There are more AI technologies in our platform and in all our solutions than most AI experts know. That’s because Empolis developed AI technologies for industrial use more than thirty years ago – when the computing power for modern machine learning was still a vision and not yet a reality. That’s why we excel at handling information intelligently even where training data for neural networks is lacking.

Three keys to most of our solutions:

Federated Search

We don’t want to disrupt existing IT landscapes – all original data stays where it is. But all relevant information from all relevant data pools is brought together in one central spot and in a linked information model. Thus, our users quickly identify sources they need in this single spot. Of course, taking into account the assigned permissions and rights from the source systems. Depending on preference, use case and source, we use sharp phrase searches, Boolean searches, fuzzy searches on text similarity, but also semantic searches to identify synonyms or concepts similar in content. Fast, powerful, flexible.

The right tool for each search

Decision trees map processes and workflows based on structured and associative search and classification procedures. For example, the existing recipe for the fruit salad can also be captured as a step in a sequence. Semantic search can easily be activated and adapted to your needs. In this way, terminology is adapted to your users’ needs in the existing data and they map simple relationships. Apples and pears go with fruit salad.

When associative search is activated, statistical correlations and similarities are also taken into account: If you search for apples, you have also often searched for pears. Full text search requires no configuration and is immediately available on all texts. Thanks to modern linguistics, apple sauce is also found when searching for apples.

Knowledge Graph

Many things just can’t be determined statistically, but they can be recorded quickly and easily in a structured manner. For example, the maximum speed limit of 50 km/h within city limits, the process required for a business trip, or who within the company has certain skills and experience. The best way to determine this is in an enterprise knowledge graph. Our knowledge graph technology comes with a unique, patented rights management feature that ensures that even in complex networks, everyone only sees what they are allowed to see. Another special feature for insiders: We know that there are not only relationships between nodes, but also between edges – our Knowledge Graph knows that too!

Draw conclusions from linked data

Most of the information we need is not explicitly recorded. Nevertheless, we can deduce a lot from existing knowledge and, draw conclusions: Peter works in project STORM, project STORM is performed on behalf of ACME Ltd., thus, Peter probably has contacts at ACME Ltd.

Knowledge graphs are not limited to simply recording and reproducing data. They map relationships in a formal model that make all company knowledge accessible.

See whitepaper

Case-based reasoning

A smart person learns from his/her own mistakes, a wise person learns from mistakes made by others, the fool from neither. And to ensure that our customers can access all of their company’s experiences at all times, we have integrated case-based reasoning into our products as a key process – recognized by the “German Informatics Society” as one of the ten significant technologies in German AI history. Case-based reasoning in a nutshell: We capture every case with all of its characteristics – regardless of whether it is a description for a technical problem or an administrative request. Additionally, we document the associated solution or decision. With this approach, we can look closely at each new case to see whether similar cases already exist and how we dealt with them. By these means, gained experience is never lost.

Solve problems based on experience

The #KI50 jury of the German Informatics Society says: “In the American legal system, searching for suitable precedents plays a substantial role. Therefore, an AI approach for case-based reasoning has been developed in the U.S. to support machine searches. In Germany, the potential of this approach for industrial applications was recognized early on by Michael M. Richter. With his work and that of his group, he has advanced the field of case-based reasoning or experience-based systems internationally. One example is the model of “knowledge containers” proposed by him, which has contributed significantly to the understanding and design of applications. As a result, a technology established in practice has emerged, which has also been reflected, for example, with the founding of the company Empolis (in 1991 first established as tec:inno), which is now a leader in this AI environment.

(German Source:

How to better utilize your knowledge!

Start now

Memberships & Certificates