[Recommender Systems][CF]Unifying User-based and Item-based CF by Similarity Fusion

Main Points

1. What is the problem that the paper wants to solve? Why is it difficult?

rating data의 sparsity 때문에 성능 안좋은 문제

 

2. What is the solution? What is the main idea?

predictions based on ratings of the same item by other users, predictions based on different item ratings made by the same user, and, third, ratings predicted based on data from other but similar users rating other but similar items. 

item+user based 각각의 장점을 합칠 수 있게 similarity 결과를 선형결합. because the different types of ratings are used in concert, while additional ratings from similar users towards similar items are employed as a background model to smooth the predictions.

data sparsity 문제 해결하기 위한 여러 method 논문을 정리해놨음

3. What is the result?

The experiments showed that our new fusion framework is effective in improving the prediction accuracy of collaborative filtering and dealing with the data sparsity problem.

 

Strengths

1. What is the main novelty that enabled the solution? 

similarity fusing

 

2. What are the good aspects of the paper? Did you learn something from the paper?

통계적 방식 활용

3. What is the impact of the paper?

?

 

Future Improvements

1. Are there weaknesses parts in the paper? How can you improve it?

실험결과 metric이 부족

2. How can you extend the paper?

 

3. How can you apply the technique to other data/problem?