Research
Our research topics are centered around large unstructered datasets, mostly textual data. We are working at the interface of the fields of information retrieval (IR) and natural language processing (NLP). We do research on topics like:
- Supporting humans searching the web (query understanding, comparative questions, user simulation and abstractive snippets)
- Analyzing large web crawls (near-duplicates and ad-hoc search)
- Identifying good arguments (argument mining and argument search)
- Analyzing and supporting human writing processes (authorship and paraphrasing)
- Understanding how we will (soon) communicate with many tools (conversational interfaces)
Ongoing Third-Party Funded Projects
- ACQuA: Answering Comparative Questions with Arguments (funded by DFG)
- SINIR: Simulating INteractive Information Retrieval (funded by DFG)
Other Projects
- Args.me : First search engine for arguments on the web
- Arguana : Argumentation Analysis (ArguAna) for the Web
- Conversational Search
- Learning to Rank from Implicit User Feedback
- Near Duplicates in Web Crawls
- Query Understanding via Entity Linking
- Total Recall in Systematic Reviews
- Web Archive Analytics
Finished Third-Party Funded Projects
- Conversational News (funded by Google Digital News Initiative)
You can find Shared Tasks in the Events section.
You can find more research projects of the Webis Group at webis.de/research .