Treelet kernel incorporating cyclic, stereo and inter pattern information in chemoinformatics
作者：
Highlights：
• Inclusion of chemical information into treelet kernel.
• Adaptation of multiple kernel learning to graph kernels based on bags of patterns.
• Two new molecular representations encoding explicit cyclic information.
• New relationship between maximum structural common subgraph and graph edit distance.
• Stereoisomerism is encoded in treelet kernel.
摘要
Highlights•Inclusion of chemical information into treelet kernel.•Adaptation of multiple kernel learning to graph kernels based on bags of patterns.•Two new molecular representations encoding explicit cyclic information.•New relationship between maximum structural common subgraph and graph edit distance.•Stereoisomerism is encoded in treelet kernel.
论文关键词：Chemoinformatics,Graph kernel,Machine learning
论文评审过程：Received 30 August 2013, Revised 26 July 2014, Accepted 29 July 2014, Available online 14 August 2014.
论文官网地址：https://doi.org/10.1016/j.patcog.2014.07.029
