A Skip-Chain Conditional Random Field for Ranking Meeting Utterances by Importance

We describe a probabilistic approach to content selection for meeting summarization. We use skipchain Conditional Random Fields (CRF) to model non-local pragmatic dependencies between paired utterances such as QUESTION-ANSWER that typically appear together in summaries, and show that these models outperform linear-chain CRFs and Bayesian models in the task. We also discuss different approaches for ranking all utterances in a sequence using CRFs. Our best performing system

• Type: Conference
• Year: 2006
• Where: EMNLP
• Authors: Michel Galley
• Keywords: Conditional Random Fields, EMNLP 2006, EMNLP 2007, Non-local Pragmatic Dependencies, Probabilistic Approach