Predicting the Priority of Social Situations for Personal Assistant Agents

Abstract

Personal assistant agents have been developed to help people in their daily lives with tasks such as agenda management. In order to provide better support, they should not only model the user's internal aspects, but also their social situation. Current research on social context tackles this by modelling the social aspects of a situation from an objective perspective. In our approach, we model these social aspects of the situation from the user's subjective perspective. We do so by using concepts from social science, and in turn apply machine learning techniques to predict the priority that the user would assign to these situations. Furthermore, we show that using these techniques allows agents to determine which features influenced these predictions. Results based on a crowd-sourcing user study suggest that our proposed model would enable personal assistant agents to differentiate between situations with high and low priority. We believe this to be a first step towards agents that better understand the user's social situation, and adapt their support accordingly.

Who’s That? - Social Situation Awareness for Behaviour Support Agents

“…The studies described in the previous section provide evidence indicating that the proposed three-level social situation awareness component is implementable in practice [Kola et al, 2021[Kola et al, , 2020b. Their results are based on user studies revolving around the use case of a socially aware personal assistant agent for calendar management.…”

Using Psychological Characteristics of Situations for Social Situation Comprehension in Support Agents

“…Another approach on how to take into account the effects of social situations on user behavior is proposed by Kola et al [2020b]. They model social situations through a set of social situation features seen from the point of view of the user.…”

“…This information can come via sensory input or interaction with the user. Kola et al [2019Kola et al [ , 2020b propose a set of features based on research from social sciences. They divide features into situation cues, namely setting, event frequency, initiator, help dynamic, and social background features describing the social relation between the user and other people in the social situation, namely role, hierarchy level, contact frequency, geographical distance, years known, relationship quality, depth of acquaintance, formality level and shared interests.…”

“…Existing approaches (e.g., [Kola et al, 2020b, Dignum and Dignum, 2014, Ajmeri et al, 2017) tackle this challenge by using some type of situation cues as input (e.g., actors, relationship characteristics, etc. ), and using information processing techniques such as machine learning or rule-based approaches to assess expected behavior.…”

“…As in Kola et al [2019Kola et al [ , 2020b, our use case is a calendar management agent for making automated decisions or suggestions to users about how to resolve conflicts in their schedule. We develop computational models that would allow such an agent to assess the priority (Level 3 information) of conflicting meetings, the idea being that the user would attend the one with the highest priority.…”