Sequence matching with subsequence analysis

摘要

This article describes an alternative approach for matching user text input in natural language processing against an existing knowledge base, consisting of semantically described words or phrases. Most common methods of natural language processing are overviewed and their main problems are outlined. A sequence matching algorithm is introduced, which deals with some of these problems. First the longest subsequences discovery algorithm is explained. Then the major components of the similarity measure are defined and the computation of concurrence and dispersion measure is presented. Results of the algorithms performance on a test set are then shown. The work is concluded with some ideas for the future and some examples where our approach can be practically used.