A Computational Model of Human Preferences for Pronoun Resolution

Olga Seminck, Pascal Amsili


Abstract
We present a cognitive computational model of pronoun resolution that reproduces the human interpretation preferences of the Subject Assignment Strategy and the Parallel Function Strategy. Our model relies on a probabilistic pronoun resolution system trained on corpus data. Factors influencing pronoun resolution are represented as features weighted by their relative importance. The importance the model gives to the preferences is in line with psycholinguistic studies. We demonstrate the cognitive plausibility of the model by running it on experimental items and simulating antecedent choice and reading times of human participants. Our model can be used as a new means to study pronoun resolution, because it captures the interaction of preferences.
Anthology ID:
E17-4006
Volume:
Proceedings of the Student Research Workshop at the 15th Conference of the European Chapter of the Association for Computational Linguistics
Month:
April
Year:
2017
Address:
Valencia, Spain
Editors:
Florian Kunneman, Uxoa Iñurrieta, John J. Camilleri, Mariona Coll Ardanuy
Venue:
EACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
53–63
Language:
URL:
https://aclanthology.org/E17-4006
DOI:
Bibkey:
Cite (ACL):
Olga Seminck and Pascal Amsili. 2017. A Computational Model of Human Preferences for Pronoun Resolution. In Proceedings of the Student Research Workshop at the 15th Conference of the European Chapter of the Association for Computational Linguistics, pages 53–63, Valencia, Spain. Association for Computational Linguistics.
Cite (Informal):
A Computational Model of Human Preferences for Pronoun Resolution (Seminck & Amsili, EACL 2017)
Copy Citation:
PDF:
https://aclanthology.org/E17-4006.pdf