Institutskolloquium

Web-Scale Domain-Specific Information Extraction

Prof. Dr. Ulf Leser, Institute for Computer Science, Humboldt Universität zu Berlin

Freitag, 13. Januar 2017 14.00 Uhr
Mathematikon, Im Neuenheimer Feld 205, Konferenzraum 05.104, 5. OG

 

Abstract

Information Extraction (IE) from unstructured texts is a technology with growing importance in many applications. Three important challenges to IE are the achievement of high quality results, scalability of methods to very large corpora, and integration of IEresults with other data for downstream analysis. In this talk, we willhighlight recent advances and open questions in these areas by drawing from extensive experiences in developing and applying IE forbiomedical research.

About the speaker

Prof. Ulf Leser holds the chair 'Knowledge Management in Bioinformatics‘ at theHumboldt University Berlin. The group on "Knowledge Management in Bioinformatics" has a proven record of successful  research and applications especially in biomedical databases, scientific workflows, statistical analysis of high - throughput experiments, and biomedical text mining. It gathers more than 15 years experience in managing and analyzing biomedical data set, especially from high - throughput experiments. Previous and current projects in this area are concerned with analysis and management of proteins and their function, phenotype data, cancer-related knowledge bases, protein-protein-interaction  networks, and sequence / microarray / ChIP-Chip / ChIP-seq data sets. The group has developed several popular integrated biomedical databases (e.g. IXDB, COLUMBA) and biomedical search engines (GeneView, Alibaba). The group currently consists of 12 scientists, all of which work on topics related to biomedical data and biomedical knowledge management. Prof. Leser currently is part of six publicly funded projects in biomedical research and a member of two publicly funded Graduate Schools and one research unit.