The problems associated with the automatic analysis of large amounts of data regarding law, science and society were discussed at the conference. PD Dr. Erich Weichselgartner, Deputy Scientific Director of the Leibniz Institute for Psychology Information (ZPID), participated in the session "Who 'owns' the text and data corpus?".
His contribution addressed the use of TDM in psychology and specifically at ZPID. "Approximately 200,000 scientific articles are indexed by bibliographic psychology databases annually," said Weichselgartner. "An interested researcher would have to read more than 10,000 pages a day." In order to get an overview in this flood of information, TDM serves as great help. For example, TDM supports machine learning algorithms which enable semantic search.
The problem is the legal situation, the question whether copyrights are violated by the automatic readout. "Voluntary access to data and publications defuses the legal problems," emphasized Weichselgartner. "ZPID supports the Open Science movement." It provides various corpora for science. Metadata of scientific papers via the search portal PubPsych, research data via the research data center PsychData and articles via the publication platform PsychOpen.
ZPID is also conducting research in the field of TDM. André Bittermann uses Topic Modeling, a computational content analysis technique, to determine current research topics in psychology. He recently demonstrated this technique at ZPID's Big Data in Psychology Conference.
An overview of the program and the speakers at the Text and Data Mining Conference in Trier can be found at text-und-data-mining.de.