Title
Extraction of Children Friendship Relations from Activity Level
Abstract
Children learn to fit into society by group interaction through which they are greatly influenced by their relations with friends. Although teachers need to observe them to assist their growth and social progress and support the development of their personalities, only experienced teachers can watch children while simultaneously providing high-quality guidance. To resolve this problem, this paper proposes a mathematical and objective method that assists teachers with observation. It uses the numerical data of activity levels recorded by pedometers from which we make tree diagrams called dendrograms based on hierarchical clustering with recorded activity levels. We also calculate the children's “breadth” and “depth” of friend relations using more than one dendrogram. After the authors recorded the children's activity levels at a kindergarten for two months to evaluate our proposed method, we found that the results generally coincided with teacher remarks and evaluations of the children.
Year
DOI
Venue
2008
10.3233/978-1-58603-900-4-368
JCKBSE
Keywords
Field
DocType
high-quality guidance,group interaction,experienced teacher,activity level,numerical data,recorded activity level,hierarchical clustering,objective method,children friendship relations,friend relation
Hierarchical clustering,Data mining,Lifelog,Friendship,Dendrogram,Computer science,Social progress,Mathematics education,Group interaction,Artificial intelligence,Cluster analysis,Personality psychology
Conference
Volume
ISSN
Citations 
180
0922-6389
0
PageRank 
References 
Authors
0.34
1
7
Name
Order
Citations
PageRank
Shigeo Kaneda16926.85
Takuya Katsuki200.34
Masahiro Shibata300.34
Hirohide Haga4247.34
Mari Ueda5155.36
Aki Kono601.01
Kimio Shintani701.35