A machine learning approach to recognizing acronyms and their expansion

摘要

The paper addresses the problem of automatically recognizing acronyms and their corresponding expansions in free format text. To deal with the problem, we propose a machine learning approach. First, all likely acronyms are identified from text. Second, candidate expansions against likely acronyms are generated from their surrounding text. Last, we employ support vector machines (SVM) to select the genuine expansions for acronyms. Experimental results show that our approach outperforms baseline method of using patterns. Experimental results also show that the trained SVM model is generic and performs well on different domains.