Digital Reasoning defines Big Data as too large, accumulating too fast, and too diverse in format and location to handle with conventional software and i
Digital Reasoning defines Big Data as too large, accumulating too fast, and too diverse in format and location to handle with conventional software and infrastructure. Big Data leads to information overload in government and industry, such as when logs and audits gather information that a business can’t track, when an influx of human, signals, satellite, and drone video intelligence cause an analyst in the IC to lose or miss important security developments, or when a government agency struggles to find trends in millions of documents. Solutions exist to better deliver documents via search and indexing, but with Synthesys, Digital Reasoning tries a different approach by automatically pulling facts, entities, relationships, and associated terms from structured and unstructured data so analysts don’t need to read all of the documents themselves.
The key to Synthesys is Digital Reasoning’s patented entity extraction software to identify the people, places, things, and whatever other categories interest users. Synthesys then augments these entities by showing how they connect as well as the strength of those connections, and determines their time and location. It can find unexpected linkages and associated terms or synonyms, even in coded language, slang, or jargon, and works with a growing number of languages. Synthesys performs text analytics without ontologies or models to drastically reduce the time required to prepare data and can therefore analyze streams of data as they come into the system. After analysis, Synthesys further helps users draw out the salient points through visualization to show entities, connections, and the nature of those connections. Analysts can drill down to find an abstract supporting the linkage or further still for the original document, context, and evidence.
Synthesys Cloud can then take these capabilities and move them into a public cloud with scalable computing and infrastructure on commodity hardware. Synthesys can do so effectively thanks to Digital Reasoning’s partnership with Cloudera, the leading provider of Apache Hadoop-based data management software and services. Cloudera’s Distribution including Apache Hadoop (CDH3) and HBase, the distributed, scalable Hadoop database, will be integrated into Version 3.1 release of Synthesys. Synthesys also integrates with Apache Cassandra, an open source database management system.
