Online Article Recommender System

Abstract

Recommender System recommends relevant items to users, based on user habits or preference. Preference does not have quantitative measure. It is subjective matter. Generally it indirectly measure by items that consumed by users in past. There is a plethora of text available on the web and there are many online platforms that provide text (article) for reading. This is an attempt to develop a Recommender System (RecSys) for the article suggestion for the online article reading to the end user by the online article service provider. RecSys will use collaborative learning, content-based learning and combination of both, i,e, hybrid learning for the recommendation process. The proposed RecSys is tested and trained on is one article sharing platform service and it has been found that the hybrid learning model performed better than other.