Tree-Based Weighted Interesting Pattern Mining Approach for Human Interaction Pattern Discovery
S. Uma, J. Suguna · International Review on Computers and Software (IRECOS) · 2013
Human Interaction is an essential aspect to recognize communicative data in various applications including medical diagnosis and human-computer interaction. Mining human Interaction in meetings is very helpful to determine reactions of persons in different scenarios. Behavior denotes the nature of the person and mining helps to examine the exhibition of one’s opinion. This research work presents a novel technique to mine frequent patterns of human interaction based on the meetings. In this proposed approach, human interaction flow in a discussion session is denoted as a tree. Tree-based interaction mining of Weighted Interesting Pattern mining algorithms are designed to analyze the structures of the trees and to extract interaction flow patterns. The experimental results shows that several interesting patterns are efficiently extracted which 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 interaction.