Data is the new oil – From Language Understanding in eSports to Knowledge Extraction in Networks
Nowadays, the collection of data is becoming cheaper and cheaper and the technological progress allows to measure nearly everything. The collected mass of data from very different domains allows for totally new insights and Data Science has started to reveal the hidden patterns by analyzing this new oil. In this lecture, results from a variety of domains ranging from eSports and social media over economics to our environment and bees will show the power of Data Science and the way to apply and develop new methods. For eSports, we adopt current natural language processing to learn something about the special language in this domain and to extract sentiments of participants when watching others playing online games. The analysis of user traces in social media systems does not only provide insights into the users behavior, it is also possible to make use of this data to extract more fine grained semantic information from various sources. Automatic detection of anomalies is central to reveal unusual behavior in economics, company networks or in sensor readings from bee hives. Making use of and getting new insights from data in all of these domains is the major challenge of research around Data Science and will be the central topic of this lecture.
Andreas Hotho is a professor at the University of Würzburg and holds the Data Science Chair. He is speaker of the new Center for Artificial Intelligence and Data Science (CAIDAS) at the JMU Würzburg. He holds a Ph.D. from the University of Karlsruhe, where he worked from 1999 to 2004 at the Institute of Applied Informatics and Formal Description Methods (AIFB) in the areas of text, data, and web mining, semantic web and information retrieval. From 2004 to 2009 he was a senior researcher at the University of Kassel and from 2011 to 2018 member of L3S in Hannover. Since 2005 he has been leading the development of the social bookmark and publication sharing platform BibSonomy.
For more than 10 years, his research group is working on topics related to data science. Basically, all types of data and analysis methods are in the center of their research interest, with the group specializing mainly in text and web data. In researching new Data Science methods for very large amounts of data, the group not only has experience in handling data from social media systems with varying degrees of structure, the analysis of text, among other things with a focus on historical novels in cooperation with Digital Humanities or on data from companies, but also works on the processing of sensor data for air pollution and develops new machine learning models for local climate models in cooperation with geography.
Andreas Hotho has published over 200 articles in journals and at conferences, co-edited special issues and books, and co-chaired workshops. The leading European conference in the field of Machine Learning and Data Mining, ECML PKDD, was successfully organized by him and colleagues in Würzburg in 2019. He received the SWSA Ten-Year Award at the International Semantic Web Conference 2018 for his work on the extraction of semantics and the Best Paper Award at the Web Conference 2015 for his analysis of user behavior on the Web using the HypTrails method. In recent years, the research work has been supported by funding from the EU, DFG and BMBF, but also by industrial collaborations.