Automated Content Rating Using Natural Language Processing Techniques
Advisor Information
Harvey Siy
Location
UNO Criss Library, Room 225
Presentation Type
Oral Presentation
Start Date
2-3-2018 9:15 AM
End Date
2-3-2018 9:30 AM
Abstract
The purpose of this project is to create an automated content rating system that uses natural language processing techniques. Individuals have differing views on what is appropriate in various forms of entertainment. With such varying degrees and definitions of objectionable content, decisions to consume forms of entertainment are made without first listening to or watching the material in question or by specifically combing through reviews. Current solutions to rate entertainment based on objectionable content includes websites that are usually geared towards movies or based on arbitrary reviews or standardized rating labels that provide limited information and flexibility. Therefore our project’s goal is to create an automated content rating system that uses natural language processing techniques to rate textual entertainment content. The system provides numerical ratings across a variety of potentially offensive categories by utilizing word weights associated with objectionable words and phrases. The system is web-based and allows the user to generate customized ratings for song lyrics, movie and tv show subtitles, personal files, and simple text. Users can view and compare content and its rating side-by-side to help them establish baseline rates and word counts and flagged words are also be provided. The system is in the process of being developed as a computer science capstone project.
Automated Content Rating Using Natural Language Processing Techniques
UNO Criss Library, Room 225
The purpose of this project is to create an automated content rating system that uses natural language processing techniques. Individuals have differing views on what is appropriate in various forms of entertainment. With such varying degrees and definitions of objectionable content, decisions to consume forms of entertainment are made without first listening to or watching the material in question or by specifically combing through reviews. Current solutions to rate entertainment based on objectionable content includes websites that are usually geared towards movies or based on arbitrary reviews or standardized rating labels that provide limited information and flexibility. Therefore our project’s goal is to create an automated content rating system that uses natural language processing techniques to rate textual entertainment content. The system provides numerical ratings across a variety of potentially offensive categories by utilizing word weights associated with objectionable words and phrases. The system is web-based and allows the user to generate customized ratings for song lyrics, movie and tv show subtitles, personal files, and simple text. Users can view and compare content and its rating side-by-side to help them establish baseline rates and word counts and flagged words are also be provided. The system is in the process of being developed as a computer science capstone project.