A graph based approach to inferring item weights for pattern mining
作者：
Highlights：
• We infer unknown items weights from a small subset of known landmark weights.
• Domain specific knowledge is used to guide the weight inference process.
• A graph model is used to model items and their interactions.
• Significant improvements in Precision and Recall over previous research.
• The novel graph partitioning algorithm shows significant gains in execution time.
摘要
•We infer unknown items weights from a small subset of known landmark weights.•Domain specific knowledge is used to guide the weight inference process.•A graph model is used to model items and their interactions.•Significant improvements in Precision and Recall over previous research.•The novel graph partitioning algorithm shows significant gains in execution time.
论文关键词：Semi supervised classification,Weighted association rule mining,WeightTransmitter model
论文评审过程：Available online 7 August 2014.
论文官网地址：https://doi.org/10.1016/j.eswa.2014.07.030
