Big data analytics for smart transportation (2)
Chung, Maybelle Huishi
Date of Issue2017-04-26
School of Computer Science and Engineering
Gathering transport data has been made easier with the cutting-edge technologies as more GPS enabled devices are able to log travel data. The amount of data is growing significantly in the coming years. It is important to make use of these data to discover the hidden insights and patterns. Singapore has adopted the use of data analysis in the approach of developing a Smart Nation City. Through data analytics on travel data, new-found knowledge of travel patterns could help Singapore in future decision-making process, leading to a more efficient and effective transport network. The aim of this project is to analyze taxi data and gain some insights on the travel patterns of taxi passengers in Singapore. This could aid taxi operators in taxi resource allocation. This project is broken into 3 major phases, Data Cleaning and Preparation, Data Mining, and Data post-processing. Data cleaning involves the tasks of removing outliers and unwanted data. Data preparation involves the tasks of retrieving taxi trips from the raw dataset. By applying the DBSCAN clustering algorithm on the data, major clusters of the taxi trips were retrieved. The results from the clustering was visualized on a map of Singapore and the travel patterns were identified. Data post-processing was conducted by getting the boundaries of the regions in Singapore. All pickups from each region was analyzed to derive significant travel patterns. The results showed that the most common pickup and drop-off locations are around the Central/Downtown areas in Singapore. There are little pickups at the North-East region of Singapore. The travel pattern of passengers who boarded the taxi at the East region would usually drop-off at Woodlands. During the non-peak hours of the day, the pickups from the west region would usually have drop-offs at the North region, while the pickups at the north regions had drop-offs to all regions of Singapore.
Final Year Project (FYP)
Nanyang Technological University