Associative Memory Base: The Power of Natural Intelligence in the Real World

Imagine a network wide community of analysts of personal assistants. — Think millions of them communicating and collaborating with you. You give them millions of documents to read. Each analyst concentrates on a single person, place, thing or event. Your community observes, stores, and remembers every association among people, places, things and events in great detail and in context. —Think millions of attributes for each. When asked a question, your community is highly collaborative. And best of all, your community memories never forget — they have total recall.

The power of Saffron’s associative memory-based approach rests in its simplicity, inspired by natural intelligence. We see things and learn how things are related when we see them together (or in sequence). We remember the coincidences. We also reason from this experience. New situations remind us of prior situations, recalling what other things, actions, or outcomes are likely connected, while imagining what’s missing or what’s new and different. As in human experience, software should also capture knowledge from direct observation of the real world.

Co-occurrence matrices and memories—instead of tables and models—are the foundation of Saffron’s associative memory base. This new form of data representation allows users to quickly extract significant results with extraordinary accuracy from massive and noisy data sets.
Since matrices provide a superior representation of the relationships in the data, they allow Saffron’s Natural Intelligence™ Platform to observe and evaluate all sources of data to establish which attributes provide relevant information for a given context. Then Saffron’s associative memory base recalls relationships, similarities or recognizes patterns based on its “experiences” at query or run time. On the other hand, the design of relational tables typically requires a priori understanding of the queries that the design need to support. This is problematic for discovery because how do you know what needs to be answered before you start? How do you know what questions will follow once you begin your analysis? By correlating all the data into entity matrices Saffron can quickly reveal hidden relationships in the data that otherwise would be lost by predetermining what is important.

In Saffron’s Natural Intelligence Platform, observed patterns in data become memories as Saffron observes the associations between them. Saffron’s autonomous, associative memory based systems learn from the constant stream of incoming memories and can make predictions based on analogy or experience—even when the question is complex, interdependency increases between the variables and the data volume skyrockets.

This is exactly how your brain works. Your brain learns patterns, recognizes learned patterns—from memory—and helps you make decisions about what to do next, all based on analogy or experience. We don’t go offline to model the environment and use a model to predict what happens next.

Why not use a memory-based system that works like your brain, making predictions quickly and accurately using analogy or experience?

Saffron’s Natural Intelligence™ Platform provides just that. It lets the data speak for itself, and allows the data to reveal information, that would otherwise remain hidden in plain sight, by using a memory-based system of entity analytics and prediction.

Saffron’s Natural Intelligence delivers:

• Rapid ingestion and processing of information with comprehension of the whole body of data;
• Effective detection of critical elements, structures and patterns in relationships of entities in transactions;
• Accurate visualization of complex models and systems;
• Precise expression of discovery;
• An unusual capacity for embracing complexity and scale, with memory and ability to learn in an integrative, intuitive, nonlinear manner.