials and Methods: The assessment on student’s entrance examination through the effectiveness of Dempster-Shafer theory can be viewed with its significant contribution by combining the previous measure of evidence. Eight student’s entrance examination results were proposed. Results: The result reveals that there were some significant findings in assessing the student’s entrance examination using mathematical theory of evidence. Those were obtained degrees of belief of Computer Science with 76.4% for student 1, Computer Science with 64.2% for student 2, Computer Science with 75.4% for student 3, Computer Science with 80.3% for student 4, Computer Science with 67.4% for student 5, Computer Science with 57.1% for student 6, Islamic Studies with 26.3% for student 7, Computer Science with 62.5% for student 8. Conclusion: In this research, mathematical theory of evidence has been successfully developed to assess student’s entrance examination and displaying the result of identification process.
Research Article
Development of a Smart Program for the Intellectual Style Inventory
Mai Sabry Saleh and Yasser Zakaria Zaki
Background and Objective: The intellectual style inventory (ISI) is a reliable and valid tool for learning style assessment. It is based on the most popular and recent theories that describe learning. The ISI introduces four styles of thinking and four styles of perception and roots them in the four cortical lobes of the brain. It is able to describe one’s natural lead in learning with respect to cortical preference in thinking and perception and produce different learning profiles, accordingly. The aim of the present study was to use potentials of software programming in converting the paper based psychometric tool of the ISI into a readymade, user-friendly program. Materials and Methods: The software was developed using visual basic and microsoft access. It was then tested on 42 volunteers working at the National Research Centre of Egypt. Volunteers successfully used the program and interesting data were collected describing their learning styles. Results collected were analyzed statistically using Statistical Package for the Social Sciences (SPSS). Descriptive statistics, comparative test and chi-square test were done for all data. Results: Females showed significantly higher score on the base right thinking style at p<0.01. Similarly, scores for the front left thinking style as first preference were significantly higher than perception style at the same lobe. The developed software program succeeded to represent the ISI and to introduce it to end user in an interesting and easy way. Conclusion: The ISI software program is recommended to be used by practitioners in fields of education, human resources, counseling, research and others. As an extension of the MBTI, the ISI could enrich the field of personality software engineering and introduce a new material that could expand knowledge about personality theory and application. The ISI has many advantages as it is designed in Arabic and English, easy, accurate, can easily run on most operating systems with simple installation process and introduces scientific reporting for individuals about their learning abilities.
Research Article
Soft Computing Based Cluster-Head Selection in Mobile Ad-Hoc Network
Jay Prakash, Deepak Kumar Gupta and Rakesh Kumar
Background and Objective: Mobile ad-hoc network (MANET) is a specific type of network that can be quickly deployed without any existing framework. Cluster formation and cluster head selection in MANET is an important issue in such networks. Clustering is one of the vital issues used in increasing the network life time by gathering the information from specific group of nodes and forwarding it to other neighbouring cluster heads. This paper propose a soft computing based approach for the selection of the cluster head in MANET. A cluster head selection model based on fuzzy logic has been devised. Cluster head selection is done on the basis of three parameters viz., residual energy, centrality and hop-count. Materials and Methods: The proposed approach has been implemented in MATLAB followed by execution of cluster-head selection based on fuzzy logic using 3 criteria viz; residual energy, hop count and centrality. Results: The benefits of this approach include reduced overhead, improved performance of cluster node selection and increased network lifetime. Conclusion: Through simulation, it has been observed that this approach outperforms over existing approaches. The model has also been analytically validated. The new important aspects can be integrating mobility and trust to the existing model as fourth parameter for cluster-head selection that helps in improving the network performance.

Frequently Asked Questions    |    Privacy Policy    |    Terms & Conditions    |    Contact Us