Research

UKP participated in the  Multilingual Opinion Analysis Task (MOAT) at the  7th NTCIR Workshop, held on 16–19 December 2008 in Tokyo, Japan.

We developed a supervised approach to the opinionated and the polarity subtasks.  We applied a sequential tagging approach at the token level and used the learned token labels in the sentence-level classification tasks.  In our formal run submissions, we utilized SVMhmm in both tasks with syntactic and lexicon-based features.

Our system ranked 2nd out of 9 systems in the opinionated sentence classification subtask and 3rd out of 5 systems in the polarity classification subtask.

A full description of the task, including results from all systems, is available in   the organizer's report.  A detailed description of our system is available in the following paper:

• Lizhen Qu and Cigdem Toprak and Niklas Jakob and Iryna Gurevych.  Sentence Level Subjectivity and Sentiment Analysis Experiments in NTCIR-7 MOAT Challenge. In Proceedings of the 7th NTCIR Workshop Meeting on Evaluation of Information Access Technologies: Information Retrieval, Question Answering, and Cross-Lingual Information Access, pages 210–217, December 2008.