Automatic Lexico-Semantic Frames Acquisition from Syntactic Parsed Tree by Using Clustering and Combining Techniques

摘要

This article describes an unsupervised strategy to acquire lexico-semantic frames (LSFs) of verbs from sentential parsed corpora (in syntactic level). LSF is a crucial linguistic resource presents a set of semantic elements for exhibiting a meaning of lexeme. The problems of acquiring LSFs consist of verb senses ambiguity, diversity of linguistic usages, and lack of completed elements in a sentence. We propose an specific clustering and combining technique to acquire frame for each verb sense and specify constraints to each frame’s slots. Our proposed clustering technique is based on the Minimum Description Length (MDL) principle and using information encoded in features of element instead of its frequency from the corpora.