Simone Paolo Ponzetto

I joined the Data and Web Science Group in February 2013 as Juniorprofessor and hold since February 2016 the Chair of Information Systems III (Enterprise Data Analysis) as full professor (W3) at the University of Mannheim, where I lead the Natural Language Processing and Information Retrieval group.

webisadb WebIsA Database released

We are releasing a very large database of hypernymy relations extracted from the Common Crawl. You can find it here. For details please have a look at the LREC paper describing it (cite it if you find our work useful).

Our WebIsA Database has been used so far as a component in TAXI, our top-performing SemEval TExEval-2 system, as well as for our system participating in the ESWC 016 Open Knowledge Extraction Challenge - for which we received a nomination for best challenge paper!

joint JOIN-T team wins SemEval TExEval-2

A system developed as part of a collaboration between the NLP group of DWS and the Language Technology group of TU Darmstadt has been ranked first in an upcoming SemEval challenge on Taxonomy Extraction Evaluation (TExEval-2).

The results of the challenge can be found here. SemEval is the premier evaluation forum the computational semantics community. This work is part of an ongoing DFG-project (JOIN-T) collaboration between the two groups.

ecir ECIR-16 paper accepted

Our work with Laura, Michael and Ben Roth on constructing query-specific knowledge base has been accepted as short paper for the 38th European Conference on Information Retrieval.

cikm CIKM-15 paper accepted

Our work with Laura and Michael on ranking entities for unstructured Web queries has been accepted as full paper for the 24th ACM International Conference on Information and Knowledge Management. You can find the paper here.

ijhac IJHAC paper accepted

Our position paper with Federico and Hiram on semi-supervised machine learning and research methods in the digital humanities has been accepted for a special issue of the International Journal of Humanities and Arts Computing, one of the premier journal venues for research on arts and humanities computing, on the topic of "The Future of Digital Methods for Complex Datasets".

joint Some recent awards...

Our ESWC-14 paper on "A Probabilistic Approach for Integrating Heterogeneous Knowledge Sources" (with Arnab Dutta and Christian Meilicke) received an Honorable Mention for the Best Student Paper Award.

Our paper on evaluating central bank preferences using FOMC transcripts (with Nicole Baerg, Will Lowe, Heiner Stuckenschmidt and Caecillia Zirn) was selected as the "most promising" submission of the NLP Unshared Task in PoliInfomatics 2014.

Older news can be found here.