Inspiration
This app has been developed in order to predict the risk of individuals being drawn into online sex work. We want this app to be a reference for adult website users and educators in this field to gauge the effect of their activity on such sites and to avoid any harm to themselves.
What it does
It predicts the risk of an innocent individual falling prey to the web of online sex-work by using their preferences and their online activity as features to understand their susceptibility to it. The data used to build this app can be found here. This database was used in the paper: “Covert online ethnography and machine learning for detecting individuals at risk of being drawn into online sex work”
How we built it
Using R and Shiny apps.
Challenges we ran into
- Inadequate data
- anonymized data-fields
- Some App functionality
- Lower Score of prediction algorithm based on training data
Accomplishments that we're proud of
Building and deploying a working machine learning model within 10 hours of the inception of its idea! And, getting 72% accuracy on our classification model on our first try.
What we learned
We learned based on the online-activity and individual characteristics there can be a high risk of teens and young adults falling prey to online sex work or the sex trade.
What's next for Online Sex-work Risk Prediction
With the availability of more data and data-sources, the online-sex-work-predictor will have more robust predictions Also, We will be using a web-API to extract data for predicting some individual is at Risk
Built With
- classification
- machine-learning
- model
- r
- shiny
Log in or sign up for Devpost to join the conversation.