The prevalence of mobile devices to individuals enhances the use of personal location data for several services. However, service providers do not necessarily have a sufficient amount of data for users. In such cases, a possible solution is integrating location data from other resources into existing records by correctly re-identify each of new trajec- tories as existing user’s one. This work structures a graph-based user re-identification method for location data. For each user, the method creates mobility features rep- resenting the frequency of visits to locations and the direction of the movement from one place to another. The similarity of location trajectories is scaled with the features. Experimental results show that our proposed method correctly matches the same user’s trajectories even if they are recorded in different time periods (e.g. before the COVID- 19 outbreak, during the lockdown caused by the pandemic). The matching accuracy is over 99 %. Attackers may abuse the user re-identification method to increase the amount of location data about users. Therefore, the project constructs a clustering- based location privacy protection mechanism (LPPM) as a countermeasure. Through experiments on actual GPS data, we ensure that the clustering-based LPPM surely reduces the matching performance of the graph-based method. Also, the project indi- cates a way to balance the data utility and privacy level out. In this project, we utilize the location data obtained through a field experiment. The data collection started in September 2019, and the data includes records obtained in the lockdown period induced by the COVID-19. This project depicts frequent daily mobility patterns found in the dataset. We show that over 88 % of daily movements can be delineated with only ten unique patterns. By analyzing the changes in the daily movement patterns, we reveal Swiss people reduced the number of daily destinations during the lockdown period.