Team: 19
School: La Cueva High
Area of Science: Behavioral and Social Sciences
Proposal: For college and class essays as well as reflections on specific topics, reading and interpreting essays can be time consuming. Additionally it can be easy to overlook plagiarism or extreme similarities between writing samples when the amount of writing to look over is very large.
To provide an easier and more efficient way of grouping essays and detecting possible plagiarism our program will analyze a large collection of writing samples for repeated words and grammatical structures. Additionally it will analyze "diction" for repeated connotations. If multiple people write using an extremely similar manner (tone as well as diction) the writing sample get grouped into a category.
This can be used to roughly detect who is working together/copying ideas on an essay or to analyze emotional response when a group of people is asked to respond to a controversial topic. Depressed tones in comparison to positive tones may also be applicable to analyzing mindsets or opinions in relation to current affairs. Additionally hate speech and inappropriate language can be detected and flagged quickly within a group of essays.
Team Members:
Sponsoring Teacher: Yolanda Lozano