Sense Disambiguation in PubMed Abstracts Using Text and Data Mining

摘要

Recently, as the size of genetic knowledge grows faster, automated classification of documents in PubMed database has become a critical issue in bio-informatics area. In order to make bio database useful to end users, it should be well organized not only in structure but also semantically, and it all starts with a classification task. Traditionally, classification of texts is done either statistically or using NLP (Natural Language Processing) techniques. Statistical approaches are efficient and fastbut usually lack deep understanding, and hence prone to ambiguity errors. Knowledge based NLP techniques, however, are very slow even though the quality of the results is usually better than that of statistical approaches. In this paper, we propose a new approach based on text mining technique.