Machine Learning Using K-Nearest Neighbor for Library Resources Classification in Agent-Based Library Recommender System

Abstract

Agent-based library recommender system is proposed with the objective to provide effective and intelligent use of library resources such as finding right book(s), relevant research journal papers, and articles. It is composed of profile agent and library recommender agent. Library recommender agent performs the main task of filtering and providing recommendations. Library resources include book records having table of contents and journal articles including abstract and keywords. This provides availability of rich set of keywords to compute similarity. The library resources are classified into fourteen categories specified in ACM computing classification system 2012. The identified category provides a way to obtain semantically related keywords for the library resources. The results of k-Nearest Neighbor (k-NN) for library recommender system are encouraging as there is improvement in the existing results. Use of ACM CCS 2012 as ontology, semantic similarity computation, implicit auto update of user profiles, and variety of users in evaluation are the features of the complete recommender system which makes it useful and novel. This paper details classification of library resources performed by library recommender agent.