“Our program uses simple graphs and color codes to show users exactly how their data could be used. For instance, some websites share geolocation data for marketing purposes, while others may not fully protect information about children. Such clauses are typically buried deep in their data protection policies,” says Hamza Harkous, a post-doc working at EPFL’s Distributed Information Systems Laboratory and the project lead.
Using TITAN X GPUs and the cuDNN-accelerated TensorFlow deep learning framework, the team trained their convolutional neural networks on over 130,000 online privacy policies from apps on the Google Play Store. For more details about their deep learning architecture, read their recent paper, “Polisis: Automated Analysis and Presentation of Privacy Policies Using Deep Learning“.
Their program called Polisis can be used for free of change as a browser extension or directly on their website by inserting the website’s url. They also have an online chatbot called PriBot where you can enter questions about a website’s data protection policy.
Read more >
- GTC session: How to Assess and Mitigate Bias in AI Models (Spring 2023)
- GTC session: BNY Mellon's Journey in Enterprise AI/ML From Development to Production (Spring 2023)
- GTC session: Foundations for Deploying Trustworthy AI (Spring 2023)
- Webinar: Applications of AI for Anomaly Detection
- Webinar: Inception Workshop 101 - Getting Started with Conversational AI
- Webinar: Detecting Cyber Threats with Unsupervised Learning