Theme Of Model-based Expert Retrieval And Mining

In 21st century, as the information society going into a knowledge society, the talents who are mastery of knowledge and creative become the most valuable resources of enterprises and organizations. Whether scientific research institutions, or social production sectors (such as enterprise, factory), all need domain experts who have rich professional knowledge, skills and experiences to organize the team, guide the research and development and thus improve the work or production efficiency. But how to find these experts by effective means increasingly becomes an important scientific question.As a retrieval task, expert finding has recently attracted much attention and becomes a hot research in the field of information processing. This article first reviews the research background and present situation of expert finding, describes its research advance and further, makes detailed analysis and summarization on its research methods such as expert modeling, link analysis, query expansion, expert evidence identification, hidden theme analysis. And also the statistics of various types of test corpus are made and the future researches in this field are given.And various methods have been proposed to rank expert candidates against topic query. The most efficient approach is document-based method. However, such kind of method lacks of the capability to capture the hidden semantic association between queries and candidates. In this paper, a hidden topic analysis based approach is proposed. It models query and supporting document as a word-topic-document association instead of the word-document association in language model. In addition, the prior knowledge of supporting document is considered to favor expert ranking. The empirical results on metadata corpus have demonstrated the model can effectively catch the semantic association between queries and candidates, thus improves the performance of expert finding. This paper also presents an assessment of automatic expert allocation strategies and ranking of experts under the condition of multi-index to achieve the evaluation of experts recommend automatically and the allocation of candidate experts fairly and reliably.There are more and more researches about the application of the hidden topic model. And the application of the model in the field of bioinformatics is a perfect example in this paper. Bioinformatics has grown about 30 years. Especially in the past 10 years, the field developed in leaps and bounds and emerged many research works. Whether as a novice, or a famous scholar in this field, would like to be able to glimpse the research situation of this field, and get an intuitive and quantified understanding. This article aims to use the hidden topic model to mine the literatures in the field of bioinformatics and discover the important research topics, to quantify the evolution of these themes to show the trend.