Project Description

Many recommendation systems have incorporated the location context as part of the preferences, although none of the existing systems considers public transit as a mode of transport. While mass transit enables effective, safe, and cheap commuting in large metropolitan areas, one of the consequences of using it is that there are only fixed places of boarding/exiting and one may need to walk to a particular location from a given station. Given the impact of weather on the mood and activities, people may decide to have the preferences for a certain type of services dynamically adjusted based on the current weather, or near-future weather forecast, and the distances to preferred locations.

In this work, we expand the functionality of Yelp-like recommendation sites by enabling users to search for places bounded by more-intuitive travel-time instead of distance. We also allow users to choose places which are accessed via public transportation, and we put the weather context into the recommendation process by letting users define event models to control the extent of walking in user-defined wea- ther conditions. We develop a web application (both desktop and mobile environments) for this purpose with user-friendly interface, and utilize existing tools such as Google Maps Direction API and OpenWeatherMap API for retrieving real-time information in the backend.