Universiti Teknologi Malaysia Institutional Repository

Tfidf and entropy for sports, news and gambling web content classification

Dawood, M. and Ibrahim, O. B. and Abali, M. (2019) Tfidf and entropy for sports, news and gambling web content classification. International Journal of Recent Technology and Engineering, 8 (1). pp. 25-30. ISSN 2277-3878

Full text not available from this repository.

Official URL: https://www.ijrte.org/wp-content/uploads/papers/v8...


The exponential increase in online data or information brought up the issue of information security. Gambling web content is one of the greatest harmful res ources that pollutes children's and adolescents ' minds by disguis ing sports and News Web pages. Gambling can be adopted by any gender or age. Several web content-based analytical approaches were proposed to prevent children from accessing these illegal web content. They are introduced to the Internet at an early age. Most approaches are weak to classify web content of high similarity such as Gambling, Sports and News Web pages. In this paper two existing term weighting schemes namely TFIDF and Entropy are used as feature selection process in filtering website. We examine the performance of both techniques via datasets and compare it with the term weighting schemes. The suitability of these term weighting schemes as the selection of features is measured according to the accuracy of the results obtained using the classification program known as Support Vector Machine (SVM). In this paper the performance of TFIDF and Entropy is judged on the basis of Accuracy. Results showed that TFIDF performed better than Entropy. On average, TFIDF obtained 97% and Entropy 91% accuracy. This study is hoped to give other researchers an insight, especially those who would like to work in the same area.

Item Type:Article
Uncontrolled Keywords:term weighting schemes, TFIDF, URL
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
ID Code:89275
Deposited By: Narimah Nawil
Deposited On:09 Feb 2021 12:26
Last Modified:09 Feb 2021 12:26

Repository Staff Only: item control page