Efficient Recommender System by Implicit Emotion Prediction
Recommender systems are widely used in almost all domains to recommend products based on user’s preference. However, there are several ongoing debates about increasing the efficiency with which recommendations are made to the user. So, nowadays, recommender systems not just considers user’s preference, but also take into account the emotional state of the user to make recommendations. This paper aims at getting user’s emotion implicitly by taking into account the time spent on different parts of the webpage. If any of these meet the predefined threshold, the user’s emotion is analysed based on mouse movement in that part of the webpage. Thus, from this emotion, one gets to know whether the user is actually interested in the content of that part of the webpage. Thus, the project aims to improve the efficiency of recommendations by providing a personalized recommendation to each user.
KeywordsAffective computing Recommender system Information overload
- 1.Madhukar, M.: Challenges & limitation in recommender systems. Int. J. Latest Trends Eng. Technol. (IJLTET) 4(3) (2014)Google Scholar
- 2.Sree Lakshmi, Soanpet, Adi Lakshmi, T.: Recommendation systems: issues and challenges. (IJCSIT) Int. J. Comput. Sci. Inf. Technol. 5(4), 5771–5772 (2014)Google Scholar
- 3.Sharma, L., Gera, A.: A survey of recommendation system: research challenges. Int. J. Eng. Trends Technol. (IJETT) 4(5) (2013)Google Scholar
- 4.Jain, S., Grover, A., Thakur, P.S., Choudhary, S.K.: Trends, Problems and Solutions of Recommender System. ISBN: 978-1-4799-8890-7/15/$31.00 ©2015 IEEEGoogle Scholar
- 5.Shikder, R., Rahaman, S., Afroze, F., Alim Al Islam, A.B.M.: Keystroke/mouse usage based emotion detection and user identification. 978-1-5090-3260-0/17/$31.00 ©2017 IEEEGoogle Scholar