Information Interactions in Outcome Prediction, Quantification and Interpretation using Stochastic Block Models
Pieces of information interact with each other, on Twitter, Reddit, Spotify, etc. When a user of these platforms has to make a choice, s/he will be influenced by this interaction. On this picture, you see which groups of information interact strongly with each other. The darker the color, the more an exposed user's decision is likely to change due to information interaction.

Information Interactions in Outcome Prediction, Quantification and Interpretation using Stochastic Block Models

2021, Jul 06    

Link to the paper

In most real-world applications, it is seldom the case that a result appears independently from an environment. In social networks, users’ behavior results from the people they interact with, news in their feed, or trending topics. In natural language, the meaning of phrases emerges from the combination of words. In general medicine, a diagnosis is established on the basis of the interaction of symptoms. Here, we propose the Interacting Mixed Membership Stochastic Block Model (IMMSBM), which investigates the role of interactions between entities (hashtags, words, memes, etc.) and quantifies their importance within the aforementioned corpora. We find that in inference tasks, taking them into account leads to average relative changes with respect to non-interacting models of up to 150\% in the probability of an outcome and greatly improves the predictions performances. Furthermore, their role greatly improves the predictive power of the model. Our findings suggest that neglecting interactions when modeling real-world phenomena might lead to incorrect conclusions being drawn.

Reference:

Information Interactions in Outcome Prediction, Quantification and Interpretation using Stochastic Block Models
G. Poux-Médard, J. Velcin, S. Loudcher, RecSys, 2021

DOI: 10.1145/3460231.3474254