Skip to Main content Skip to Navigation
Conference papers

On a Topic Model for Sentences

Abstract : Probabilistic topic models are generative models that describe the content of documents by discovering the latent topics underlying them. However, the structure of the textual input, and for instance the grouping of words in coherent text spans such as sentences, contains much information which is generally lost with these models. In this paper, we propose sentenceLDA, an extension of LDA whose goal is to overcome this limitation by incorporating the structure of the text in the generative and inference processes. We illustrate the advantages of sentenceLDA by comparing it with LDA using both intrinsic (perplexity) and extrinsic (text classification) evaluation tasks on different text collections.
Complete list of metadata
Contributor : Massih-Reza Amini Connect in order to contact the contributor
Submitted on : Friday, November 4, 2016 - 12:34:04 AM
Last modification on : Wednesday, November 3, 2021 - 6:46:39 AM


  • HAL Id : hal-01391903, version 1


Georgios Balikas, Massih-Reza Amini, Marianne Clausel. On a Topic Model for Sentences. SIGIR'16 ACM SIGIR conference on Research and Development in Information Retrieval, Jul 2016, Pise, Italy. ⟨hal-01391903⟩



Record views