User Context and Community Modelling using Proximity Signals and Cyber, Physical, Social Behaviours

Nov 23, 2021

The modeling of social and mobility networks continues to gain importance in a variety of fields ranging from epidemiology social group and community detection [1, 2], to user movement and behavior understanding [3, 4, 5].

In the Simultaneous Extraction of Context and Community (SECC) model that is proposed by Nguyen et al. [1, 2], each context is represented as a multinomial distribution, which indicates the participating level of the users to such context. To detect communities, it then computes the clusters of multinomial distribution to discover proximity contexts and users.

In our project, in collaboration with Scentre Group (Westfield malls), we analysed visitor behaviour in shopping malls, with joint learning of semantic behaviours from cyber, physical, and social behaviours [4,5], for contextual recommendations with tripartite graph models [6].

[1] Nguyen, T., Nguyen, V., Salim, F. D., & Phung, D. (2016, March). SECC: Simultaneous extraction of context and community from pervasive signals. In 2016 IEEE International Conference on Pervasive Computing and Communications (PerCom) (pp. 1-9). IEEE.

[2] Nguyen, T., Nguyen, V., Salim, F. D., Le, D. V., & Phung, D. (2017). A Simultaneous Extraction of Context and Community from pervasive signals using nested Dirichlet process. Pervasive and Mobile Computing, 38, 396-417.

[3] Ren, Y., Tomko, M., Salim, F. D., Chan, J., & Sanderson, M. (2018). Understanding the predictability of user demographics from cyber-physical-social behaviours in indoor retail spaces. EPJ Data Science, 7, 1-21.

[4] Manpreet Kaur, Flora D. Salim, Yongli Ren, Jeffrey Chan, Martin Tomko, and Mark Sanderson. 2020. Joint Modelling of Cyber Activities and Physical Context to Improve Prediction of Visitor Behaviors. ACM Trans. Sen. Netw. 16, 3, Article 28 (August 2020), 25 pages. DOI:https://doi.org/10.1145/3393692

[5] Manpreet Kaur, Flora D. Salim, Yongli Ren, Jeffrey Chan, Martin Tomko, and Mark Sanderson. 2018. Shopping intent recognition and location prediction from cyber-physical activities via wi-fi logs. In Proceedings of the 5th Conference on Systems for Built Environments (BuildSys ‘18). Association for Computing Machinery, New York, NY, USA, 130–139. DOI:https://doi.org/10.1145/3276774.3276786

[6] Y. Ren, M. Tomko, F. D. Salim, J. Chan, C. L. A. Clarke and M. Sanderson, “A Location-Query-Browse Graph for Contextual Recommendation,” in IEEE Transactions on Knowledge and Data Engineering, vol. 30, no. 2, pp. 204-218, 1 Feb. 2018, doi: 10.1109/TKDE.2017.2766059. https://ieeexplore.ieee.org/document/8081816


Main Participants

Related Projects

Smart Parking for High Demand Areas

Show More

Parking Availability Prediction

Show More

© 2021 Flora Salim - CRUISE Research Group.