Discovering Common Pathways Across Users¿ Habits in Mobility Data

• Type: Article in International Conference Proceedings Book
• Date: 2019

Authors

Thiago Andrade Silva (Author) Other; Cancela, B (Author) Other; João Gama (Author) FEP

Conference proceedings International

Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Pages: 410-421

19th EPIA Conference on Artificial Intelligence, EPIA 2019

3 September 2019 through 6 September 2019

Indexing

Scopus - 2 Citations

Other information

Authenticus ID: P-00R-4JZ

DOI: 10.1007/978-3-030-30244-3_34

Abstract (EN): Different activities are performed by people during the day and many aspects of life are associated with places of human mobility patterns. Among those activities, there are some that are recurrent and demand displacement of the individual between regular places like going to work, going to school, going back home from wherever the individual is located. To accomplish these recurrent daily activities, people tend to follow regular paths with similar temporal and spatial characteristics. In this paper, we propose a method for discovering common pathways across users¿ habits. By using density-based clustering algorithms, we detect the users¿ most preferable locations and apply a Gaussian Mixture Model (GMM) over these locations to automatically separate the trajectories that follow patterns of days and hours, in order to discover the representations of individual¿s habits. Over the set of users¿ habits, we search for the trajectories that are more common among them by using the Longest Common Sub-sequence (LCSS) algorithm considering the distance that pairs of users travel on the same path. To evaluate the proposed method we use a real-world GPS dataset. The results show that the method is able to find common routes between users that have similar habits paving the way for future recommendation, prediction and carpooling research techniques. © 2019, Springer Nature Switzerland AG.

Language: English

Type (Professor's evaluation): Scientific