Cortical.io proposes a new paradigm for natural language understanding (NLU) that overcomes the limitations of other artificial intelligence (AI) approaches. Semantic Folding is inspired by the latest findings on the way the brain processes information. This novel theory introduces a new data representation, the Semantic Fingerprint, which encodes meaning explicitly, including all senses and contexts. The system understands the relatedness of two items by measuring the overlap of their fingerprints. As a result, it is very fast, reliable, and easy to implement—a breakthrough technology that leverages the intelligence of the brain to enable the natural language understanding of big-text data.
How does it work?
To begin with, a semantic space called a Retina Database is created via unsupervised learning of the reference material (for example, contracts, medical textbooks, financial documents, support requests). A Retina Database can be trained with different text collections to specialize in specific topics or a particular language domain.
In the next step, the Cortical.io Retina engine is used to convert the text repository into semantic fingerprints, numerical representations that capture the meaning behind natural language.
- Semantic fingerprints can be generated for language elements like words, sentences and entire documents.
- Any two pieces of text can be compared, regardless of length or language.
- Computational operations can be performed on the meaning of text data.
While traditional NLP systems are based on word frequency calculations, the Cortical.io Retina Engine uses a substantially finer-grained representation for every word: 16,000 semantic features are captured for every term.
Semantic fingerprints allow direct comparison of the meanings of any two pieces of text, showing thousands of semantic relations.
With Semantic Folding, semantic spaces are stable across languages, enabling direct comparison of text across languages without machine translation.
Semantic fingerprints are encoded in the form of Sparse Distributed Representations (SDRs): a data structure made up of a large number of individual bits, each of which can be turned on or off. The meaning of a fingerprint is determined by the behavior of these bits, with each bit contributing a small amount to the overall meaning.Learn more about semantic folding Learn more about sparse distributed representations