Summarizing Decisions in Spoken Meetings
This work addresses the challenge of summarizing decisions in meetings, which is incremental as it builds on existing summarization techniques for a specific domain.
The paper tackles the problem of summarizing decisions in spoken meetings by producing concise decision abstracts, comparing token-level and dialogue act-level methods with supervised and unsupervised frameworks. In supervised settings, token-level summaries with discourse context approach an upper bound, while in unsupervised settings, LDA-based topic models achieve 0.22 ROUGE-F1, comparable to 0.23 with SVMs.
This paper addresses the problem of summarizing decisions in spoken meetings: our goal is to produce a concise {\it decision abstract} for each meeting decision. We explore and compare token-level and dialogue act-level automatic summarization methods using both unsupervised and supervised learning frameworks. In the supervised summarization setting, and given true clusterings of decision-related utterances, we find that token-level summaries that employ discourse context can approach an upper bound for decision abstracts derived directly from dialogue acts. In the unsupervised summarization setting,we find that summaries based on unsupervised partitioning of decision-related utterances perform comparably to those based on partitions generated using supervised techniques (0.22 ROUGE-F1 using LDA-based topic models vs. 0.23 using SVMs).