		FeedBack \| Contact Us \| Links \| Site Map

Welcome to Journal of Theoretical and Applied Information Technology

Submit Paper / Call for Papers

Journal receives papers in continuous flow and we will consider articles from a wide range of Information Technology disciplines encompassing the most basic research to the most innovative technologies. Please submit your papers electronically to our submission system at http://jatit.org/submit_paper.php in an MSWord, Pdf or compatible format so that they may be evaluated for publication in the upcoming issue. This journal uses a blinded review process; please remember to include all your personal identifiable information in the manuscript before submitting it for review, we will edit the necessary information at our side. Submissions to JATIT should be full research / review papers (properly indicated below main title).

Journal of Theoretical and Applied Information Technology May 2011 | Vol. 27. No.1

Paper ID:	14148 -JATIT
Title:	NEURAL NETWORKS AND INVESTOR SENTIMENT MEASURES FOR STOCK MARKET TREND PREDICTION
Author:	SALIM LAHMIRI
Source:	Journal of Theoretical and Applied Information Technology Vol 27. No. 1 -- 2011
Abstract	Soft computing methods and various sentiment indicators are employed to conduct out-of-sample predictions of the future sign of the stock market returns. In particular, we assess the performance of the probabilistic neural network (PNN) against the back-propagation neural network (BPNN) in predicting technology stocks and NYSE up and down moves. Genetic algorithms (GA) are employed to optimize the topologies of the BPNN. Our results from Granger causality tests show strong evidence that all stock returns are strongly related to at least one of the sentiment variables. In addition, the results from simulations show that the GA-BPNN is more capable of distinguishing between market ups and downs than the PNN. Finally, the simulations show that trading given decision rules (for example; buy stock if predicted return is higher than a given threshold) yields to higher accuracy than predicting the stock market ups and downs.
Keywords	Artificial Intelligence, Classification, Stock Market
Full Text

Paper ID:	14181 -JATIT
Title:	PRACTICAL GUIDELINES FOR SUCCESSFUL ERP TESTING
Author:	AMEL AL-HOSSAN, ABDULLAH S. AL-MUDIMIGH
Source:	Journal of Theoretical and Applied Information Technology Vol 27. No. 1 -- 2011
Abstract	Testing of Enterprise Resource Planning (ERP) system comprises the activities that are used to validate business processes and the rules that govern them and removing or reducing maximum operational risk within available resources and time schedule constraints. This paper reviews some researches performed in ERP testing to identify the common challenges, failures and the proposed solutions. It also presents guidelines to help achieving success ERP testing. To support this paper with real life example, we included the testing methodologies and the best practices that are applied in the Government Resource Planning (GRP) system of King Saud University.
Keywords	MADAR, ERP, Testing, Hasib, Organization, GRP
Full Text

Paper ID:	14168 -JATIT
Title:	A COMPARATIVE ANALYSIS BETWEEN K-MEDOIDS AND FUZZY C-MEANS CLUSTERING ALGORITHMS FOR STATISTICALLY DISTRIBUTED DATA POINTS
Author:	T.VELMURUGAN, T.SANTHANAM
Source:	Journal of Theoretical and Applied Information Technology Vol 27. No. 1 -- 2011
Abstract	Data clustering is a process of putting similar data into groups. A clustering algorithm partitions a data set into several groups such that the similarity within a group is larger than among groups. In the field of data mining, various clustering algorithms are proved for their clustering quality. This research work deals with, two of the most representative clustering algorithms namely centroid based K-Medoids and representative object based Fuzzy C-Means are described and analyzed based on their basic approach using the distance between two data points. For both the algorithms, a set of n data points are given in a two-dimensional space and an integer K (the number of clusters) and the problem is to determine a set of n points in the given space called centers, so as to minimize the mean squared distance from each data point to its nearest center. The performance of the algorithms is investigated during different execution of the program for the given input data points. Based on experimental results the algorithms are compared regarding their clustering quality and their performance, which depends on the time complexity between the various numbers of clusters chosen by the end user. The total elapsed time to cluster all the data points and Clustering time for each cluster are also calculated in milliseconds and the results compared with one another.
Keywords	K-Medoids Algorithm, Fuzzy C-Means Algorithm, Cluster Analysis, Data Analysis
Full Text

Paper ID:	14205 -JATIT
Title:	KNOWLEDGE BASED APPROACH FOR ALIGNMENT PROBLEMS
Author:	S.SAMUNDEESWARI, M.THIYAGARAJAN
Source:	Journal of Theoretical and Applied Information Technology Vol 27. No. 1 -- 2011
Abstract	Machine vision is an area in which all problems related to image analysis are handled in a different outlook. In analyzing biomedical images and other coherent imaging systems one is interested in identifying the part from the whole. This is done usually by adopting different similarity measures like joint entropy. Here a knowledge base is created on which an affine transmission having specific translation and rotation are used to complete the solution to the above problem besides the use of statistical ZKIP (zero knowledge interactive protocol) based on mutual information differences. This method solves the problem with 95% confidence level while compared with earlier techniques.
Keywords	Alignment Problem, Affine Transformation, AI and Machine Vision, Joint Entropy
Full Text

Paper ID:	14174 -JATIT
Title:	PROBABILISTIC NEURAL NETWORK – A BETTER SOLUTION FOR NOISE CLASSIFICATION
Author:	T. SANTHANAM1, S. RADHIKA
Source:	Journal of Theoretical and Applied Information Technology Vol 27. No. 1 -- 2011
Abstract	Classification is one of the common tasks of human behavior. Classification problems arise when an entity needs to be assigned into a predefined set based on a number of features associated with that entity. Neural Network models prove to be a competitive alternative to traditional classifiers for many practical classification problems. Noise classification in digital image processing is a must so as to identify the suitable filters for smoothing the image for further processing. The use of Probabilistic Neural Network to classify the noise present in an image after extracting the statistical features like skewness and kurtosis is explored in this article. When the noises are classified accurately, identification of the filter becomes an easy task.
Keywords	Probabilistic Neural Network, Noise Classification, Statistical Features
Full Text

Paper ID:	14169 -JATIT
Title:	GENE EXPRESSION ANALYSIS FOR TYPE-2 DIABETES MELLITUS – A STUDY ON DIABETES WITH AND WITHOUT PARENTAL HISTORY
Author:	CHANDRA SEKHAR VASAMSETTY, Dr. SRINIVASA RAO PERI, Dr. ALLAM APPA RAO, Dr. K. SRINIVAS, CHINTA SOMESWARARAO
Source:	Journal of Theoretical and Applied Information Technology Vol 27. No. 1 -- 2011
Abstract	Diabetes mellitus, simply referred to as diabetes, is a group of metabolic diseases. There exits more than one type of diabetes, with each type having its own risks. Among the different types, types 1 and 2 are the most common ones. The cause of diabetes depends on the type. In each case, combinations of genetic and environmental influences are responsible for causing diabetes. Type 2 diabetes is primarily due to lifestyle factors and genetics. Microarray analysis is a method for analyzing expression levels of multiple genes at once. This method is especially suitable for identifying and classifying genes whose expression level differs in two samples. The present work focuses on identifying and classifying genes that cause type-II diabetes with two different samples, one with parental history and other without parental history. Mahalanobis Distance, Minimum Co-variance Determinant are the statistical methods used for identifying multivariate and univariate outliers for the identified inflammatory genes, the functional classification is performed by using Gene Ontology and pathway analysis. It is observed that 38 differentially expressed genes were identified out of 39400 genes tested between diabetes with and without parental history.
Keywords	Type-2 Diabetes mellitus, Mahalanobis Distance, Gene Ontology, pathway analysis, Microarray analysis
Full Text

Paper ID:	14186 -JATIT
Title:	MINING CUSTOMER DATA FOR DECISION MAKING USING NEW HYBRID CLASSIFICATION ALGORITHM
Author:	Aurangzeb Khan, Baharum Baharudin, Khairullah Khan
Source:	Journal of Theoretical and Applied Information Technology Vol 27. No. 1 -- 2011
Abstract	Classification and patterns extraction from customer data is very important for business support and decision making. Timely identification of newly emerging trends is needed in business process. Sales patterns from inventory data indicate market trends and can be used in forecasting which has great potential for decision making, strategic planning and market competition. The objectives in this paper are to get better decision making for improving sale, services and quality as to identify the reasons of dead stock, slow-moving, and fast-moving products, which is useful mechanism for business support, investment and surveillance. In this paper we proposed an algorithm for mining patterns of huge stock data to predict factors affecting the sale of products. In the first phase, we divide the stock data in three different clusters on the basis of product categories and sold quantities i.e. Dead-Stock (DS), Slow-Moving (SM) and Fast-Moving (FM) using K-means algorithm. In the second phase we have proposed Most Frequent Pattern (MFP) algorithm to find frequencies of property values of the corresponding items. MFP provides frequent patterns of item attributes in each category of products and also gives sales trend in a compact form. The experimental result shows that the proposed hybrid k-mean plus MFP algorithm can generate more useful pattern from large stock data.
Keywords	Stock Data, Most Frequent Patterns, Clustering, Decision Making
Full Text