Minimizing Web Diversion Using Query Classification and Text Mining

The process of obtaining relevant search outcomes through web search engines is often complicated due to the usage of short, ambiguous, and noisy queries. Moreover, there is the absence of standard taxonomies and detailed semantics in target categories frequently. To address this challenge, this paper proposes a methodology that classifies these web pages based on their content, thereby resulting in an improved search system for end users. Unlike pure-text categorization, web page classification must contend with a wide range of noisy information present on web pages. The system in this paper aims to classify the search results for users interested in specific categories. The proposed system in this paper intends to provide better search result pages for users who have interests in the intended categories. Furthermore, according to the experimental results, the proposed system achieves an approximate of 86.71% for the best-performing model.