Which NLU has the best campus

Natural Language Understanding (NLU)

Methods for processing texts and spoken language are often summarized under the generic term Natural Language Processing (NLP). The subject area NLP also includes text generation and that contentUnderstand of texts and language, this sub-area is called Natural Language Understanding (NLU).

Our NLU team uses a wide variety of NLP processes, but focuses on developing methods and solutions for the complete content capture of texts (NLU). An essential component of NLU is Artificial Intelligence (AI), which enables our algorithms to understand the relationships between words and documents. This enables our NLU.Solutions to intelligently search through documents, compare them with one another and extract relevant information from them in a targeted manner.

NLU methods are based on semantic representations of texts. These can understand and depict associations and connections between words, e.g. that the word invoice is related to the word payment. These semantic representations exceed the possibilities of the classic rule-based methods of text mining. Nevertheless, these text mining methods and tools, such as regular expressions (RE), syntactic or semantic rules and ontologies or knowledge graphs, are also important for our systems.

The NLU team not only uses the two components (AI and knowledge, e.g. in the form of rules), but also connects them with each other. At Fraunhofer IAIS we call this solution Informed Machine Learning or hybrid AI. Hybrid AI means that less training data is required and yet reliable results are achieved. With the same amount of training data, hybrid AI achieves better results than conventional methods.

You can find more information in our AI glossary