Ongoing digitization is causing rapid growth in the availability of text-based data at an ever increasing rate. This means that it is no easy task to prepare, process and analyze this overwhelming amount of information and data. We offer sophisticated natural language processing (NLP) software solutions as an effective tool for the automated processing of natural language texts. Behind these solutions lies great potential for making business processes more efficient, less resource-intensive and more cost-effective. The central technical components are language models, for which multilingual models are also available.
Intelligent and flexible search
If unstructured data repositories make it difficult to find relevant information and specifying precise keywords in a search query is not practical or desirable, NLP language models can enable a flexible search (e.g. as an additional search option) that compares the meaning of the search query with the information from the data repository and presents those results that are semantically the most similar to the search query. Furthermore, the language models can be adapted to a domain-specific language.
Automatic processing and mapping
NLP methods can automatically analyze natural language text, check it, and prepare it for further process steps. Furthermore, it is possible to assign instances or information to a text according to its meaning, e.g. the responsible department to a customer inquiry or the relevant legal texts to a described issue. The automation of repetitive activities allows employees to focus on their core competencies.
ExANT
The aim of this project is the reliable automated translation of natural language requirements for technical systems into a machine-usable form. Various NLP tools - including open source systems with OHB Digital Services as the knowledge carrier - are to be applied to natural language requirements from real space projects. This project is funded by the German Federal Ministry for Economic Affairs and Energy as part of the National Program for Space and Innovation and is technically and administratively supported by the German Space Agency at DLR. Further information on ExANT
Semantic Finder
At the core of this research project is a domain-specific semantic search engine that, in the context of space-related requirements documents, takes into account the meaning of individual words in a text in contrast to conventional keyword-based search. Thus, this modular software solution performs searches of documents and concrete information without having to rely on the input of specific keywords. As a result, employees can find the required documents without any need for expert knowledge and do not have to search in (unstructured) data repositories.