Generation of iiRDS Meta Data through Linguistic Processing

A joint project by Siemens AG and Linguistic Content AG has successfully shown that iiRDS meta data can efficiently be generated from available meta data by using linguistic analysis. Two talks at the 2019 tekom and TC World conference in Stuttgart have presented the details of the approach.
The talk SIKiiRDS - Das Pilotprojekt von Siemens Digital Industries (SIKiiRDS – A Pilot Project by Siemens Digital Industries) has sketched the overall scenario. The overall goal of the project has been to provide information in a standardized form for internal use as well as for the end user in "Content Search & Delivery" portals.
The talk Einsatz von Sprachtechnologie zur automatischen Erzeugung von iiRDS-Metadaten (Language Technology for Automatic Generation of iiRDS Metadata) has focused on the linguistic analysis which automatically derives iiRDS annotations out of existing titles, indexes and so called key questions. The automatically derived iiRDS classes encompass subclasses of “ProductLifeCyclePhase" and "InformationSubject" as well as candidates for the classes "Component" und "ProductFunction". Possible next steps are the annotation of the core content as well as additional classification according to VDI 2770.
The potential benefits of deep linguistic analysis have also been the focus of the third talk of the IAI Linguistic Content AG. The topic has been terminology evaluation: Terminologie aufräumen mit linguistischen Analyseverfahren (Validating Terminologies with the help of Linguistic Processing).

We Have Moved

Since 1st of January 2019, IAI Linguistic Content AG has been located right in the city centre of Saarbrücken, Rathausplatz 8.

IAILC AG at tekom Fair and tcworld Conference 2018

The focus of our presentations at the tekom annual fair and tcworld conference in Stuttgart was on meta data extraction and legacy data recovery, though not neglecting linguistic services such as terminology extraction, terminology evaluation or language quality checking.

In his tool presentation, Paul Schmidt focused on Automatic Metadata Extraction for Content Delivery based on linguistic analysis routines. All presentations attracted much interest.

tekom Fair and tcworld Conference 2017

Cleaning Up Legacy Product Data Is Feasible

This was the conclusion of the talk Verjüngungskur für sprachliche Altdaten (Rejuvenating Treatment for Legacy Data) presented by Axel Theofilidis at the tekom/TC World conference 2017. Legacy product data may exhibit specific properties due to limitations of early IT systems such as texts written in capital letters only, without umlauts, with lots of orthographic errors and the like. Such data can be cleaned and recovered to a degree of 95 % and even more by using high quality linguistic processing tools. The tiny rest may easily be corrected and adapted intellectually.

 

 

Oct 11, 16:00 - 16:30
LT Industry Summit, 9-11 Oct, Brussels
Dr. Paul Schmidt

The compilation of a high quality terminological resource is a must for many scenarios. It is a basis for high quality technical communication, it is a prerequisite for efficient translation in technical domains, it facilitates the use of language technology such as authoring memories or translation memories. However, the compilation of a standardized high quality terminological resource is a time consuming and expensive process. Language technology has had this problem in its focus for a long time resulting in term extraction tools galore. We would like to present another one which claims to have a superior quality through its excellent linguistic basis. In addition, the tool is easily accessible through a web shop which makes it available for everyone that needs such a service.

‘Termextraktion in den Warenkorb’ (‘put term extraction into the cart’) was IAILC AG’s motto for this year’s tekom fair. IAILC AG presented a recently opened web shop which provides term extraction for German and English. Many visitors appreciated its easy accessibility, its cost effectiveness and above all the quality of the results.  

Linguistic services providers, translators, terminologists, and technical writers expressed confidence that there is a great potential in our approach to terminology services.

Also other services such as terminology evaluation, automatic indexation, compilation of word clouds from large document collections as well as the calculation of document similarities were of great interest to visitors. 

 

Luther Bible 2017

Lemmatisation of more than 800 000 items

Deutsche Bibelgesellschaft is publishing a new version of the Luther Bible, the Luther Bible 2017. IAI Linguistic Content AG has provided a lemmatisation for an electronic version of it with more than 800 000 items lemmatised. The lemmatisation allows for an efficient lemma based search.