Exploring the Characteristics of Multi-Party Dialogues

This paper describes novel results on the characteristics of three-party dialogues by quantitatively comparing them with those of two-party. In previous dialogue research, two-party dialogues are mainly focussed because data collection of multi-party dialogues is difficult and there are very few theories handling them, although research on multi-party dialogues is expected to be of much use in building computer supported collaborative work environments and computer assisted instruction systems. In this paper, firstly we describe our data collection method of multi-party dialogues using a meeting scheduling task, which enables us to compare three-party dialogues with those of two party. Then we quantitively compare these two kinds of dialogues such as the number of characters and turns and patterns of information exchanges. Lastly we show that patterns of information exchanges in speaker alternation and initiative-taking can be used to characterise three-party dialogues. 1 I n t r o d u c t i o n Previous research on dialogue has been mostly focussing on two-party human-human dialogue for developing practical human-computer dialogue systems. However, our everyday communicative activities involves not only two-party communicative situations but also those of more than two-party (we call this multi-party). For example, it is not unusual for us to chitchat with more than one friend, or business meetings are usually held among more than two participants. Recently advances of computer and networking technologies enable us to examine the possibility of using computers to assist effective communication in business meetings. As well as this line of computer assisted communication research, autonomous programs called 'agents', which enable users to effectively use computers for solving problems, have been extensively studied. In this research trend, 'agent' is supposed to be distributed among computers, and how they cooperate for problem solving is one of the most important research topics. Previous studies on two party dialogue can be of some use to the above important computer related communication research, but research on multi-party interaction can contribute more directly to the advances of the above research. Furthermore, research on multi-party dialogue is expected to make us understand the nature of human communication in combination with the previous and ongoing research on two-party dialogue. The purpose of this paper is to quantitively show the characteristics of multi-party dialogues in comparison with those of two-party using actual dialogue data. In exploring the characteristics of multi-party dialogues to those of two-party, we will concentrate on the following problems. W h a t pat terns of informat ion exchanges do conversat ional participants fo rm? When abstracting the types of speech acts, in two-party dialogues, the pattern of information exchanges is that the first and second speakers alternately contribute (A-B-A-B . . . ) . But in multiparty dialogues, for example, in three-party dialogues, dialogue does not seem to proceed as A-B-C-A-B-C . . . , since this pattern seems to be too inefficient if B tells C what B are told by A, which C will be told the same content twice, and too efficient and strict if A, B and C always initiate new topics in this order, in which they have no