Knowledge graphs

Talk to your data

The amount of data available in companies is growing and growing. Quick access to relevant information can offer decisive advantages for companies.

Knowledge graphs structure the data and information, enable semantic linking and are in many cases the basis for making Artificial Intelligence applications explainable and delivering results that are comprehensible to humans.

Retrieve information efficiently and conveniently

In combination with a chatbot, acoustic speech recognition and speech synthesis, knowledge graphs are the basis of voice assistants that can answer content-related questions and provide important information in seconds.

Fraunhofer IAIS's unique technology incorporates years of experience from research and countless customer projects.

Use of knowledge graphs in companies

Step 1

Integrate and understand information

Information available in the company, whether in the form of internet pages or brochures (also in paper form), is fed into the knowledge graph.

Through the use of Artificial Intelligence, the speech recognition is oriented towards real conversations of people and in natural language (Natural Language Understanding). The system can also understand less common terms, dialects or colloquial language.

 

Step 2

Structuring knowledge

A knowledge graph is able to combine the most diverse sources of information into a dynamic knowledge base.

Associations and connections between words are also taken into account appropriately, for example, that the word invoice is related to the word payment in terms of content.

Step 3

Generate responses

Finally, an answer is generated to match the question posed. The structured data is converted into text and output. You have a query and refer to the previous answer? This is also no problem for the system.

The output can be both an acoustic signal and text. You can also discover our voice assistance solution Allinga.

Test our knowledge graph demonstrator now

The chatbot in the KI.Showroom of the KI.NRW competence platform demonstrates the functionality of an online dialog system based on a knowledge graph from Fraunhofer IAIS. An AI-based dialog system ("question answering system") makes the search for information more efficient and convenient for the user. The KI.NRW demonstrator illustrates such an AI-based chatbot for querying Corona case numbers worldwide. The animation explains step by step how it works.

Try out now