Active Learning and Deep Learning for the Cold-Start Problem in Recommendation System: A Comparative Study

With the advancement of technology there are millions of songs available on the internet and this creates problem for a person to choose from this vast pool of songs. So, there should be some middleman who must do this task on behalf of user and present most relevant songs that perfectly fits the user’s taste. This task is done by recommendation system. Music recommendation system predicts the user liking towards a particular song based on the listening history and profile. Most of the music recommendation system available today will give most recently played song or songs which have overall highest rating as suggestions to users but these suggestions are not personalized. The paper purposes how the recommendation systems can be used to give personalized suggestions to each and every user with the help of collaborative filtering which uses user similarity to give suggestions. The paper aims at implementing this idea and solving the cold start problem using content based filtering at the start.

Download Full-text

Active Learning and User Segmentation for the Cold-start Problem in Recommendation Systems

Proceedings of the 11th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management ◽

10.5220/0008162901130123 ◽

2019 ◽

Author(s):

Rabaa Alabdulrahman ◽

Herna Viktor ◽

Eric Paquet

Keyword(s):

Active Learning ◽

Recommendation Systems ◽

Cold Start ◽

Cold Start Problem

Download Full-text

Pairwise Preference Regression on Movie Recommendation System

Indonesian Journal on Computing (Indo-JC) ◽

10.21108/indojc.2019.4.1.255 ◽

2019 ◽

Vol 4 (1) ◽

pp. 57

Author(s):

Rita Rismala ◽

Rudy Prabowo ◽

Agung Toto Wibowo

Keyword(s):

Standard Deviation ◽

Recommendation System ◽

Cold Start ◽

Pairwise Preference ◽

Movie Recommendation ◽

New Item ◽

Cold Start Problem

Recommendation System is able to help users to choose items, including movies, that match their interests. One of the problems faced by recommendation system is cold-start problem. Cold start problem can be categorized into three types, they are: recommending existed item for new user, recommending new item for existed user, and recommending new item for new user. Pairwise preference regression is a method that directly deals with cold-start problem. This method can suggest a recommendation, not only for users who have no historical rating, but also for those who only have less demographic info. From the experiment result, the best score of Normalized Discounted Cumulative Gain (nDGC) from the system is 0.8484. The standard deviation of rating resulted by the recommendation system is 1.24, the average is 3.82. Consequently, the distribution of recommendation result is around rating 5 to 3. Those results mean that this recommendation system is good to solving cold-start problem in movie recommendation system.

Download Full-text