Comparing Two Novel Machine Learning Approaches For Temporal Information Extraction

Temporal Expression in the document is an important structures in any natural language text document. Temporal information processing is challenging research area in recent years. Temporal information processing is a fundamental task to be done for applications like question answering, temporal information retrieval, text summarization, and exploring search results in timeline manner. In this paper author have compared two novel machine learning approaches, Conditional Random Fields and Hidden Markov Model for temporal information extraction. The author have achieved average precision, recall and f-measure 92.5% , 96.19% 94.28% , 89.16%, 91.46%, 90.19%, in CRF and HMM respectively.