AN EFFECTIVE TAXI RECOMMENDER SYSTEM |
Author(s): |
Siddhesh R. Bhagat , MMCOE, Pune University; Sagar K. Golhar, MMCOE, Pune University; Sahil S. Bagde, MMCOE, Pune University; Rohit Gavade, MMCOE, Pune University; Prof. Shubhada Mone, MMCOE, Pune University |
Keywords: |
Recommendation System, Android application, GPS, Temporal, Spatial, K-means, Data Mining |
Abstract |
In this research, we are creating an android-web based taxi booking application. Our system is a real time android application which not only allows the passenger to book a taxi on demand, but its uniqueness lies in the point that it also helps the taxi driver to earn more profit by mining the historical GPS trajectories. Our system helps the taxi drivers in finding the next cruising location where they can find passengers yielding maximum profits. The different factors considered in the system are distance between the current location and the recommended location, waiting time to get the next passenger, expected fare for the trip and most importantly the historical data of a location for providing the recommendations. We are using the historical GPS data to analyze and retrieve the habitation of residents in area. Also we are using clustering analysis and spatial-temporal analysis to make groups of locations based on the mentioned factors. Temporal and spatial issues are important factors which the taxi drivers often consider in the real world for determining passenger distribution. A location-to-location graph model, referred to as pick-drop location and the next passenger pick-drop location is used in the system. This model is also used to calculate the trip fare from the current location and the destination location. |
Other Details |
Paper ID: IJSRDV3I2096 Published in: Volume : 3, Issue : 2 Publication Date: 01/05/2015 Page(s): 135-137 |
Article Preview |
|
|