Linear Classifiers for Context-aware Place Suggestions Implemented on Google Map

Abstract

Mobile applications such as Google Maps can provide suggestions for nearby locations. However, some issues with personalized presentation and recommendations and suggested locations are not ordered. This paper proposes context-awareness on place types using linear classifiers. The context-aware ubiquitous support is concerned with recommending nearby locations based on rating and distance. We use the scenario of types of places to process the recommendation. Two experiment studies were conducted, and the results showed that our approach is significantly better than a normal Google places search. Overall, the users were satisfied with our approach.