UKP Home

Two UKP teams participated at this year’s Cross-Language Evaluation Forum competition.

Elisabeth Wolf took part in the Robust-WSD track. This track aims at exploring the contribution of Word Sense Disambiguation to monolingual and multilingual Information Retrieval. Elisabeth’s system - based upon the work by Bernhard and Gurevych (ACL 2009) - achieved the highest Mean Average Precision (MAP) score in monolingual retrieval. This work is based on the Emmy-Noether project “Mining Lexical-Semantic Knowledge from Dynamic and Linguistic Sources and Integration into Question Answering for Discourse-Based Knowledge Acquisition in eLearning” funded by the DFG. For more details, see the working notes paper.

Dr. György Szarvas and Benjamin Herbert joined the CLEF-IP track launched in 2009 to investigate IR techniques for patent retrieval. More than 1M patent documents derived from EPO sources, covering English, French, and German patents with at least 100,000 documents in each language are utilized to find patent documents that constitute prior art to a given patent. UKP ended up in the upper part of the ranking based on all participating systems. This work is done as part of the “Semantics- and Emotion-Based Conversation Management in Customer Support” project funded by the BMBF.

For more details, see the working notes paper.

Background: Cross-Language Evaluation Forum (CLEF) promotes R&D in multilingual information access by

• developing an infrastructure for the testing, tuning and evaluation of information retrieval systems operating on European languages in both monolingual and cross-language contexts, and
• creating test-suites of reusable data which can be employed by system developers for benchmarking purposes.