Improving the Performance of Gene Mention Recognition System using Reformed Lexicon-based Support Vector Machine

In this paper, we propose a gene mention recog- nition system for biomedical literature using Support Vector Machine based on a reformed lexicon. Then we present an ensemble of rule-based post-processing modules, a integrity check module, a boundary check module, an abbreviation resolution module and a name pruning module, to improve the performance further. The newly developed lexicon is composed of uni-indicating and co-indicating words inside gene mention phrases. With the carefully designed lexicon, the characters of gene mentions can be extracted to support the recognition. Based on this lexicon and post-processing modules, our system can recognize gene mentions in biomedical literature with fairly high accuracy, which can achieve the precision of 85.07%, recall of 83.68% and balanced Ffl=1 score of 84.37.