VAL

Problem and Target Audience

• Gender inequality issue - women don’t have safe spaces online
• A harmless video, with no ill intent produces
• Men feel entitled to make comments about a woman’s body which are demeaning
• Men who see more objectifying content report stronger attitudes supporting violence against women than men who are not exposed to as much objectifying content (Wright & Tokunaga, 2016).

Solution Overview

• She provides a dashboard that summarizes
  • Post analytics
  • Flagged profiles
  • Account Protection
  • Digital Wellbeing
• Allows users to identify which post is gaining the most unwanted attention
  • Potential preventative measures
• Which commenters are contributing the most harmful comments
  • Users can block/remove these flagged profiles based on the model
• Trained model can learn words that classify words as degrading, which can then be applied to flag commenters that are engaging in inappropriate behavior

AI Implementation

• ML model - SVM
• Boost obscene words
• Upsampled the instances that are classified as malicious
  • To improve the model’s performance when predicting these classes

Impact and Vision

• Women can have safer online communities
• Do not have to fear men
• Have better peace of mind/not be exposed to harmful comments that could be damaging to a woman’s mental state

Built With

• python