Intelligent Recommendation System Based On GPS Trajectory Of Taxi

At present, the city taxi because of blind looking for passengers and caused a serious waste of resources of the taxi, increased fuel consumption, which aggravated the environmental pollution, but also reduces the income of taxi drivers. While there are a large number of passengers looking for the taxi everywhere and give your travel inconvenience. This situation is mainly because the taxi drivers and the passenger’s behavior is blind, the lack of unified guidance, which has brought great distress to the passengers and taxi drivers.Currently, regulatory and safety-based application requirements, many domestic cities’ taxi fitted with GPS positioning device, so the taxi trajectory analysis and its applications has become a hot research topic.In this context, using the Xi’an taxi historical trajectory, through data mining and analysis, established a taxi recommendation system for the purpose of improving the taxi response probability and recommending the optimal path. We developed various forms of interactive software, and carried out test. The specific work completed as follows:(1)This paper presents a grid-based passenger behavior clustering algorithm. Through the passengers get off the case analyzed within the sub-period fixed, we extract the location data most likely to carry the passengers, and as a basis for the taxi recommended driving route to the taxi drivers.(2)According to the GPS track data, analysis and extracted the path from taxi driver experience, and established a recommendation algorithm based on the historical data and the current data.(3)The first to use Oracle database and MATLAB software for data computing and processing; and then, using SAE cloud as platform, and Baidu Maps API as interfaces.Were developed terminal software in web,allowing users get data more friendly.