Expert finding is hot topic discussed in co-author networks. Traditional language models only compares query terms with the documents of candidate for expert finding and ignores venues (conferences or journals) in which the paper is published, hi this paper we propose novel influence language models which consider the importance of venues in which the papers of candidates are published. If the paper is published in a high level venue and another is published in a low level venue then these two papers should not have same weight-age for finding experts. The paper which is published in high level venue is more valuable than the paper which is published in low level venue. Experimental results show that our proposed models outperform the existing models.
展开▼