into a predefined set of categories based on its contents of keywords. Text classification is an extended type of text categorization where the text is further categorized into sub- categories. Many algorithms have been proposed and implemented to solve the problem of English text categorization and classification. However, few studies have been carried out for categorizing and classifying Arabic text. Compared to English, the Arabic text classification is considered as a very challenging due to the Arabic language complex linguistic structure and its highly derivational nature where morphology plays a very important role. This paper proposes a new method for Arabic text classification in which a document is compared with pre-defined documents categories based on its contents using the TF.IDF method (Term Frequency times Inverse Document Frequency) measure, then the document is classified into the appropriate sub-category using Chi Square measure..

/pdf/arabic-text-classification-algorithm-using-tfidf-and-chi-1lbtohlqrb.pdf

Arabic Text Classification Algorithm using TFIDF and Chi Square Measurements

Feature selection is necessary for effective text classification. Dataset preprocessing is essential to make upright result and effective performance. This paper investigates the effectiveness of using feature selection. In this paper we have been compared the performance between different classifiers in different situations using feature selection with stemming, and without stemming.Evaluation used a BBC Arabic dataset, different classification algorithms such as decision tree (D.T), K-nearest neighbors (KNN), Naive Bayesian (NB) method and Naive Bayes Multinomial(NBM) classifier were used. The experimental results are presented in term of precision, recall, F-Measures, accuracy and time to build model.

/pdf/different-classification-algorithms-based-on-arabic-text-31ma3q4ce2.pdf

Different Classification Algorithms Based on Arabic Text Classification: Feature Selection Comparative Study

Purpose – Citation analysis is an important measure for the assessment of quality and impact of academic entities (authors, papers and publication venues) used for ranking of research articles, authors and publication venues. It is a common observation that high-level publication venues, with few exceptions (Nature, Science and PLOS ONE), are usually topic specific. The purpose of this paper is to investigate the claim correlation analysis between topic specificity and citation count of different types of publication venues (journals, conferences and workshops). Design/methodology/approach – The topic specificity was calculated using the information theoretic measure of entropy (which tells us about the disorder of the system). The authors computed the entropy of the titles of the papers published in each venue type to investigate their topic specificity. Findings – It was observed that venues usually with higher citations (high-level publication venues) have low entropy and venues with lesser citations (not-high-level publication venues) have high entropy. Low entropy means less disorder and more specific to topic and vice versa. The input data considered here were DBLP-V7 data set for the last 10 years. Experimental analysis shows that topic specificity and citation count of publication venues are negatively correlated to each other. Originality/value – This paper is the first attempt to discover correlation between topic sensitivity and citation counts of publication venues. It also used topic specificity as a feature to rank academic entities.

/pdf/linked-open-data-of-bibliometric-networks-analytics-research-5cnvga818d.pdf

Linked open data of bibliometric networks: analytics research for personalized library services

Documents indexing is the main step in a conventional document classification or information retrieval framework. This study aims to highlight the influence of features' type on the efficiency of a classification system. Empirical results on Arabic dataset reveal that the choice of extracted feature's type has a significant impact on conserving semantic information and improving classification accuracy, especially with the morphological complexity of the Arabic language. Precision, recall, F-measure and accuracy are the metrics adopted to compare the efficiency of the proposed indexing system.

Stemming versus multi-words indexing for Arabic documents classification

This research paper demonstrates the work accomplished in the last phase of the ongoing research project with an objective of developing a system for moving Arabic video text extraction for efficient content-based indexing and searching. The novelty of this paper is the technique used for concatenation of the individual stand alone Arabic characters which are extracted and recognized from image frames. Unicode format of Arabic characters is used for concatenation of extracted characters which is never done before. The concatenated characters are written into the text file in incessant way. This text files are indexed using Lucene and search for the desired string is done in a faster and precise manner.

Concatenation Technique for Extracted Arabic Characters for Efficient Content-based Indexing and Searching

The continuous information explosion through the Internet and all information sources makes it necessary to perform all information processing activities automatically in quick and reliable manners. In this paper, we proposed and implemented a method to automatically create and Index for books written in Arabic language. The process depends largely on text summarization and

Abdulrahman Al Molijy

Papers

Indexing of Arabic documents automatically based on lexical analysis