A coherence model for sentence ordering

Houda Oufaida (1) , Philippe Blache (1) , Omar Nouali (2)

1 LPL - Laboratoire Parole et Langage
2 CERIST - Centre de recherche sur l'Information Scientifique et Technique

Text generation applications such as machine translation and automatic summarization require an additional post-processing step to enhance readability and coherence of output texts. In this work, we identify a set of coherence features from different levels of discourse analysis. Features have either positive or negative input to the output coherence. We propose a new model that combines these features to produce more coherent summaries for our target application: extractive summariza-tion. The model use a genetic algorithm to search for a better ordering of the extracted sentences to form output summaries. Experimentations on two datasets using an automatic coherence assessment measure show promising results.

Domaines: Intelligence artificielle [cs.AI]; Interface homme-machine [cs.HC]; Traitement du texte et du document; Linguistique; Sciences cognitives

Houda Oufaida, Philippe Blache, Omar Nouali. A coherence model for sentence ordering. NLDB-2019, 2019, Manchester, United Kingdom. ⟨10.1007/978-3-030-23281-8⟩. ⟨hal-02299211⟩