Personalized Content Recommendation Engine for Web Publishing Services Using Textmining and Predictive Analytics

Personalized Content Recommendation Engine for Web Publishing Services Using Textmining and Predictive Analytics

Başar Öztayşi (Istanbul Technical University, Turkey), Ahmet Tezcan Tekin (Kontra Digital, Turkey), Cansu Özdikicioğlu (Kontra Digital, Turkey) and Kerim Caner Tümkaya (Kontra Digital, Turkey)
Copyright: © 2017 |Pages: 12
DOI: 10.4018/978-1-5225-2148-8.ch007
OnDemand PDF Download:
No Current Special Offers


Recommendation systems have become very important especially for internet based business such as e-commerce and web publishing. While content based filtering and collaborative filtering are most commonly used groups in recommendation systems there are still researches for new approaches. In this study, a personalized recommendation system based on text mining and predictive analytics is proposed for a real world web publishing company. The approach given in this chapter first preprocesses existing web contents, integrate the structured data with history of a specific user and create an extended TDM for the user. Then this data is used for prediction of the users interest in new content. In order to reach that point, SVM, K-NN and Naïve Bayesian methods are used. Finally, the best performing method is used for determining the interest level of the user in a new content. Based on the forecasted interest levels the system recommends among the alternatives.
Chapter Preview


Recommendation Systems have been widely investigated in industrial, academic, and educational fields. There are many approaches in literature. In the literature, the best known approaches are content based filtering and collaborative filtering (CBF). In CBF, items are grouped in specific properties. When users register a system firstly user profile is created for each of them. The profile is defined by items which are examined, liked or bought by users before. Based on this user profile, list of item recommendation is defined. Schein & Popescul (2002) underline that content based filtering ignore the preferences of the other users. Each recommendation engine has its own advantages and disadvantages when compared with others. In previous decades hybrid approaches are proposed in order to eliminate their disadvantages (Resnick et al.,1997). Recommendation systems can be classified in five main categories. These categories can be given as:

Complete Chapter List

Search this Book: