Institute of Transportation, MOTC

        The purpose of this study is to estimate significant locations such as home and workplace. Due to the distribution of these significant location, roughly defines the spatial structure and activity patterns of urban area. The movement pattern and significant location of residents exhibit pretty much regularity in terms of sequence and time. Estimated significant location information is useful and important in many applications, including urban planning, transportation and other related research. There are three main contributions of this study

1.  This study designed Bearing Difference Method to optimize phenomenon of signal oscillation.
2. This study designs and improves a series of processes for estimating the user’s home and workplace to get more accurate.
3. The model with the best predictive performance is extracted through the experimental design to facilitate the estimation of the user's home and workplace.

       According to the actual test and model verification results of the mobile phone data, it is indicated that the Bearing Difference Method can effectively improve the accuracy of the predicted location. In addition, using the model designed in this study is the most accurate. Among the 27 users, the average error between the estimated and the actual home location is 148 meters. and the average error between the estimated and actual workplace is 269 meters. It shows that the results of this study are suitable for application in the field of transportation.