Mining Group Nonverbal Conversational Patterns Using Probabilistic Topic Models
Mining Group Nonverbal Conversational Patterns Using Probabilistic Topic Models
复制标题
DOI：
10.1109/tmm.2010.2065218
复制
发表时间：
2010-12
影响因子：
7.3
通讯作者：
D. Jayagopi;D. Gática-Pérez
中科院分区：
文献类型：
--
作者：
D. Jayagopi;D. Gática-Pérez
群体会话行为的自动发现是社会计算中的一个重要问题。在本文中，我们提出了一种方法来解决这个问题，通过定义一个新的组描述符称为袋组非语言模式（NVP）定义的简短的观察组的互动，并通过使用原则的概率主题模型来发现主题。拟议的袋组NVP允许融合的个人线索，并促进不同规模的群体的最终比较。主题模型的使用有助于聚类组交互，并量化它们在正式概率意义上彼此之间的差异。在增强多方交互（AMI）会议语料库上发现的行为主题的结果被证明是有意义的，使用人类注释与多个观察员。我们的方法有利于“基于组行为”检索组会话段，而不需要任何以前的标签。
The automatic discovery of group conversational behavior is a relevant problem in social computing. In this paper, we present an approach to address this problem by defining a novel group descriptor called bag of group-nonverbal-patterns (NVPs) defined on brief observations of group interaction, and by using principled probabilistic topic models to discover topics. The proposed bag of group NVPs allows fusion of individual cues and facilitates the eventual comparison of groups of varying sizes. The use of topic models helps to cluster group interactions and to quantify how different they are from each other in a formal probabilistic sense. Results of behavioral topics discovered on the Augmented Multi-Party Interaction (AMI) meeting corpus are shown to be meaningful using human annotation with multiple observers. Our method facilitates “group behavior-based” retrieval of group conversational segments without the need of any previous labeling.