Uncertainty Estimation for Twitter Inference
Meng Hsiu Tsai, Nicole Marie Ely, Yingfeng Wang · 2021 International Conference on Computational Science and Computational Intelligence (CSCI) · 2021
Twitter provides a platform for users to express their opinions in the form of Twitter messages called tweets. Analyzing tweets from a specific city or demographical group requires geographical or demographical information. Unfortunately, most Twitter users do not provide these details in their profiles. To overcome this challenge, tools have been developed for inferring users' geographical and demographical information from Twitter data. Using inference results is risky due to the lack of uncertainty estimation of these results. Here, we present a framework to estimate uncertainties of Twitter inference results. The effectiveness of this framework is verified in experiments.