AIMU: Actionable Items for Meeting Understanding

AIMU: Actionable Items for Meeting Understanding Yun-Nung Chen and Dilek Hakkani-T¨ur Microsoft Research, Redmond, WA, USA y.v.chen@ieee.org, dilek@ieee.org Abstract With emerging conversational data, automated content analysis is needed for better data interpretation, so that it is accurately understood and can be effectively integrated and utilized in various applications. ICSI meeting corpus is a publicly released data set of multi-party meetings in an organization that has been released over a decade ago, and has been fostering meeting understanding research since then. The original data collection includes transcription of participant turns as well as meta-data annotations, such as disfluencies and dialog act tags. This paper presents an extended set of annotations for the ICSI meeting corpus with a goal of deeply understanding meeting conversations, where participant turns are annotated by actionable items that could be performed by an automated meeting assistant. In addition to the user utterances that contain an actionable item, annotations also include the arguments associated with the actionable item. The set of actionable items are determined by aligning human-human interactions to human-machine interactions, where a data annotation schema designed for a virtual personal assistant (human-machine genre) is adapted to the meetings domain (human-human genre). The data set is formed by annotating participants’ utterances in meetings with potential intents/actions considering their contexts. The set of actions target what could be accomplished by an automated meeting assistant, such as taking a note of action items that a participant commits to, or finding emails or topic related documents that were mentioned during the meeting. A total of 10 defined intents/actions are considered as actionable items in meetings. Turns that include actionable intents were annotated for 22 public ICSI meetings, that include a total of 21K utterances, segmented by speaker turns. Participants’ spoken turns, possible actions along with associated arguments and their vector representations as computed by convolutional deep structured semantic models are included in the data set for future research. We present a detailed statistical analysis of the data set and analyze the performance of applying convolutional deep structured semantic models for an actionable item detection task. The data is available at http://research.microsoft.com/projects/meetingunderstanding/. Keywords: actionable item, meeting understanding, convolotional deep structured semantic model (CDSSM), embeddings. 1. Introduction Meetings pose unique knowledge sharing opportunities, and have been a commonly accepted practice to coordi- nate the work of multiple parties in organizations. With the surge of smart phones, computing devices have been easily accessible during meetings and real-time informa- tion search has been a common part of regular conversa- tions (Brown et al., 2015). Furthermore, recent improve- ments in conversational speech recognition suggest the pos- sibility of automatic speech recognition and understanding on continual, in the background, audio recording of conver- sations (McMillan et al., 2015). In meetings, discussions could be a rich resource for identifying participants’ next actions and helping them to accomplish those in a seamless way during or after the meeting, without interrupting the discussions. This paper proposes a novel task of actionable item de- tection in meetings, with the goal …

Content truncated to 500 words