Temporal Relations Learning with a Bootstrapped Cross-document Classifier
Mirroshandel Seyed Abolghasem, Gholamreza Ghassem-Sani · Frontiers in artificial intelligence and applications · 2010
The ability to accurately classify temporal relation between events is an important task for a large number of natural language processing applications such as Question Answering (QA), Summarization, and Information Extraction. This paper presents a weakly-supervised machine learning approach for classification of temporal relation between events. In the first stage, the algorithm learns a general classifier from an annotated corpus. Then, it applies the hypothesis of “one type of temporal relation per discourse” and expands the scope of “discourse” from a single document to a cluster of topically-related documents. By combining the global information of such a cluster with local decisions of a general classifier, we propose a novel bootstrapping cross-document classifier to extract temporal relations between events. Our experiments show that without any additional annotated data, the accuracy of the proposed algorithm is at least 7% higher than that of the pattern based state of the art system.