Application of Machine Learning Techniques to Identify Non-Inf... | Helene Renninger | RRS Live 2024 скачать в хорошем качестве

Application of Machine Learning Techniques to Identify Non-Inf... | Helene Renninger | RRS Live 2024

Helene Renninger is a Junior Management Information Systems, Economics major in the Randall Research Scholars Program (RRSP). Their research project presentation, "Application of Machine Learning Techniques to Identify Non-Informative Images," was completed under the advisement of Dr. Nick Freeman from the Information Systems, Statistics, Management Department and Dr. Greg Bott from the Information Systems, Statistics, Management Science Department. Project Description: This research project belongs to a larger initiative called STANDD, Sex Trafficking Analytics for Network Detection and Disruption. STANDD's goal is to identify potential victims of human trafficking by scraping commercial escort sites for post data. Each ad posting contains text, images, phone numbers, and additional pieces of data. The disparate data (e.g., text, images) associated with these ads must be linked in order to identify individuals and perform valid analyses. Generic data, e.g., memes or generic text, is a significant issue as it can lead to false linkages of different individuals. This research investigates machine learning methods to identify image types commonly encountered in escort ads that lead to high levels of false links.