*metrics are based on benchmarking during pilots
Behavox analytical engine is able to process any type of structured (logs, browsing history etc.) and unstructured (emails, chats, phone calls, voice messages etc.) data. Processing involves full indexation for search, and data enrichment for running machine learning algorithms.
Full activity log and collaboration with commenting. All alerts are shown in case management. Highlighting and full explanation of the case logic, in addition we provide links to similar previous enforcement cases.
Employees can be added to the watchlist, both allowing the users quick access to their communications and cases involving them and allowing the system to run specialized detection algorithms applicable exclusively to employees currently under heightened supervision.
Users can search through all available types of data at once employing advanced search capabilities e.g. use fuzzy logic, search for certain sentiment, emotions, search within attachments etc.
Risk scoring indicates the probability of misconduct and is based on the track record of employees' compliance breaches. It allows users to identify and focus on the biggest potential risks instantly.
Allows users to see the complete map of any employee's professional relationships. Proprietary and patented modelling is used to calculate relationships between people based on the content and meta data.
Automated analysis of employee's behavior and conduct based on the historical observation of the individual and the relevant peer group. Allows the system to automatically detect behavioral anomalies and alert users to them.
Many regulated firms are required to conduct random sampling analysis of the retained employee communications. Behavox achieves this 10 times more efficiently by smart filtering of irrelevant content from the sample.
The Behavox scenario (detection algorithm) engine provides compliance teams with an easy-to-use infrastructure for creating scenarios using our proprietary standard syntax (Blink); these scenarios can then be pointed at a defined group of monitored employees.
Users can experiment with different scenarios and evaluate the quality of the results that would have been generated on their own historical data set. This process does not impact the production set-up of the system.
Behavox maintains and develops a library of out-of-the-box back-tested misconduct detection algorithms for the benefit of the whole industry. Clients of Behavox can contribute to the scenario repository and use any existing algorithms as well as new ones added each month based on market and regulatory change.
Behavox provides transparent real-time integration with its solutions via REST API, allowing users to integrate any kind of data feed into the system's data processing cycle.
© 2014-2016 Behavox