Using convolutional neural networks and LSTM to predict places at…
The team built heatmaps and machine learning models that calculate safe routes for major Indian cities with award-winning organization Safecity. 34 Omdena Collaborators leveraged the latest Artificial Intelligence tools to build solutions for preventing Sexual Harassment.
“UN Women states that 1 in 3 women face some kind of sexual assault at least once in their lifetime.” In Safecity’s experience, the statistics in India seem to be extremely high. A rape occurs every 20 mins. Yet most women and girls do not talk about this abuse for multiple reasons such as fear of society, culture, victim-blaming, fear of the police, tedious formal procedures, etc. Safecity’s mission is to empower women who have experienced sexual abuse and they have done a great job gathering one of the biggest global data sets on sexual harassment cases. In the challenge, the Omdena community built several solutions to help women at risk of sexual abuse.
The community leveraged open-source data as well as Safecity’s database to build heatmaps and safe routes using Nearby Search, Directions API, and Grid Coverage techniques. Part of the solution is a sexual harassment category classifier with 93 percent accuracy and the application of convolutional neural networks and LSTM to predict places at a high risk of sexual harassment incidents.
All articles from the challenge can be found on our blog below.
Safecity platform and app
Safecity´s platform functions as a crowdsourced place where individuals can share their personal stories of sexual harassment and abuse in public spaces, categorizing them by place, time, and type of harassment. This data then gets aggregated into hot spots on a map that indicate trends at a local level. Our AI-driven safe routes and heatmaps automate this process and result in an efficiency and accuracy boost.
We are thanking all community collaborators for the amazing work done!
Safecity is an award-winning platform that crowdsources personal stories of sexual harassment and abuse in public spaces. Safecity has collected over 10,000 harassment stories from over 50 cities in India, Kenya, Cameroon, and Nepal.
Spell is a powerful platform for building and managing machine learning projects. Spell takes care of infrastructure, making machine learning projects easier to start, faster to get results, more organized and safer.
Want to work with us?Learn more