The Cortical Engine for Processing Text

CEPT Systems introduces a new way to process information represented by text.

Try our demos »Free API-Key »Contact us »

Information retrieval beyond statistics

The CEPT approach is inspired by the latest findings on the way the human cortex works. It breaks with traditional methods based on pure word count statistics. The innovation is fully disclosed in pending European and US patent applications.

Semantic Fingerprinting

The CEPT-Retina produces semantic fingerprints of language and thereby represents a new fundamental alternative to capture the inner semantics of natural language. These fingerprints can represent words, documents or the information needs of users. This approach helps get more relevant search results and classify information more efficiently. The CEPT-Retina also enables the making of intelligent decisions based on human-generated text input.

How it works

Words can be represented as fingerprints

A picture of cats or dogs represents a good example of a traditional semantic fingerprint. An audio recording would be another example that captures a different (audio) dimension. These kinds of semantic representation are handled by image resp. sound analysis, a process that is computationally intensive but feasible for some applications.

From Symbols to Numeric representations

The words CAT and DOG are symbolic representations of the entities of cats and dogs. To give meaning to these symbols, we need a dictionary as we are unable to interpret the representations on themselves. The CEPT-Retina transforms the symbol for CAT into its semantic fingerprint shown below in red, the same for DOG shown in blue. The overlay of the 2 fingerprints enables direct (visual) comparison of semantic relatedness. Therefore, we refer to them as semantic fingerprints.

The CEPT product line

Who uses our technology

The CEPT-Retina addresses industrial clients, service providers and individual application developers that are trying to improve the effectiveness and speed of their information retrieval processes, products and applications. It helps them enhance and extend the value and relevance of their data.


The first version of the CEPT-Retina allows the use of word-level semantic fingerprints. The CEPT-Retina will continue to evolve to extend the functions available at the word level but also to embrace successively the next levels (document fingerprints, user fingerprints and fingerprints of any entity that can be described with natural language). The CEPT-Retina is a language independent technology. It can be used in other languages, such as Spanish, German, French or Russian, etc.

Use Cases

The CEPT-Retina can be used for term expansion: a word is queried and expanded by the semantic fingerprinting engine to a list of context relevant terms. These words can then be used for disambiguation, classification, annotation and so on. The CEPT-Retina can also be used to request a comparative visual overlay of multiple terms. This helps users using analytics in establishing the most relevant concepts ranked by priority or labelling landscape maps. Users can also extract features for classification or annotation purposes.

How to access the CEPT-API

The CEPT-Retina is offered through an API and deployed using a Software-as-a-Service model. This approach has the advantage of keeping the lab-to-market time to the minimum. This also allows us to significantly reduce initial on-boarding and running costs.

The CEPT-Technology can be integrated very easily into your environment by using the CEPT Service-API either from your website, your intranet, your business processing engine or even your desktop applications.

Join our API user community and find out how the CEPT-Retina can unleash Meaning-Based Computing in your products and services!

Go to the Dev Portal now »


CEPT Systems GmbH was founded in 2011 in Vienna, Austria, as a science-based start-up with the mission to develop and commercialize a breakthrough, patent pending technology: the CEPT-Retina. Our small team combines many years of experience in the areas of information retrieval and analysis with a good knowledge of the professional information market.


Francisco E. De Sousa Webber, co-founder

Francisco took an interest in information technology as a medical student specializing in genetics and serology at the University of Vienna. He participated in various research projects at the Vienna Serological Institute and was heavily involved in medical data processing. He took part in such projects as establishing and organizing Austria's dialysis register database and creating a patient documentation system for the university clinic.

In the mid-1990s, he worked with Konrad Becker to found Vienna's Institute for New Culture Technologies and Public Netbase - at that time Austria's only free public-access internet server - thus establishing an international competency platform for the critical use of information and communication technologies.

In 2005, Francisco founded Matrixware Information Services, a company that developed the first standardised database of patents under the name of Alexandria where he acted as a CEO.

He also initiated the foundation of the Information Retrieval Facility in 2006, a not-for-profit research institute, with the goal to reduce the gap between science and industry.

Daniel Schreiber, co-founder

Daniel has many years of experience as managing director in different IT-services and media companies.

He was last CFO of Matrixware Information Services and treasurer of the Information Retrieval Facility.

In his early years, he founded and managed two trading companies, put in place a European distribution network and introduced marketing and logistics structure for the products of an American company.

max.recall, development partner

max.recall is an Austrian company offering consulting and development services for search and content analytics solutions such as vertical search or social media analysis.