Sentiment analysis of social network data for cold-start relief in recommender systems

Abstract

Recommender systems have been used in e-commerce to increase conversion due to matching product offer and consumer preferences. Cold-start is the situation of a new user about whom there is no information to make suitable recommendations. Texts published by the user in social networks are a good source of information to reduce the cold-start issue. However, the valence of the emotion in a text must be considered in the recommendation so that no product is recommended based on a negative opinion. This paper proposes a recommendation process that includes sentiment analysis to textual data extracted from Facebook and Twitter and present results of an experiment in which this algorithm is used to reduce the cold-start issue.