A multi-instance learning wrapper based on the Rocchio classifier for web index recommendation

作者：

Highlights：

• We propose a new multi-instance learning algorithm for web index recommendation.

• Compared to previously proposed methods, our algorithm has low computational cost.

• Our method outperforms state-of-the-art solutions on benchmark data, achieving in particular a high precision.

摘要

•We propose a new multi-instance learning algorithm for web index recommendation.•Compared to previously proposed methods, our algorithm has low computational cost.•Our method outperforms state-of-the-art solutions on benchmark data, achieving in particular a high precision.

论文关键词：Multi-instance learning,Multi-instance wrapper,Recommender system,Rocchio classifier,Web mining

论文评审过程：Received 31 July 2013, Revised 23 October 2013, Accepted 8 January 2014, Available online 29 January 2014.

论文官网地址：https://doi.org/10.1016/j.knosys.2014.01.008