Sequence Package Analysis: A New Natural Language Understanding Method for Performing Data Mining of Help-Line Calls and Doctor-Patient Interviews

摘要

Designers of audio mining programs must confront the complexities of natural language dialog, which is replete with ambiguities, circumlocutions and ellipses. Speakers often make requests, lodge complaints, or report on problems in such roundabout ways that attempts to find a statistically probable word match between the application vocabulary and the user's speech can yield unsatisfactory results. One solution to this problem is the application of Sequence Package Analysis (SPA) to perform intelligent data mining. The purpose of this paper is to demonstrate how Sequence Package Analysis, as a new AI-based natural language understanding method, can be used to gather important business intelligence by learning what customers are really complaining about when they make contact with customer call centers. The paper also looks at how SPA can be used in mining tape recordings of doctorpatient interviews to uncover important medical history data, often buried in the ambiguity of patient dialog. The paper, however, not only looks at the applications of SPA (e.g., industrial and medical), but demonstrates that SPA, as a new data mining technique, is an innovative heuristic approach to data mining that can bring machines one step closer to understanding humans.