Pro-Active Systems and Influenceable Users: Simulating Pro-Activity in Task-oriented Dialogues

We investigate proactivity, the capacity of a dialogue system to provide relevant information even when not explicitly requested, in the context of task-oriented dialogues. We propose to extend the current task-oriented framework, and have investigated four aspects of proactivity: (i) the degree of proactivity provided by the system during a dialogue; (ii) the propensity of the user to be influenced by the system proactivity; (iii) the complexity of the domain ontology; (iv) the relation between user needs and application domain, in terms of expected failure situations. Under the hypothesis that proactivity helps to increase effectiveness and efficiency of dialogues, we set up a framework based on dialogue simulations, and experimented the four aspects mentioned above. Although the current implementation allows to simulate a limited amount of dialogue phenomena (e.g., system initiative only), we are able to show that proactivity might have strong effects on dialogues, reducing up to 60% of dialogue turns in an application domain of medium complexity.

Introduction. Proactivity is a fundamental characteristic of human-human collaborative dialogues, consisting in the attitude of the speakers to provide information that can be used to achieve the goal of the dialogue, even when such information is not explicitly requested. This attitude obeys to the so called principles of cooperative dialogue, which have been summarized in the popular Grices’s maxims (Grice, 1975). Among the others, proactivity is very common in instruction-giving dialogues. For instance, assuming the following dialogue between speakers A and B, we say that It arrives at 1245 is a proactive response, because information given by B is more than was asked for by A, and because B guesses that this is the sort of information A might also need, and so offers it unsolicited. The major effect of proactivity is that, by anticipating user needs (e.g., a question like What time this train will arrive in London? in our example), it avoids long dialogues, making communication more efficient. Proactive behaviours share some aspects with other communicative devices.

Discussion / Conclusion. Which are the features that affect a proactive behaviour in task-oriented dialogues? The aspects proposed in Section 4 proactive unit, system proactivity rate, user influenceability and complexity of the application domain are all relevant for describing proactive dialogues. Particularly, the notion of proactive unit has been shown to be essential to make operative our simulation of proactivity. How can we model such features within the architecture of current dialogue systems? While the current modeling of task-oriented dialogues provides a general framework, specific extensions were necessary to model our intuitions about the attitude of the user (i.e., influenceability) and the role of the system (i.e., proactivity). However, simulating proactivity revealed itself as more complex than initially expected, and, due to the issues described in Section 5.6, we had to adopt quite simplistic solutions. How can we show the impact of proactivity on task-oriented dialogues, in terms of effectiveness and efficiency of the dialogues?