A probabilistic framework from information extraction models

摘要

Information extraction (IE) is the problem of constructing a knowledge base from a corpus of text documents. In recent years, uncertain data applications have grown in importance in the large number of real-world applications, and IE as an uncertain data source. This paper investigated the uncertain data represent and presented a probabilistic framework from IE model that adapting principles of a state-of-the-art statistical model-semi-Conditional Random Fields (semi-CRFs), which provides a sound probability distribution over extractions.