RECOMMENDER SYSTEM ON STRUCTURED DATA

Recommender system for documents ranking

The publications of research papers are increasing exponentially and to find the relevant paper to a research area or according a user query is crucial task. It is due to the large corpus of text data in different formats. If a user query on search engine to read a research paper, it will provide him a bundle of papers, now to find the relevancy of the paper; user will read one by one each paper. There are some techniques already exist as TF*IDF with many variations and content match search cited by and co-citation. Document ranking and the vector space model is nearest to our system. In this paper, we used a new approach for recommender system , it examine identify paper section then assign weights to each section , applying Paper section weights to determine what documents are more relevant to a user query. When papers retrieved then rank it on the basis of section weights. each paper ha four main sections are Paper Title, Abstract, Keywords, Introduction and Conclusion. In online survey we get best results against paper content match.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.