Topic analysis to identify communities from Twitter (*)

Potential supervisors: 

Social networks have become one of the most used tools to communicate peoples' thought, opinions and interests. Based on these information, communities detection can be useful in order to understand and analyze a group of citizen. Thus, the main idea of this project is to detect groups of citizens that share common interest and opinion based on a topic analysis of large-scale tweets. This analysis is based on machine learning techniques.

Useful skills:

  • Programming skills, and preferably with some experience with machine learning techniques
  • Some experience of working with text data would be useful.


Vargas-Calderón, V., & Camargo, J. E. (2019). Characterization of citizens using word2vec and latent topic analysis in a large set of tweets. Cities, 92, 187-196.

Monish, P., Kumari, S., & Babu, C. N. (2018, July). Automated Topic Modeling and Sentiment Analysis of Tweets on SparkR. In 2018 9th International Conference on Computing, Communication and Networking Technologies (ICCCNT) (pp. 1-7). IEEE.

Zhang, L., Wu, Z., Bu, Z., Jiang, Y., & Cao, J. (2018). A pattern-based topic detection and analysis system on Chinese tweets. Journal of computational science, 28, 369-381.

Mangal, Nimita, Rajdeep Niyogi, and Alfredo Milani. (2016). "Analysis of users’ interest based on tweets." International Conference on Computational Science and Its Applications. Springer, Cham.

Valverde-Rebaza, Jorge, and Alneu de Andrade Lopes. (2013). "Exploiting behaviors of communities of twitter users for link prediction." Social Network Analysis and Mining 3.4, 1063-1074.

(*) This project is not available during 2021




Date range: 
October, 2019 to October, 2020