Öffentliche Vorlesung
Veranstaltet im Rahmen der Europäischen Sommeruniversität in Digitalen Geisteswissenschaften „Kulturen & Technologien“
______________________
Why does for some words the context of usage change? Often we can identify phases of intensive context changes in which clues to possible causes can be found. Examples of such phases are:
- controversial discussions in which the proponents or opponents of a position use central words in completely different contexts;
- punctual events such as natural disasters, historical events or technological changes, in which not only new words are introduced but at the same time old words are used in new contexts;
- cyclical events such as „Olympia“, in which the location of the event and the names of the actors acting or affected change in a time cycle determined by the event.
For the recognition of context changes, the lecture introduces the measure of context volatility. This makes it possible to quantify the rates of context change of words in a diachronic corpus over a given period of time and to identify exploratively those words whose context of use has changed conspicuously over this period. Normalising the rate of change, words of different frequency classes can also be compared with each other. Compared to the pure frequency analysis, low-frequency words can thus also be included in the analysis, for example, for the detection of weak signals, which are often indicated by early changes in the context of usage of low-frequency words. In addition to the highly volatile words, words whose contexts of usage change little over a longer period of time can also be identified, for example, because they are part of a formulaic language in a technical or legal context, or because they represent a political or social consensus in the period under consideration.
______________________
Gerhard Heyer is emeritus professor at the computer science department of Leipzig University. He has studied Mathematical Logic and Philosophy at Cambridge University (with support by the Britisch Council), and General Linguistics at the University of the Ruhr, where he received his Ph.D. in 1983. After research on AI based natural language processing at the University of Michigan, Ann Arbor, he has been working as a systems specialist and manager in the software industry and a company of his own that was one of the first to develop a translation memory system.
His field of research is focussed on automatic semantic processing of natural language text. In addition to numerous publications on this topic – including the German text-book „Text Mining: Wissensrohstoff Text“ (W3L-Verlag, 32011, revised edition scheduled for winter 2021 at Springer Campus) – he has also been conducting several research projects. Worth mentioning are in particular his contributions to the research infrastructure for the humanities and social sciences CLARIN-D, the joint work with GESIS on the DFG-funded interactive Leipzig Corpus Miner (http://ilcm.informatik.uni-leipzig.de/), and the application of machine learning for OCR and HTR (DFG-Verbundprojekt OCR-D, Koordinierte Förderinitiative zur Weiterentwicklung von Verfahren für die Optical-Character-Recognition).
Prof. Heyer is a member of numerous scientific advisory boards (including a.o. the KompetenzwerkD of the Saxonian Academy of Sciences) as well as advisor to the executive board of the Institute for Applied Informatics (InfAI), an adjunct institute of Leipzig University. He has also served as a member of the scientific advisory council of the GESIS Institute IZ from 1997 until 2006, and was a member of the GESIS Kuratorium from 2006 to 2007.
________________________