Natural Intelligence is Human

Saffron develops software based on principles of memory-based representation and reasoning to make applications operate more brain-like. Our principal end users are human analysts who understand how to transform data into intelligence products. Human experts have learned through the experience of actions and results how to best address each situation. If technology is to assist us in transforming data into information, and information into actionable knowledge, then it needs to also be more natural to the way we work: it needs to be more brain-like. A more revolutionary approach, inspired by natural systems, is required to address the growing problem of data and how to help us transform data into intelligence. This is what Saffron Technology delivers – Natural Intelligence.

There are innumerable examples of associative memory from ordinary life. It is the way we think; we do not think in tables. When we are choosing a restaurant, we certainly do not look at a table of ethnicity (Thai versus Italian), price, or convenience. Instead, we intuitively integrate a variety of factors. We note how often and when we last ate there. We add in the quality of the food and service. Most of us calculate the price and the time of day. And we often adjust for the patrons (children versus adults). Our brains associate many of these factors to arrive at a choice.

Throughout our lives, we use associative memory to identify colleagues from various parts of our past.  It is the relationships, which matter: are they a friend of a spouse, or from work or college?  Relational databases lose much of this valuable information a priori, as the restrictions of columns and fields are pre-defined, often somewhat arbitrarily due to the constraints on size, structure, and management of maintenance. Saffron not only reads free text to identify relationships, it is almost unlimitedly scalable and will still return relationships based on frequency, hence statistical strength.

Memories Rather Than Models

Intelligence includes the ability to associate how things are related to other things. These associations are learned from experience. It’s not a new concept – Aristotle believed that knowledge was defined by associationism, and the birth of Psychology was also founded as the elemental study of how sensations, ideas, and actions are associated to each other. Although the human brain is more complex than the above concept allows, we’re always looking for new ways to make computers work more like our brains to better assist us.

As We May Think

Dr. Vannevar Bush is notably the first to articulate the idea of a human-like associative memory for computers in his article “As We May Think” published during July 1945 in the Atlantic Monthly. Bush suggested, “[That] The human mind…operated by association. Selection by association, rather than indexing, may yet be mechanized.” Sixty years later computing is still fixed on indexing rather than associating.

Models Limit Information

Computers were designed to index and retrieve documents and records according to their content; this is easy work for computers. Before Saffron entered the market, representing and remembering how everything is potentially related to everything else was hard because the process did not naturally scale well.

Traditional methods address this scaling challenge by reducing the information that resides in a population of data, into an abstract model or set of rules. Rules are simplified reductions of the available information and lose all the actual relationships and exceptions within the data. Statistics, another traditional method, reduce all the information residing in the data into a formula. Whether using rules or statistics, the work of knowledge engineering is to fit the data to a simpler model. These models may be representative of yesterday’s normal cases, but today’s problems require knowledge of the exceptions – whether the problem is a terrorist hidden in a sea of normality, or a drug’s effect on any specific patient. Rich and detailed information about each individual or entity is what matters.  Solving these problems requires a different approach.

Memories Between Data and Models

Saffron Technology’s Natural Intelligence approach is founded on our patented associative memory technology, which uniquely addresses the problems unresolved by traditional methods. Unprocessed data or content is too raw, searching through massive data sets and reading large volumes of documents to make sense of it is difficult for humans. To solve this problem Saffron memories store information as associations between everything in the data, rather than fitting data to a model that reduces information.

Saffron’s associative memory technology recalls all relevant associations, experiences, cases, and evidence – as they relate to the specific case, question or situation.