University Links: Home Page | Site Map
Covenant University Repository

Knowledge Discovery in Online Repositories: A Text Mining Approach

Afolabi, I. T. and Musa, G. A. and Ayo, C. K. and Sofoluwe, A. B. (2008) Knowledge Discovery in Online Repositories: A Text Mining Approach. European Journal of Scientific Research, 22 (2). pp. 241-250. ISSN 1450-216X

[img] PDF
Download (178kB)


Before the advent of the Internet, the newspapers were the prominent instrument of mobilization for independence and political struggles. Since independence in Nigeria, the political class has adopted newspapers as a medium of Political Competition and Communication. Consequently, most political information exists in unstructured form and hence the need to tap into it using text mining algorithm. This paper implements a text mining algorithm on some unstructured data format in some newspapers. The algorithm involves the following natural language processing techniques: tokenization, text filtering and refinement. As a follow-up to the natural language techniques, association rule mining technique of data mining is used to extract knowledge using the Modified Generating Association Rules based on Weighting scheme (GARW). The main contributions of the technique are that it integrates information retrieval scheme (Term Frequency Inverse Document Frequency) (for keyword/feature selection that automatically selects the most discriminative keywords for use in association rules generation) with Data Mining technique for association rules discovery. The program is applied to Pre-Election information gotten from the website of the Nigerian Guardian newspaper. The extracted association rules contained important features and described the informative news included in the documents collection when related to the concluded 2007 presidential election. The system presented useful information that could help sanitize the polity as well as protect the nascent democracy.

Item Type: Article
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
Depositing User: Mr Adewole Adewumi
Date Deposited: 29 Nov 2012 01:17
Last Modified: 18 Feb 2014 11:16

Actions (login required)

View Item View Item