Mining Tree Based Frequent Pattern from Human Interaction in Meeting

In modern life, interactions between human beings are frequently occurring in meetings, where topics are discussed and new decisions are made. To understand and interpret how people interact in a meeting discussion study of semantic knowledge is important. It becomes possible to extract frequent patterns of human interaction based on the captured content of face-to-face meetings by using mining method. Human interactions are categorized as proposing an idea, giving comments, expressing a positive opinion, acknowledgement; indicate the intention of user towards a topic or role in a discussion.Tree is used for representing a human interaction flow in a discussion session. Tree-based interaction mining algorithms are studied to analyze the structures of the trees and to extract interaction flow patterns. It can successfully extract several interesting patterns which is 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.