On a Topic Model for Sentences - Archive ouverte HAL Access content directly
Conference Papers Year :

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.
No file

Dates and versions

hal-01391903 , version 1 (04-11-2016)

Identifiers

  • HAL Id : hal-01391903 , version 1

Cite

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⟩
295 View
0 Download

Share

Gmail Facebook Twitter LinkedIn More