Calendar agent with commonsense reasoning

In this work, we describe a Calendar Agent that uses Commonsense reasoning to generate sub-tasks based on the events that are entered by the user in the calendar. In order to generate the relevant sub-tasks of an event, we utilize ConceptNet which has 250,000 elements of commonsense knowledge. Our calendar agent recommends the best time slot for a given event based on the profile and the behavior of the user. This calendar agent was developed based on Prometheus Methodology using Prometheus Design Tool. Among the features of the calendar agent is its ability to solve timing conflict, filling in the event entry automatically based on the given information and generating sub-tasks that the user has to complete prior to the event. The calendar agent also tries to learn the behavior of the users to provide better recommendation.