Community Member Retrieval on Social Media using Textual Information
This addresses community detection on social media for users or platforms, but it is incremental as it builds on existing unsupervised methods with a novel proxy task.
The paper tackled community membership detection using only text features with few positive examples by introducing an unsupervised proxy task called user re-identification for learning user embeddings. Experiments on 16 communities showed these embeddings were more effective for identification than common unsupervised representations.
This paper addresses the problem of community membership detection using only text features in a scenario where a small number of positive labeled examples defines the community. The solution introduces an unsupervised proxy task for learning user embeddings: user re-identification. Experiments with 16 different communities show that the resulting embeddings are more effective for community membership identification than common unsupervised representations.