An Application of Recommendation Systems to Cluster-based Pattern Recognition

Ankush  Singh; Bhavana  Pillai

Abstract

Information overload is a common problem faced by internet users; so, the recommendation system is becoming increasingly important to this service. Collective filtering is used by recommendation systems to offer recommendations to individuals or for products based on their shared interests. Data or people should be grouped into clusters so that they have more in common with those in the same cluster and stand out from those in other clusters. Parameters like average, correlation, mutual information, etc. are used in pattern discovery and matching. To extract and organise patterns into categories, PR may be used as a classification approach. PR is a way of categorising. Public relations (PR) is a hybrid method. In this survey, we investigated data mining, clustering, recommendation systems, and pattern recognition. We defined numerous clustering and pattern recognition algorithms and discussed their applications.

Abstract

Keywords