A Topic Extraction Method on the Flow of Conversation in Meetings

Abstract

In this paper, we present a new topic extraction method for meetings according to the flow of conversation. Our method extracts appropriate topic words according to their importance in the conversation in a time series using text data taken from meetings. Since meetings take up a great deal of time, one of the most important issues for organizations and companies is to improve meeting efficiency. Therefore, we should analyze the contents of the meetings, but in order to do that, it is important to be able to automatically extract the most important topics made during each meeting. The changes in the importance of a topic can be seen in a time series, so it is necessary to utilize topic extraction according to its importance in time series variation during a meeting. We can then find the topic words in a meeting according to their importance in the time series variation by using our method.