Hindi Compound Verbs and their Automatic Extraction

We analyse Hindi complex predicates and propose linguistic tests for their detection. This analysis enables us to identify a category of V+V complex predicates called lexical compound verbs (LCpdVs) which need to be stored in the dictionary. Based on the linguistic analysis, a simple automatic method has been devised for extracting LCpdVs from corpora. We achieve an accuracy of around 98% in this task. The LCpdVs thus extracted may be used to automatically augment lexical resources like wordnets, an otherwise time consuming and labourintensive process