International Journal of Advanced Research in Computer and Communication Engineering

A monthly peer-reviewed online and print journal

ISSN Online 2278-1021
ISSN Print 2319-5940

Abstract: In text mining applications, side-information is obtainable at the side of the text documents. Such side-information types are document provenance data, the links within the document, user-access behavior from net logs, or alternative non-textual attributes that are embedded into the text document. Such attributes might contain an incredible quantity of data for bunch functions. However, the relative importance of this side-information is also tough to estimate, particularly once a number of the data is noisy. In such cases, it is risky to include side-information into the mining method; as a result of it will either improve the standard of the illustration for the mining method, or will add noise to the method. Therefore, a principled means is needed to perform the mining method, thus on maximize the benefits from mistreatment this aspect information. The proposed system developing an application for recommendations of reports articles to the readers of a news portal. This paper designs an algorithmic rule which combines classical partitioning algorithms with probabilistic models so as to form an efficient clustering approach.

Keywords: Data Mining, Ontology Mining, Classification Model, Clustering, Automatic Analysis


PDF | DOI: 10.17148/IJARCCE.2018.71026