Find
Raven can recognize and download illicit videos from various online sources such as social media platforms.
Classify
Raven utilizes advanced machine learning algorithms to automatically classify videos into categories that it is trained for.
Analyze
Raven has advanced video processing and video recognition algorithms for quickly recognizing and extracting various objects of interest from videos.
Video recognition & analysis
Hundreds of hours of videos are uploaded every minute. Many of the videos are extremist propaganda or violate laws or the terms of service of social media platforms.
Raven is an intelligent software using machine learning to find, classify, and analyse videos with extremist content. It combines advanced machine learning, image object recognition, and crawling.
This is why we created Raven.
Efficiency
Increased throughput to locate multimedia content on social platforms.
Wellbeing
Eases the work of human analysts by reducing the time spent on analysing video content
Safer world
Provides links and video analytics to law enforcement and counterterrorism agencies to keep society safer.
Mission
Our mission is to make the world a safer place to live in.
Technology
Raven is built with a combination of automatic data collection, human input from subject experts, and advanced machine learning algorithms.
Data Collection
Raven features a crawler capable of scraping from various social media channels. Raven has collected several thousands of videos and downloads more continuously.
Subject experts
Raven works with leading experts in the field who provide valuable input in order to refine our crawling and classification.
Algorithms
Raven is equipped with advanced machine learning algorithms for video classification and image recognition/object detection. These algorithms are trained and tested using thousands of videos.
About Raven
Raven aims to make the world safer through video intelligence software tools to improve online security.
70% of all Internet traffic nowadays is video. Many of the videos are extremist propaganda or violate laws and the terms of service of the media platforms they are uploaded on.
To fight these phenomena, we created Raven, an intelligent software using machine learning to find, analyze, and classify videos with illicit content.
It combines advanced machine learning, image/object recognition, and crawling.
Raven has been developed in partnership with extremism experts and trained using thousands of videos in order to identify, classify and analyze extremist content.
Team
We are cybersecurity specialists with a passion for data science and machine learning applied in video analytics.
Raven Science is a research-led technology start-up that develops intelligent software using machine learning to find, classify, and analyse videos. It combines advanced machine learning, image object recognition, and crawling.
Raven has been developed by academics at City, University of London in partnership with extremism experts and trained using thousands of videos in order to identify, classify and analyse extremist content. We maintain a strong partnership with researchers from City, University of London, who are experts in cybersecurity, machine learning, image processing, cultural analysis, and wellbeing.
We host visiting academics, PhD and MSc students, presenting work at conferences.
Use cases
Video forensics for LEAs
Law enforcement agencies (LEAs) use Raven for triaging multimedia material. It is essentially a prevention and detection/forensics tool, depending on its usage. Our Report Generation feature enables LEAs to use Raven to generate reports that ultimately become evidence in their investigations.
Social media companies
Social media companies and online video content depositories can use Raven to identify content that violates their terms of service. It enables them to go through tons of content very fast, thus helping human analysts in making the final decision regarding any given video/image.
Features
Raven downloads videos from social media platforms like YouTube and Twitter and then applies video recognition algorithms to classify video frames using sample materials used in forensic analysis and counter-terrorism.
Based on specialized machine learning algorithms, Raven can analyse video files and detect elements like:
Faces
Text in English and Arabic
Logos
Weapons
An estimate of the amount of propaganda found in the video.
Fast, Detailed, Full
Contact
Email: info@ravenscience.com
Adress: Northampton Square, London EC1V 0HB, United Kingdom