Automatic detection of parallel sentences ...
Document type :
Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès avec actes
Title :
Automatic detection of parallel sentences from comparable biomedical texts
Author(s) :
Cardon, Rémi [Auteur]
Savoirs, Textes, Langage (STL) - UMR 8163 [STL]
Grabar, Natalia [Auteur]
Savoirs, Textes, Langage (STL) - UMR 8163 [STL]
Savoirs, Textes, Langage (STL) - UMR 8163 [STL]
Grabar, Natalia [Auteur]

Savoirs, Textes, Langage (STL) - UMR 8163 [STL]
Conference title :
CICLING 2019
City :
La Rochelle
Country :
France
Start date of the conference :
2019-04-07
HAL domain(s) :
Informatique [cs]
English abstract : [en]
Parallel sentences provide semantically similar information which can vary on a given dimension, such as language or register. Parallel sentences with register variation (like expert and non-expert documents) can be exploited ...
Show more >Parallel sentences provide semantically similar information which can vary on a given dimension, such as language or register. Parallel sentences with register variation (like expert and non-expert documents) can be exploited for the automatic text simplification. The aim of automatic text simplification is to better access and understand a given information. In the biomedical field, simplification may permit patients to understand medical and health texts. Yet, there is currently no such available resources. We propose to exploit comparable corpora which are distinguished by their registers (specialized and simplified versions) to detect and align parallel sentences. These corpora are in French and are related to the biomedical area. Our purpose is to state whether a given pair of specialized and simplified sentences is to be aligned or not. Manually created reference data show 0.76 inter-annotator agreement. We treat this task as binary classification (alignment/non-alignment). We perform experiments on balanced and imbalanced data. The results on balanced data reach up to 0.96 F-Measure. On imbalanced data, the results are lower but remain competitive when using classification models train on balanced data. Besides, among the three datasets exploited (se-mantic equivalence and inclusions), the detection of equivalence pairs is more efficient.Show less >
Show more >Parallel sentences provide semantically similar information which can vary on a given dimension, such as language or register. Parallel sentences with register variation (like expert and non-expert documents) can be exploited for the automatic text simplification. The aim of automatic text simplification is to better access and understand a given information. In the biomedical field, simplification may permit patients to understand medical and health texts. Yet, there is currently no such available resources. We propose to exploit comparable corpora which are distinguished by their registers (specialized and simplified versions) to detect and align parallel sentences. These corpora are in French and are related to the biomedical area. Our purpose is to state whether a given pair of specialized and simplified sentences is to be aligned or not. Manually created reference data show 0.76 inter-annotator agreement. We treat this task as binary classification (alignment/non-alignment). We perform experiments on balanced and imbalanced data. The results on balanced data reach up to 0.96 F-Measure. On imbalanced data, the results are lower but remain competitive when using classification models train on balanced data. Besides, among the three datasets exploited (se-mantic equivalence and inclusions), the detection of equivalence pairs is more efficient.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Collections :
Source :
Files
- https://hal.archives-ouvertes.fr/hal-02430419/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02430419/document
- Open access
- Access the document
- document
- Open access
- Access the document
- cardon-CICLING2019.pdf
- Open access
- Access the document