Recent News Recommender Using Browser’s History

Abstract

Identifying users’ interests to use in a recommender system is a difficult process because it needs a huge amount of data to analyze. On the other hand, each user has many historical data stored by the browser in his computer such as the visited pages including its title, visit date, URLs…etc. However, these data are not manipulated or analyzed properly to be used in recommender systems. In this research, we are proposing to use natural language processing techniques on the browser’s historical data, which collected from user’s actions on the browser to improve the recommendation systems in a way that helps to recommend and select news from recent news site based on the user’s interests and history. We experimented with our technique by developing an application that extracts and analyzes the historical data from the browser’s history. In addition, use the previously analyzed data in a recommender system. The experiments showed good results as discussed in the implementation part of this paper. The use of the browser’s history in recommender systems will enhance the recommendation process. This data can be used in the recommendation of news and other domains such as the advertisements, research topics, etc.