Topic Shift Detection for Mixed Initiative Response

Conversational systems have become a part and parcel of our everyday life and virtual assistants like Amazon’s Alexa1, Google Home2 or Apple’s Siri 3 are soon becoming conventional household items (Terzopoulos and Satratzemi, 2020). Most of the conversational systems were built with the primary goal of accessing information, completing tasks, or executing transactions. However, recent conversational agents are transitioning towards a novel hybrid of both task-oriented and a non-task-oriented systems (Akasaki and Kaji, 2017) from the earlier models that resembled factual information systems (Leuski et al., 2006). But with this transition, they are failing to engage in complex information seeking tasks and conversations where multiple turns tend to get involved (Trippas et al., 2020). These new-age open-domain dialogue systems also suffer from a different kind of user behaviour called “anomalous state of knowledge” (Belkin and Vickery, 1985) where the user has vague information requirements and is often unable to articulate it with enough precision. This leads to the user deviating from their original path and traversing into a sub-topic without their knowledge (Larsson, 2017). Thus, we need a context-dependent user guidance without presupposing a strict hierarchy of plans and task goals of the user. Such a guidance, without topic information provided beforehand, is a difficult task to achieve in an open-domain system.

We create a novel model which can, with a precision of 84%, predict the utterances that belong to the major topic and those which are deviating from the same, without a predetermined topic set.