The Durm Project

Printer-friendly versionPDF version

1. Overview

The Durm project, carried out from 2004-2006 at the Institute for Program Structures and Data Organization (IPD) at the University of Karlsruhe, Germany, investigated the use of advanced semantic technologies for cultural heritage data management. The goal was to support end users, in particular users from building history and architecture, with tools that go beyond classical information retrieval techniques. Experiments were carried out on the historical Handbuch der Architektur (Handbook on Architecture).

2. Project Results

Some of the major project results include the Durm corpus, the Durm Wiki, and the Durm German Lemmatizer, as well as a number of publications.

2.1. The Durm Wiki

One of the major results from this project is our strategy for transforming cultural heritage data into a Wiki format. A public version (albeit with a restricted set of features due to the shared web hosting environment) is available online.

2.2. The Durm Corpus

As part of the project, we digitized a complete book of the encyclopedia and transformed it into several formats. Both scanned page images and the marked-up text are freely available.

2.3. The Durm German Lemmatizer

Due to the rather special language of the Durm corpus, namely 100-year old architectural texts, we needed a lemmatizer that can accurately compute lemmas even for outdated, technical terminology. Since freely available resources for German are rather scarce, we developed a self-learning context-aware lemmatizer for German, which is available under an open source license.

2.4. Publications

A complete list of publications from this project is available here.

3. Project Members

Project supervision:

• Peter C. Lockemann (IPD, University of Karlsruhe, Germany)
• René Witte

The following group members contributed to this project:

Tom Gitzinger

Index Generation, Wiki/NLP Connection

Thomas Kappler

Durm Wiki, Durm Corpus, Wiki Bot

Ralf Krestel

Automatic Summarization, Ontology Population

Qiangqiang Li

NLP Pipelines, Ontology Population

Praharshana Perera

Index Generation, German Lemmatization.

4. Acknowledgments

This project was funded by the German research foundation (DFG) under the title "Entstehungswissen" (LO296/18-1).

5. Feedback

For questions, comments, etc., please use the Durm Forum.