HIERARCHICAL ORGANIZATION FOR DATA USING PARTIAL LEAST SQUARE ANALYSIS(PLS-TREE)

PROBLEM TO BE SOLVED: To perform clustering of large amounts of data in a relatively short processing time.SOLUTION: A method includes supplying a first data matrix and a second data matrix, each of the first and second data matrixes including one or more variables and a plurality of data points. Al...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: SVANTE BJARNE WOLD, TRYGG JOHAN, LENNART ERIKSSON
Format: Patent
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:PROBLEM TO BE SOLVED: To perform clustering of large amounts of data in a relatively short processing time.SOLUTION: A method includes supplying a first data matrix and a second data matrix, each of the first and second data matrixes including one or more variables and a plurality of data points. Also, this method includes: determining a first score from the first data matrix by using partial least square (PLS) analysis or orthogonal PLS(OPLS) analysis; and clustering the first and second data matrixes into a first group and a second group (by, for example, in line unit), the clustering being executed on the basis of the distribution of the first and second groups related to the sorted first score, the distribution of the first data matrix and the distribution of the first and second data matrixes.