Siren Platform User Guide

What’s new in Siren 10.3

Siren Platform

  • Investigate and Federate/Elasticsearch compatibility: Siren 10.3 now allows users to upgrade to a newer Investigate version without having to deal with the complexity of simultaneously upgrading their Elasticsearch and Federate backend.

Siren Investigate

  • Dashboard 360: Formerly, a dashboard's visualizations were limited to a single table (search). Siren 10.3 introduces a significant new feature, Dashboard 360. You can now define an internal 'dashboard data model' which enables a dashboard's visualizations to be based on different searches, and to coherently display data from connected records at each step. This provides an integrated 360-degree view showing what is connected, both directly and indirectly.

  • Topic Clustering: Topic Clustering is a new visualization in Siren 10.3 which performs significance and clustering analysis on full-text fields, highlighting significant terms whose frequency increases when filters and queries are applied.

  • Neo4j Import Wizard [Beta]: Neo4j is a graph database management system, which uses nodes, relations, and properties to represent data. Siren’s Neo4j Import Wizard makes ingesting and reflecting Neo4j data a straightforward process, just like importing from any other datasource.

  • Graph Browser enhancements: Graph Browser now works with intuitive drag-and-drop operations from dashboards, and enhanced configuration saving.

Siren Machine Learning (ML)

  • Siren ML deep learning-based prediction and anomaly detection: Siren Machine Learning merges Siren’s relational/cross-backend perspective with advanced TensorFlow deep learning technology. It provides two types of machine learning (ML) model—Anomaly Detection and Future Prediction.

    Siren ML requires a Docker installation; it also requires a plugin which provides the UI in Siren Investigate.

Siren Entity Resolution (ER) [Beta]

  • Siren Entity Resolution (ER): A beta feature that uses AI/Machine Learning to recognize that records across different tables, datasources, and schemas may be referring to the same entity (which could be a person, company, or any other thing). Siren ER (beta) can remove ambiguity  and reveal previously hidden relationships.

    Siren ER is currently implemented as a Docker installation.

Siren Federate

  • Connecting to remote Elasticsearch clusters: Siren Federate now provides the capability to query data from a remote Elasticsearch cluster through the Remote Clusters Module and the Federate Connector APIs.