[学术交流] 2012年6月26日:A two-step Recommendation Algorithm via..

题目: A two-step Recommendation Algorithm via Iterative Local Least Squ

摘要 Local Least Squares(LLS) method is one of three main methods in microarray missing value imputation. Based on the idea of this approach, we proposed a two-step recommendation algorithm via Iterative Local Least Squares(ILLS) which is the iterative way of LLS. In our work,  we preprocess the “ratings” matrix through ProbS which can convert, the sparse data to a dense one. Then we  use ILLS to estimate those missing values. Finally, the recommendation list is generated. Experimental results on the three datasets: MovieLens, Netflix, RYM, suggest that the proposed method can enhance the algorithmic accuracy of AUC. Especially, it performs much better in dense datasets. Furthermore, since  this methods can improve those missing value more accurately via iteration which  might show light in discovering those inactive users’ purchasing intention and eventually solving  cold-start  problem.

报告人: 刘金虎 博士 时间:2012年6月26日 18H30 地点:110