A LDA-Based Topic Classification Approach from Highly Imperfect Automatic Transcriptions - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

A LDA-Based Topic Classification Approach from Highly Imperfect Automatic Transcriptions

Mohamed Morchid
Richard Dufour
Georges Linarès

Résumé

Although the current transcription systems could achieve high recognition performance, they still have a lot of difficulties to transcribe speech in very noisy environments. The transcription quality has a direct impact on classification tasks using text features. In this paper, we propose to identify themes of telephone conversation services with the classical Term Frequency-Inverse Document Frequency using Gini purity criteria (TF-IDF-Gini) method and with a Latent Dirichlet Allocation (LDA) approach. These approaches are coupled with a Support Vector Machine (SVM) classification to resolve theme identification problem. Results show the effectiveness of the proposed LDA-based method compared to the classical TF-IDF-Gini approach in the context of highly imperfect automatic transcriptions. Finally , we discuss the impact of discriminative and non-discriminative words extracted by both methods in terms of transcription accuracy.
Fichier non déposé

Dates et versions

hal-01319771 , version 1 (23-05-2016)

Identifiants

  • HAL Id : hal-01319771 , version 1

Citer

Mohamed Morchid, Richard Dufour, Georges Linarès. A LDA-Based Topic Classification Approach from Highly Imperfect Automatic Transcriptions. LREC, May 2014, Reykjavik, Iceland. ⟨hal-01319771⟩

Collections

UNIV-AVIGNON LIA
52 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More