The latest version of Siren’s ‘investigative intelligence’ technology includes five new artificial intelligence (AI) capabilities: entity resolution, deep learning-based predictive analytics and alerting, deep learning-based times series anomaly detection, real-time topic clustering for textual corpuses and associative model technology for dashboards
The new AI enhancements all feature in the 10.3 version release of the core Siren product.
Siren’s investigative intelligence technology is, as it sounds, used for big data crunching in use cases involving fraud and crime investigation in finance and law enforcement.
Away from the ‘sexier’ end of crime-fighting, Siren can also be used to ‘investigate’ operational analytics (ITOps, cybersecurity, SIEM, telecommunications, IoT) and data discovery (knowledge graph/enterprise knowledge exploration and search).
Five ‘elements’ of AI
#1 — Dr. Giovanni Tummarello, co-founder and chief product officer at Siren explains self-correcting entity resolution as, “The AI/ML ability to recognise that records across different tables and data sources, using different schemas and different languages, are in fact talking about the same entity (person, company).”
#2 — Deep learning-based predictive analytics and alerting is real-time forecasting of operational data streams and alerting on expected future crossing of set thresholds.
#3 — Deep learning-based time series anomaly detection is the capability of learning from data to recognize and alert for anomalous behavior.
“Unlike other offerings on the market, Siren offers this based on automatic model selection ML, powered by dockerized TensorFlow backed APIs with seamless front-end integration,” said Tummarello.
#4 — Real-time topic clustering for textual corpuses is used on documents which have text (reports, emails, news articles, etc.), Siren now provides a built-in real-time visual interactive clustering exploration UI which is a good capability for news monitoring, investigative textual data discovery and e-discovery.
#5 — Finally, for now, associative relational technology (known in product terms as ‘Dashboard 360) is the ability to see in a single view a full ‘associative model’ 360-degree picture.