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.
Highlights
Pronoun resolution has been studied in the frame of theories of formal grammar, corpus studies, experimental psycholinguistic studies and NLP systems.1 But much of the findings made about the phenomenon are not shared between these disciplines
This paper takes a step towards more interdisciplinarity between the fields of NLP and psycholinguistics by building a cognitive computational model of pronoun resolution
With the ambiguous items from Crawley et al (1990) we observed that the Subject Assignment Strategy applies as a default strategy when the Parallel Function Strategy is not available
Summary
Pronoun resolution has been studied in the frame of theories of formal grammar, corpus studies, experimental psycholinguistic studies and NLP systems. But much of the findings made about the phenomenon are not shared between these disciplines. Pronoun resolution has been studied in the frame of theories of formal grammar, corpus studies, experimental psycholinguistic studies and NLP systems.. This paper takes a step towards more interdisciplinarity between the fields of NLP and psycholinguistics by building a cognitive computational model of pronoun resolution. As Keller (2010) argues convincingly, both the domains of NLP and psycholinguistics can benefit from such models. There is a very rich psycholinguistic literature of which researchers in the domain of NLP are often not aware. NLP techniques might improve if this literature is taken into account. Cognitive computational models are a new means to perform psycholinguistic research: by implementing different models that represent different theories, a comparison can be made by looking at the behavior of the models on actual experimental human data
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
