Earticle

초록

영어

Recommendation systems sort out the information of user’s concerns for supporting decision-making. Today, recommendation systems have a very close relationship with our modern society. However, despite the large amount of information available due to information technological advancement, finding information specific to the user's concern is getting more difficult. In order to handle such issues, the importance of the recommendation system has become apparent. Collaborative filtering is one of the referral systems, which automatically predicts the users’ interest based on the information on preference collected from a considerable number of people. However, accuracy issues come to the fore as an insufficient amount of information collected. This paper derived a regression equation using collaborative filtering of user preference information and official movies information to solve the problems, thereby proposing a movie recommendation system. By adding user preference information to the regression equation using only objective movie information, accuracy has been increased by 20%, and the recall ratio by 9%. It has been shown that utilizing preference information increases accuracy for recommendation of movies.

목차

Abstract
 1. Introduction
 2. Related Research
  2.1. Recommendation System
  2.2. Collaborative Filtering
  2.3. Cold-Start
  2.4. Multiple Regression Analysis
 3. Suggesting System
  3.1. Data
  3.2. Probability Prediction
 4. Result
 5. Conclusion
 References

키워드

저자

• Sang-Hyun You [ Graduate school of Software, Soongsil University, Seoul, Korea ]
• Jeawon Park [ Graduate school of Software, Soongsil University, Seoul, Korea ]
• Jaehyun Choi [ Graduate school of Software, Soongsil University, Seoul, Korea ] Corresponding author

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Multimedia and Ubiquitous Engineering

• 간기

  월간

• pISSN

  1975-0080

• 수록기간

  2008~2016

• 등재여부

  SCOPUS

• 십진분류

  KDC 505 DDC 605