top of page
Business Meeting

                                                   

Text Intelligence Solution (TIS) Overview

A broad range of visualisation tools are available including tables, maps, charts and timelines.
Many of SDOES-Sintelix’s out-of-the-box capabilities are world-leading and every aspect of SDOES-Sintelix can be configured to the customer’s needs from the user interface. SDOES-Sintelix provides automation of time and resource-intensive activities, enabling unstructured data to be leveraged at speed and scale. SDOES-Sintelix can be used to create structured data from human readable data.  This structured data supports multidimensional analysis of open source data and internal data holdings. SDOES-Sintelix’s API and its 125+ connectors can deliver data to conventional structured data systems.

sol 1.JPG

Text Intelligence Solution (TIS)

 

Capabilities

Solution2.png

One Platform, Many Applications

Ingest any Document

  • Ingest over 1500 file types

  • Email, PST, OST files, server connection

  • Open and dark web harvesting

  • Social media … Twitter, Facebook etc

  • OCR – Optical Character Recognition

  • Audio & video transcription

Extract Information & Networks

  • Extract entities, relationships and properties

  • Geotagging and clickable maps

  • Classification, tagging & predictive coding

  • Entity resolution & community detection

  • Topic extraction and mapping

  • World best scores

  • Immediate value and easy configuration

 Explore, Analyse & Report

  • Advanced contextual Search

    • on text, metadata, entities, search result clustering, synonyms, favorites, history

  • Alerts - emailed direct to you

  • Document taxonomies & BCS trees

  • Network visualization and editing

  • Drill down to documents

  • Open export to multiple formats

  • Configurable “Spreadsheet for text” capability

  • Reporting via an advanced wiki

Technology, Integration

  • Java server with Web browser clients

  • Multi user HTML 5 UIs + over 70 web services

  • Integration to IBM i2, sharePoint, 125+ platforms

  • Scalable to multiple terabytes of data

  • Full claims-based security

  • LDAP, Active directory; single sign on

  • No software limit to data sizes

  • No software limit to network sizes

Speed (8-core processor)

  • 30 pages of text per core per second

  • 2.6 million pages per core per day

  • 80 million entities per core per day

Solution3.png
Solution 4.png
solution 6.png
solution 7.png
solution 9.png
solution 8.png

 

Processing Chains SDOES-Sintelix can be thought of as a system for collecting, normalising and structuring unstructured data. The system accepts data in over 1,500 different file formats. The first stage of input processing is normalisation input data formats into a standard “document model”.  After normalisation, the document processing pipeline and the network generating pipeline (diagram, above) rapidly extract information and construct a network representing all the extracted information and relationships.

Document collections store normalised text data and source data, if required.  SDOES-Sintelix enriches the normalised documents with metadata, entities, relationships and classifications.  Enriched SDOES-Sintelix documents can be searched via simple keyword search or by using the advanced features of SDOES-Sintelix’s search interface. Collections and networks scale independently in the system making it possible to handle big data.

Networks combine the information SDOES-Sintelix extracts from documents to expose the explicit and implied linkages.  SDOES-Sintelix Networks store the information that drives visualisations: tables, link charts, timelines and maps.  SDOES-Sintelix’s advanced entity resolution capability allow networks to be fused with information derived from incoming data. 

Data Storage, Scalability: SDOES-Sintelix contains two main stores: for documents and facts derived from them and for networks derived from structured and unstructured data input.  SDOES-Sintelix’s storage is based on the highly regarded lucene search engine, which has been extended to handle big data sets. The document fact store houses the normalised and source documents, together with the references, entities, relationships and tags derived from them.  The network store is a graph database and provides source-supported data ready for analysis via SDOES-Sintelix visualisation tools.

SDOES-Sintelix requires no support from any third-party database, enabling it to be deployed quickly and easily and provide value immediately. The system can absorb processor cores indefinitely, providing linearly increasing input processing speed and capacity. Scaling is only limited by underlying hardware.  The primary data objects in SDOES-Sintelix are document collections, networks and configurations.

Configurations.  SDOES-Sintelix is highly configurable and contains built-in development environments to make the development each type of configuration rapid and testable.

Workgroup and Enterprise Capabilities: SDOES-Sintelix “projects” are a versatile mechanism for operating multiple workspaces and workgroups.  They allow shareable work spaces to be isolated, backed up, restored and exported.  Projects have their own security settings and may contain many user-defined document collections, networks and configurations. SDOES-Sintelix is equipped a fully configurable claims-based security capability.

Deployment: SDOES-Sintelix deploys on either Windows or UNIX family operating systems including Linux and OS X.  It is a client-server system; the client is your web browser and the server is written in Java and runs on the Java Virtual Machine (JVM). SDOES-Sintelix can be accessed from any modern platform or via APIs and operates as easily in the cloud as on premise.  The minimum memory requirement is 2 GB and two cores.

SDOES-Sintelix is unparalleled in providing a complete platform that handles the end-to-end unstructured data needs of defence, intelligence, and law enforcement. SDOES-Sintelix automates manual work and its highly performant capabilities can be combined to create new efficient approaches and high-productivity workflows. Commonly, vendors only provide fragments of the desired overall process with limited use cases causing continued challenges.

 

Deploying SDOES-Sintelix avoids the cost, delay and risk of integrating its internal components. Whereas others create silos for single use cases and integrate multiple solutions from different providers, resulting in ongoing complexity and cost. SDOES-Sintelix offers immediate value, use case independence, and open data import/export – so maximising value and return on investment. SDOES-Sintelix continues to rapidly grow with customer requests directly influencing future developments. We delight in progressively improving Sintelix to accommodate clients evolving needs and provide steadily increasing return on investment.

Solution 10.JPG

Architecture and Deployment

bottom of page