Book recommended formulation based on multiple bibliographic information

We proposed the formulation for discovering the candidate bibliographic records for the primary resource by integrating multiple bibliographic parameters including call number, subject heading, and title's keyword. To imitate the searching procedure of librarians, we can customize weights for assigning different priority for different bibliographic parameters. The precision and recall of the retrieval lists were compared to the performance of librarians. The results show that the quality of the recommended list formulated by our proposed system is comparable to the list from librarian. However, the performance of the proposed system can be improved by integrating NLP algorithm, and more bibliographic information.