Eyas El-Qawasmeh

Performance of KNN and SVM classifiers on full word Arabic articles

This paper reports a comparative study of two machine learning methods on Arabic text categorization. Based on a collection of news articles as a training set, and another set of news articles as a testing set, we evaluated K nearest neighbor (KNN) algorithm, and support vector machines (SVM) algorithm. We used the full word features and considered the tf.idf as the weighting method for feature selection, and CHI statistics as a ranking metric. Experiments showed that both methods were of superior performance on the test corpus while SVM showed a better micro average F1 and prediction time.

Improving arabic text categorization using decision trees

This paper presents the results of classifying Arabic text documents using a decision tree algorithm. Experiments are performed over two self collected data corpus and the results show that the suggested hybrid approach of Document Frequency Thresholding using an embedded information gain criterion of the decision tree algorithm is the preferable feature selection criterion. The study concluded that the effectiveness of the improved classifier is very good and gives generalization accuracy about 0.93 for the scientific corpus and 0.91 for the literary corpus and we also conclude that the effectiveness of the decision tree classifier was increased as we increase the training size, and the nature of the corpus has such a influence on the classifier performance.

Semantic Analysis of Web Pages Using Web Patterns

This paper introduces a novel method for semantic analysis of Web pages. Analysis is performed with regard to unwritten and empirically proven agreement between users and Web designers using Web patterns. This method is based on extraction of patterns which are characteristics for concrete domain. Patterns provide formalization of the agreement and allow assignment of semantics to parts of Web pages. Experimental results verify the effectives of the proposed method

Compression of small text files

This paper suggests a novel compression scheme for small text files. The proposed scheme depends on Boolean minimization of binary data accompanied with the adoption of Burrows-Wheeler transformation (BWT) algorithm. Compression of small text files must fulfil special requirements since they have small context. The use of Boolean minimization and Burrows-Wheeler transformation generate better context information for compression with standard algorithms. We tested the suggested scheme on collections of small and medium-sized files. The testing results showed that proposed scheme improve the compression ratio over other existing methods.

Semantic Annotation of Web Pages Using Web Patterns

This paper introduces a novel method for semantic annotation of web pages. We perform semantic annotation with regard to unwritten and empirically proven agreement between users and web designers using web patterns. This method is based on extraction of patterns, which are characteristic for a particular domain. A pattern provides formalization of the agreement and allows assigning semantics to parts of web pages. We will introduce experiments with this method and show its benefits for querying the web.

Vector space model for Arabic information retrieval — application to “Hadith” indexing

The Arabic language is one of the most important languages because it is the sacred and liturgical language of Islam, one of the influential monotheistic religions of our times. In the post-9/11 aftermath, Islam suddenly dominated western actuality for the remaining years of the present decade. Al-Qurpsilaan - The Reading par Excellence - and ldquoHadithrdquo - Saying - represent the two fundamental scriptural sources of Islamic Legislation. Specifically, ldquoHadithrdquo, or Prophetic Traditions, are sayings and doings of the Prophet of Islam (Peace and Blessings be upon Him). Researchers need automatic search tools within large ldquoHadithrdquo databasesto access one of the original sources of Islam. For this purpose, we describe the development of AuthenTique, an updated automatic text mining search tool, based on the vector space model (VSM). The aim is to allow the provision of a list of ldquoHadithsrdquo classified according to their degrees of similarity based on a given userpsilas query.

Digital Enterprise and Information Systems

This volume constitutes the refereed proceedings of the International Conference on Digital Enterprise and Information Systems, held in London during July 20 - 22, 2011. The 70 revised full papers presented were carefully reviewed and selected. They are organized in topical sections on cryptography and data protection, embedded systems and software, information technology management, e-business applications and software, critical computing and storage, distributed and parallel applications, digital management products, image processing, digital enterprises, XML-based languages, digital libraries, and data mining.

A quadtree-based representation technique for indexing and retrieval of image databases

Abstract Currently, several approaches for image indexing based on the quadtrees exist. In this paper, we propose a new organization for image databases combined with the corresponding algorithm for image search by example. The suggested organization uses the quadtrees to splits the database into multi-subsets by adding some extra fields to facilitate the image search. We suggest a centroid partial match algorithm to process the search query. The algorithm selects random points from an image in a circular uniform movement to check for image match. The proposed organization searches a subset of the image database rather than the whole database. It is flexible since the number of subsets in the database is variable. The centroid image algorithm permits the search regardless of the image size. Both the database organization and the centroid algorithm guarantee that the precision and the recall maximum values are achievable.

A Comparative Study of Statistical Feature Reduction Methods for Arabic Text Categorization

Feature reduction methods have been successfully applied to text categorization. In this paper, we perform a comparative study on three feature reduction methods for text categorization, including Document Frequency (DF), Term Frequency Inverse Document Frequency (TFIDF) and Latent Semantic Analyses (LSA). Our feature set is relatively large (since there are thousands of different terms in different texts files). We propose the use of the previous feature reduction methods as a preprocessor of Back-Propagation Neural Network (BPNN) to reduce the input data on training process. The experimental results on an Arabic data set demonstrate that among the three dimensionality reduction techniques proposed, TFIDF was found to be the most effective in reducing the dimensionality of the feature space.

New method for ranking arabic web sites using ontology concepts

Recently the numbers of Arabic web sites are rapidly increasing in the World Wide Web. Existing search engines retrieve information based on keywords, so huge number of irrelevant information is retrieved for users. By the appearing of the second generation of the World Wide Web the semantic web, many suitable approaches to retrieve information that depends on semantic can be built. Semantic web provides data models and languages such as Resource Description Framework (RDF), and Web Ontology Language (WOL) that facilitate building ontology within a specific domain. Ontology can capture concepts for specific domain. In addition, it can capture properties of these concepts, and their relationships. Therefore, they help machines to deal with data domain semantically. This paper proposes a new approach to measure the relevancy of Arabic documents to the user query using ontology concepts. In this paper we built Arabic ontology concepts for the electronic commerce domain in Arabic language. Ontology concepts are used to find new approach for ranking Arabic documents, and show the effectiveness of ontology in retrieving relevant documents.