Year: 2022
Type: dissertation
Abstract: Recent transformer-based language models pre-trained on huge text corpora have shown great \nsuccess in performing downstream Natural Language Processing (NLP) tasks such as text \nsummarization when fine-tuned on smaller labeled datasets. However, the impact of fine-tuning \non improving the performance of pre-trained language models in summarizing movie storylines \nhave not been explored. Moreover, there is a lack of extensive labelled datasets containing movies \nstorylines to allow pre-trained language models delving deeper in this realm. In this research work \nwe propose a novel labelled dataset containing IMDB movie storylines alongside their summaries \nfor teaching pre-trained language models how to perform text summarization on movie storylines. \nFurthermore, we showcase the potential of this dataset by fine-tuning a T5-base model with the \nuse of this dataset. Our results show that fine-tuning a T5-base model on this dataset can \nsignificantly improve the performance in summarizing movie storylines
Author: Ghasemi, Hojat
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6991113646