We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.IR

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Computer Science > Information Retrieval

Title: Improved Topic modeling in Twitter through Community Pooling

Abstract: Social networks play a fundamental role in propagation of information and news. Characterizing the content of the messages becomes vital for different tasks, like breaking news detection, personalized message recommendation, fake users detection, information flow characterization and others. However, Twitter posts are short and often less coherent than other text documents, which makes it challenging to apply text mining algorithms to these datasets efficiently. Tweet-pooling (aggregating tweets into longer documents) has been shown to improve automatic topic decomposition, but the performance achieved in this task varies depending on the pooling method.
In this paper, we propose a new pooling scheme for topic modeling in Twitter, which groups tweets whose authors belong to the same community (group of users who mainly interact with each other but not with other groups) on a user interaction graph. We present a complete evaluation of this methodology, state of the art schemes and previous pooling models in terms of the cluster quality, document retrieval tasks performance and supervised machine learning classification score. Results show that our Community polling method outperformed other methods on the majority of metrics in two heterogeneous datasets, while also reducing the running time. This is useful when dealing with big amounts of noisy and short user-generated social media texts. Overall, our findings contribute to an improved methodology for identifying the latent topics in a Twitter dataset, without the need of modifying the basic machinery of a topic decomposition model.
Subjects: Information Retrieval (cs.IR); Machine Learning (cs.LG)
Cite as: arXiv:2201.00690 [cs.IR]
  (or arXiv:2201.00690v1 [cs.IR] for this version)

Submission history

From: Federico Albanese [view email]
[v1] Mon, 20 Dec 2021 17:05:32 GMT (10kb)

Link back to: arXiv, form interface, contact.