Applying Deep Learning to text classification for sentiment analysis
Molloy Faculty Mentor
Helen Dang
Presenter Major
Computer Science
Presentation Type
Oral
Location
Hays Theater, Wilbur Arts Building, Molloy University
Start Date
1-5-2026 11:41 AM
End Date
1-5-2026 11:47 AM
Description (Abstract)
In business, understanding customer behavior and emotions is one of the core factors driving success. Particularly within the entertainment industry, encompassing media such as films, videos, and various other products. Today, with the advancement of Artificial Intelligence and its application across diverse sectors including entertainment. This study focuses specifically on the problem of text classification. By addressing this task, we aim to detect the sentiments of users or customers across a range of domains. We employ deep learning models - a specialized subfield of Artificial Intelligence - applied to publicly available sample datasets to conduct experiments using various models and parameters. This research aspires to contribute to the field of forecasting in general, and to forecasting within the entertainment sector in particular.
The study utilizes publicly available, de-identified data sets and does not require IRB.
Keywords
Machine Learning, Business
Related Pillar(s)
Study
Applying Deep Learning to text classification for sentiment analysis
Hays Theater, Wilbur Arts Building, Molloy University
In business, understanding customer behavior and emotions is one of the core factors driving success. Particularly within the entertainment industry, encompassing media such as films, videos, and various other products. Today, with the advancement of Artificial Intelligence and its application across diverse sectors including entertainment. This study focuses specifically on the problem of text classification. By addressing this task, we aim to detect the sentiments of users or customers across a range of domains. We employ deep learning models - a specialized subfield of Artificial Intelligence - applied to publicly available sample datasets to conduct experiments using various models and parameters. This research aspires to contribute to the field of forecasting in general, and to forecasting within the entertainment sector in particular.
The study utilizes publicly available, de-identified data sets and does not require IRB.
