POI (Point Of Interest) Name Recognition And The Application In Dialog System

In the growing popularity of smart terminals technical background, living information service industry has developed rapidly. As the basis of living information service, POI (Point of Interest) is applied more widely, such as location search services, geographic information search. From the perspective of natural language processing, it is necessary to identify POI name effectively.POI name consists of two parts:the location names and organization names, both of which are part of named entity. However, not all organization are the POI, we should extract the organization name with a geographic entity to filter organization names. So we can do further research and work for POI name recognition in the basis of location and organization names recognition.Location name recognition and organization name recognition is an important part of named entity recognition, many domestic and foreign scholars and experts conducted in-depth research on the entity name recognition. The current entity name recognition methods are:rule-based approach, statistical-based methods, rule and statistical combination method. In the statistical-based methods, there’s common approach such as hidden Markov model (HMM), Conditional Random Field Model (CRF) and Maximum Entropy Markov model (MEMM) and so on.This paper firstly do the work of location names and organization names recognition, then made improvements for the characteristics of POI names, and our work is successfully applied in a practical dialogue system for navigation, and achieved good results.The main contribution of this paper is as follows:1、Propose the method of POI name recognition to identify POI name, and design the system into two steps;2、Apply the module of POI name recognition in dialogue systems.