tailieunhanh - Báo cáo khoa học: "Decision detection using hierarchical graphical models"

We investigate hierarchical graphical models (HGMs) for automatically detecting decisions in multi-party discussions. Several types of dialogue act (DA) are distinguished on the basis of their roles in formulating decisions. HGMs enable us to model dependencies between observed features of discussions, decision DAs, and subdialogues that result in a decision. For the task of detecting decision regions, an HGM classifier was found to outperform non-hierarchical graphical models and support vector machines, raising the F1-score to from . . | Decision detection using hierarchical graphical models Trung H. Bui CSLI Stanford University Stanford CA 94305 UsA thbui@ Stanley Peters CSLI Stanford University Stanford CA 94305 UsA peters@ Abstract We investigate hierarchical graphical models HGMs for automatically detecting decisions in multi-party discussions. Several types of dialogue act DA are distinguished on the basis of their roles in formulating decisions. HGMs enable us to model dependencies between observed features of discussions decision DAs and subdialogues that result in a decision. For the task of detecting decision regions an HGM classifier was found to outperform non-hierarchical graphical models and support vector machines raising the F1-score to from . 1 Introduction In work environments people share information and make decisions in multi-party conversations known as meetings. The demand for systems that can automatically process information contained in audio and video recordings of meetings is growing rapidly. Our own research and that of other contemporary projects Janin et al. 2004 aim at meeting this demand. We are currently investigating the automatic detection of decision discussions. Our approach involves distinguishing between different dialogue act DA types based on their role in the decisionmaking process. These DA types are called Decision Dialogue Acts DDAs . Groups of DDAs combine to form a decision region. Recent work Bui et al. 2009 showed that Directed Graphical Models DGMs outperform other machine learning techniques such as Support Vector Machines SVMs for detecting individual DDAs. However the proposed models which were non-hierarchical did not significantly improve identification of decision regions. This paper tests whether giving DGMs hierarchical structure making them HGMs can improve their performance at this task compared with non-hierarchical DGMs. We proceed as follows. Section 2 discusses related work and section 3 our data .