A User Query-Centered Recommender System from Public Repository

Abstract

Query-based User Information Categorization and Extraction (QICE) methods allow the classical query extraction with its knowledge obtained from useful resources. Open Data encodes machine-readable user facts from different sources, including third-party, that play a vital role in this QICE. Mining techniques from documents available in free sources and constructing the user text based on the user query with its knowledge and analysis are the core research problems in the public repositories and query understanding tasks such as query pattern analysis and user information requirements. However, the public repositories encode the user information through Wikipedia and web pages which are static, and these do not understand the user requirement perspectives. These static web pages have many quality issues with user query information, such as information extracted, complete data representing, time of query retrieval, and correctness of the information categorized based on the user query. QICE methods are, therefore, facing problem user query variances and type of user query confusability. In this paper, a query recommender system proposes developing a technique for user-query-centered knowledgeable integration and addressing the challenges of knowledge mining of the Twitter social network by extracting the knowledge from query log data. The proposed User-Query Centered Recommender System (UQCRS) is applied to exploit different measures to demonstrate the efficiency of recommendations delivered. The proposed algorithm exhibits an effective result to the search shortcuts issues. And to provide a performance comparison of the proposed system, the comprehensive evaluation is compared with well-known methods and demonstrates the impact of the results.