06-08-2013, 04:28 PM
Tree-Based Mining for Discovering Patterns of Human Interaction in Meetings
Tree-Based Mining .pdf (Size: 700.2 KB / Downloads: 22)
Abstract
Discovering semantic knowledge is significant for understanding and interpreting how people interact in a meeting
discussion. In this paper, we propose a mining method to extract frequent patterns of human interaction based on the captured content
of face-to-face meetings. Human interactions, such as proposing an idea, giving comments, and expressing a positive opinion, indicate
user intention toward a topic or role in a discussion. Human interaction flow in a discussion session is represented as a tree. Tree-
based interaction mining algorithms are designed to analyze the structures of the trees and to extract interaction flow patterns. The
experimental results show that we can successfully extract several interesting patterns that are useful for the interpretation of human
behavior in meeting discussions, such as determining frequent interactions, typical interaction flows, and relationships between
different types of interactions.
INTRODUCTION
UMAN interaction is one of the most important
characteristics of group social dynamics in meetings.
We are developing a smart meeting system for capturing
human interactions and recognizing their types, such as
proposing an idea, giving comments, expressing a positive
opinion, and requesting information [1]. To further under-
stand and interpret human interactions in meetings, we
need to discover higher level semantic knowledge about
them, such as which interactions often occur in a discussion,
what interaction flow a discussion usually follows, and
what relationships exist among interactions. This knowl-
edge likely describes important patterns of interaction. We
also can regard it as a grammar of meeting discussion.
Data mining, which is a powerful method of discovering
new knowledge, has been widely adopted in many fields,
such as bioinformatics, marketing, and security [2]. In this
study, we investigate data mining techniques to detect and
analyze frequent interaction patterns; we hope to discover
various types of new knowledge on interactions.
HUMAN INTERACTION AND INTERACTION FLOW
Human Interaction Definition and Recognition
Human interactions in a meeting discussion are defined as
social behaviors or communicative actions taken by meeting
participants corresponding to the current topic. Various
interactions imply different user roles, attitudes, and
intentions about a topic during a discussion. The definition
of interaction types naturally varies according to usage. In
this paper, we mainly focus on the task-oriented interac-
tions that address task-related aspect. The other commu-
nicative actions that concern the meeting and the group
itself (e.g., when someone invited another participant to
take the floor) are not included. For generalizability, we
create a set of interaction types based on a standard
utterance-unit tagging scheme [27]: propose, comment,
acknowledgement, requestInfo, askOpinion, posOpinion, and
negOpinion. The detailed meanings are as follows: propose-
a user proposes an idea with respect to a topic; comment—a
user comments on a proposal, or answers a question;
acknowledgement—a user confirms someone else’s comment
or explanation, e.g., “yeah,” “uh huh,” and “OK;” reques-
tInfo—a user requests unknown information about a topic;
askOpinion-a user asks someone else’s opinion about a
proposal; posOpinion—a user expresses a positive opinion,
i.e., supports a proposal; and negOpinion—a user expresses
a negative opinion, i.e., disagrees with a proposal.
STUDIES
Data Set
Our studies involve four real meetings, lasting 20 minutes
on average. Multiple devices, such as video cameras,
microphones, and motion sensors, were used for capturing
the meetings. The four meetings include one PC purchase
meeting (26 min, discussing PCs to be ordered for the
laboratory, such as types, configuration, size, weight,
manufacturer, etc.), one trip-planning meeting (18 min,
discussing time, place, activities, and transportation for a
summer trip), one soccer preparation meeting (23 min,
talking about the players and their roles and positions in
an upcoming match), and one job selection meeting
(10 min, talking about factors that will be considered in
seeking a job, such as salary, working place, employer,
position, interest, etc.). Each meeting had four participants
seated around a table. Human interactions were detected
by using a multimodal approach [28]. In order to use a
correct data for mining.
CONCLUSION
We proposed a tree-based mining method for discovering
frequent patterns of human interaction in meeting discus-
sions. The mining results would be useful for summariza-
tion, indexing, and comparison of meeting records. They
also can be used for interpretation of human interaction in
meetings. In the future, we will develop several applica-
tions based on the discovered patterns. We also plan to
explore embedded tree mining for hidden interaction
pattern discovery. Embedded subtrees are a generalization
of induced subtrees, which allow not only direct parent-
child branches, but ancestor-descendant branches [3]. For
example, when there is an interaction of propose, there
always follows a comment, directly or indirectly. Finally, we
plan to incorporate more meeting content in both amount
and category.