How to formulate an information enquiry for a telephone (robot) operator: an analysis of call centre conversations (Iuliia Avgustis, 13.05.2020)

Telephone robots are now widely used in call centres, as they dramatically reduce human operators' workload. However, callers still often prefer to talk to a human operator, as robots cannot always comprehend the enquiry. Understanding features of enquiries formulated for artificial agents can clarify what callers expect a robot to understand and, therefore, what robots should be able to process. This paper explores how people formulate an information enquiry to a telephone robot used in a major Russian city’s call centre. This call centre answers citizens’ questions considering different administrative issues such as official documents and working hours of public institutions. These enquiry formulations for the robot are then compared with formulations that callers provide for human operators. The study is based on 60 call recordings with human operators and 103 call recordings with the artificial agent. All calls were transcribed and analysed using the method of conversation analysis. The analysis shows that interaction with the robot operator has a particular conversational character, as callers are aware of different competences robots and human operators have. When talking with the robot, callers produce “robot-like” questions and responses themselves to simplify a task for the robot. However, in the talk with human operators, callers tend to provide longer and more complicated enquiries and operators often help them to formulate a concrete question. These findings have important implications for developing telephone robots and other artificial agents.

Iuliia Avgustis has received her M.A. in Sociology at the Moscow School of Social and Economic Sciences (Russia) at a program in collaboration with the University of Manchester (UK) in 2017. She is currently a doctoral researcher in the Smart Communication project at the University of Oulu, Finland, in which she investigates smartphone use among young adults in Russia using ethnomethodology and multimodal conversational analysis. Iuliia is also interested in interaction with artificial agents and in video games.