2002. Topic identification in natural language dialogues using neural networks.pdf

Topic Identification In Natural Language Dialogues Using Neural Networks Krista Lagus and Jukka Kuusisto Neural Networks Research Centre, Helsinki University of Technology P.O.Box 9800, FIN-02015 HUT, Finland krista.lagus@hut.fi Abstract In human–computer interaction sys- tems using natural language, the recognition of the topic from user’s utterances is an important task. We examine two different perspectives to the problem of topic analysis needed for carrying out a success- ful dialogue. First, we apply self- organized document maps for mod- eling the broader subject of dis- course based on the occurrence of content words in the dialogue con- text. On a Finnish corpus of 57 dialogues the method is shown to work well for recognizing subjects of longer dialogue segments, whereas for individual utterances the sub- ject recognition history should per- haps be taken into account. Sec- ond, we attempt to identify topically relevant words in the utterances and thus locate the old information (’topic words’) and new information (’focus words’). For this we define a probabilistic model and compare dif- ferent methods for model parameter estimation on a corpus of 189 dia- logues. Moreover, the utilization of information regarding the position of the word in the utterance is found to improve the results. 1 Introduction The analysis of the topic of a sentence or a document is an important task for many nat- ural language applications. For example, in interactive dialogue systems that attempt to carry out and answer requests made by cus- tomers, the response strategy employed may depend on the topic of the request (Jokinen et al., 2002). In large vocabulary speech recog- nition knowledge of the topic can, in general, be utilized for adjusting the language model used (see, e.g., (Iyer and Ostendorf, 1999)). We describe two approaches to analyzing the topical information, namely the use of topically ordered document maps for analyz- ing the overall topic of dialogue segments, a