Title :
Visualizing similar text documents based on 3D dendrogram
Author :
Kinoshita, Tomohito ; Ohkubo, Tomoyuki ; Kobayashil, Kazuyuki ; Watanabe, Kajiro ; Kurihara, Yosuke
Author_Institution :
Fac. of Eng., Hosei Univ., Tokyo, Japan
Abstract :
This paper describes a new data visualization method using a three-dimensional dendrogram to And the relationship between similar text documents. For the detection of similarities in documents, we introduce two basic evaluation functions, "LCS" and "SED." We constructed and developed the proposed algorithm in MATLAB language, and conducted a comparison measurement and examined the evaluations. The validity of the proposed method can be verified by applying actual documents to demonstrate the 3D visualization.
Keywords :
data visualisation; mathematics computing; text analysis; LCS; MATLAB language; SED; data visualization method; longest common subsequence; shortest edit distance; similar text document visuallization; three-dimensional dendrogram; Algorithm design and analysis; Clustering algorithms; Data visualization; Detection algorithms; Software algorithms; Three dimensional displays; Visualization; 3D visualization; LCS; SED; dendrogram;
Conference_Titel :
SICE Annual Conference 2010, Proceedings of
Conference_Location :
Taipei
Print_ISBN :
978-1-4244-7642-8
