Rezumat articol ediţie STUDIA UNIVERSITATIS BABE┼×-BOLYAI

În partea de jos este prezentat rezumatul articolului selectat. Pentru revenire la cuprinsul ediţiei din care face parte acest articol, se accesează linkul din titlu. Pentru vizualizarea tuturor articolelor din arhivă la care este autor/coautor unul din autorii de mai jos, se accesează linkul din numele autorului.

 
       
         
    STUDIA INFORMATICA - Ediţia nr.1 din 2016  
         
  Articol:   DISCOVERING PATTERNS IN DATA USING ORDINAL DATA ANALYSIS.

Autori:  HORIA F. POP.
 
       
         
  Rezumat:  VIEW PDF: DISCOVERING PATTERNS IN DATA USING ORDINAL DATA ANALYSIS

Discovering patterns in data is becoming more and more important for different fields of research. The analysis of ordinal data is sensitive and requires special attention. In order to analyze ordinal data, we may use various criteria. In our paper, we present a solution by using different linkage criteria (ward, median, centroid, weighted, complete and single linkage method) with agglomerative clustering algorithms. To evaluate and interpret our results we have considered some internal and external evaluation indexes for clustering (also known as cluster analysis). The experiments reveal different comparative results. To validate our clustering results, we used pair-counting measures (Jaccard, Recall, Rand and Fowlkes-Mallows indexes), BCubed-based measures (F1-Measure), set- matching-based measures and editing-distance measures (Purity, Precision and Recall) for external evaluation and Silhouette index for analyzing intrinsic characteristics of a clustering (internal evaluation). The comparative experiments for different linkage methods suggest that for an ordinal data set, by using ward linkage methods we achieve more accurate results in terms of cluster validity than others linkage criteria applied to our data set.

Keywords and phrases: ordinal data analysis, agglomerative clustering.

2010 Mathematics Subject Classification. 68T10, 62H30, 62-07.
 
         
     
         
         
      Revenire la pagina precedentă