A Real-Time Adaptive Location-based Recommender System Integrating Personalities

The task of selecting an appropriate venue for a diverse group of people can be quite cumbersome as all the individuals in a group have differences in terms of travel time and distance and all these individuals further have their own characteristic preferences when it comes to venues like restaurants, cafes, etc. The majority of the current systems take into account either the user’s personality or trip distance on a non-real-time basis. This work aims to create a real-time adaptive venue recommendation system by integrating three dimensions -user locations, personal preferences, and the venue’s personality. The final solution space will contain a selection of venues as suggestions to the group. These venues will be convenient for all individuals present in the group in terms of travel time, distance to be travelled, and personal preferences. The proposed approach works at three levels. First, an adaptive K-means clustering algorithm is used to analyze individual locations in real time and determine a convenient geographic range around a centroid. Secondly, the Myers-Briggs Type Indicator is used for personality classification, mapping user personalities to venue personalities with an accuracy of 84%. Finally, venues within the designated geographic range are mapped to the group of individuals and suggested as options. The system was tested with sample groups and yielded positive results for venue suggestions. This approach aims to provide a comprehensive and efficient solution for selecting an appropriate venue for a diverse group of individuals by considering their locations, personal preferences, and venue personalities in real time.