Illegal behavior detection
The business goal
Our client is a small Canada-based software development company.
The business goal of the project was to create an application that could discover and track illegal behavior based on digital footprints (news, chatter, etc. in the web). The app would be presented as a website that will search for personal information. The gathered information would only be relevant to illegal actions.
Technical side
On the technical side, we have developed a system, that searches specific sites to extract relevant information, detect personal names and extract facts about possible illegal activity. For this, we used the following technologies: Python, ZeroMQ based crawler/post processor to run many scans simultaneously. We also used Selenium to scrape websites(mostly JavaScript enabled, we need to interpret JS to make scrapping successful). CoreNLP used for individual name detection (NER).
Result
As a result, our client has presented the application to potential investors for further development and has taken part in several conferences/panels.
Moscow, 117335,
Russia