Identifying temporality of word senses based on minimum cuts

The ability to capture time information is\nessential to many natural language processing and information retrieval applications. Therefore, a lexical resource associating word senses to their temporal orientation might be crucial for the computational tasks aiming at the interpretation of\nlanguage of time in texts. In this paper,\nwe propose a semi-supervised minimum\ncuts strategy that makes use of WordNet\nglosses and semantic relations to supplement WordNet entries with temporal information. Intrinsic and extrinsic evaluations\nshow that our approach outperforms prior\nsemi-supervised non-graph classifiers.