Biologically inspired, Saffron Technology’s Natural Intelligence platform is a new, patented approach to data analysis that reduces the time Analysts and Investigators spend in research and discovery while quickly, upon recall, uncovering the unknown relationships and similarities in their data. Saffron’s Natural Intelligence platform is like having millions of personal assistants working for you to read, remember and automatically recall every person, place, thing, and event in your data, in context of a subject, question or idea to be explored. Analysts are finally free to do what they do best: analysis.

• Optimizes entity analytics for representation and recall of knowledge through associations and similarities:
• Remembers relationships between people, places and things.
• Maintains perpetual awareness of all the data to retain, recall, and communicate information between people, places and things in context of other people, places and things and their attributes.
• Facilitates pattern and “link queries” to support prediction and pattern completion and prediction.
• Solves the inherent challenges of traditional data mining
• Observesmassive data sets – structured, unstructured and semi-structured.
• Updates knowledge in “Real-time”, rather than batch.
• Simplifies the data integration process.
• Eliminates the need for end users to have statistical modeling expertise or Structure Query Language development skills.
• Removes dependence on models, which are inherently error prone and hard to adapt to dynamic environments.
• Removes data boundaries and opens up information previously unavailable and unknowable.
• Identifies: non-obvious associations and relationships from structured and unstructured data:
• Entity relationships that exist but are not present at the surface data.
• Relationships that are expected but are missing
• Relationships that are present but are not expected
• Provides natural memory-based representation and reasoning:
• Incremental: Start from zero, learn case-by-case.
• Non-parametric: No knob-tweaking to build.
• Malleable: Adapt on the fly to new features.
• No over-training:  Doesn’t degrade as more data is seen.
• Novelty/Anomaly Detection: Identifies what it doesn’t know based on experience.
• Unified representation: Various inferences are computed at query-time.

Products

SaffronEnterprise™: provides the underlying associative memory base engine for memory representation and recall.   This memory-based engine captures and stores associations using the same basic representations that form the core of human memory. It is based on a patented associative memory technology with a 100% Pure Java™ Application Programming Interface (API) that is easily embedded into existing applications and systems, empowering them with highly adaptive, intelligent functionality.

SaffronAnalyst™:  serves as the Analysts interface to the Associative Memory base and accelerates the time to immediately actionable intelligence. Discover relationships and similarities in massive volumes of electronically stored data, in context, in a collaborative, actively communicative application.

Associated to:
Quickly read across structured and unstructured data sources to identify relationships between people, places and things. These relationships are context dependent, allowing the user to focus on the relevant relationships rather than the typical “ball of yarn” produced by conventional tools. The source material is also indexed by these associative memories down to the sentence (“snippet”) level for unstructured sources, and record level for structured sources, and is displayed as evidence of the entities and links. Provides new and interesting discoveries that the analysts may not have know to ask for, the evidence that supports the suggested relationships and the ability to share results with other analysts.

Similar to:
Quickly reads across databases to identify groups of significant similarity between people, places, and things. No rules, no models.  SaffronScope™ applies a set of new patent-pending methods to find information-based similarities and unusual groupings. SaffronScope analyzes all the available attribute information in the data to find duplicates, variants, aliases and other analogues with a high degree of accuracy, reducing false alarms, and enabling analysts to focus on what matters.