We are happy to support a leading manufacturer and supplier of operations control systems for railway traffic. LangTec’s contribution in this project is to support with dedicated expertise in the development of scripting solutions on Linux systems and the virtualisation of central control systems.

LangTec participated in the Hessian Ministry of Economics Innovation Competition. One problem we want to tackle is to simplify the conversion of driving licenses for non-EU residents. A non-EU driving license is only valid for six months from the first arrival in Germany. After that, one must convert their driving license to a German driving license. However, the steps for one country can be different from one to another.

Therefore, we would like to propose a driving license conversion system to help facilitate this process. Our existing product, AlatUno, a multi-lingual PDF filling system, assists users in filling the necessary application form. In addition, based on the uploaded driving license, it will automatically detect the country and decide which steps are required.

We presented our solution in the final round of the Innovation Contest of the Hessian Ministry of Economics. Our competition was extremely strong and professional. Still, at the end of the day, the jury voted for us and our solution AlatUno was awarded the gold medal in this competition!

Last week we had a team lunch to say goodbye to Soumya, who will be working from India in the future. We had a delicious lunch at AuthenTikka and enjoyed getting together in real life.

In order to facilitate the settlement process between the harbour master and shipping companies, LangTec, with expert knowledge in NLP, develops a tailor-made parser for documents with commercial, technical, legal and safety information on tankers for our shipping client in South-East Asia. The processed documents include standardised questionnaires “Questionnaire 88” and harbour timesheets in a number of different input formats such as text, images or webpages. LangTec’s solution comprises information ingestion, parsing, analysis, normalisation and disambiguation of process-relevant data and is offered as a REST-compliant web service.

The annotation and analysis software CATMA is one of the core components of the BMBF-funded research project forTEXT at the TU Darmstadt. forTEXT brings major extensions and optimizations for the git/Gitlab based backend of CATMA. This includes a simplified layout of CATMA’s underlying git-projects and their file and folder structure, an efficient indexing of text and annotations with regards to memory consumption and performance and a new and easy collaboration mode for the analysis of common annotation results without the need of lengthy merge processes.

LangTec supports this development with the expertise in software engineering, Java and graph-based data structures.

Hamburg was at its best self with the most excited winds and waves for our Christmas party and guess where we kicked-off the party!? We met at the Elbe river dock and we headed out for our journey on a harbour cruise along the river! The river ride was extremely adventurous with huge waves rocking our boat and winds fluttering the boat-roof, added on top of it was an engaging live commentary by a very humorous sailor.

After finally setting our foot on land, we dined at Philipps restaurant and tried some authentic German-style Christmas dishes. We had a very pleasant dinner with face-to-face conversations without a screen separating us! We really had a very successful evening with delicious food, drinks, splendid ambience and of course, light-hearted talks. 🙂

As part of the cooperation with Prof. Dr. Hestermann in the research project Police Press Releases, two of our computational linguists visited Marcomedia University Hamburg with the aim of introducing journalism students to the field of semantic text analysis. The automated processing of large amounts of data offers great potential for journalistic projects, as explained in the lecture using the research project of police press reports as an example: In order to be able to make statements about the origin of persons mentioned in police press reports, the distinction between suspects and victims is central to the quality of the statements. In implementing this distinction, the students actively support LangTec by annotating data sets that are later used to train the machine classification model. The seminar concludes with a lively discussion about difficulties and requirements for high-quality annotations and future uses of semantic text analysis.

The report of the Marcomedia University can be read here.

Our research project ADAPI, which we are working on with the Fraunhofer Institutes IDMT and IBMT, has been granted an extension until May 2022. We are looking forward to the first results of our tests of ADAPI in the clinic.