The paper presents difficulties of measuring clustering quality for symbolic data (such as lack of a "traditional" data matrix). Some hints concerning the usage of known indexes for such kind of data are given and indexes designed exclusively for symbolic data are described. Finally, after the presentation of simulation results, some proposals for choosing the most adequate indexes for popular classification algorithms are given.
CITATION STYLE
Dudek, A. (2007). Cluster quality indexes for symbolic classification - An examination. In Studies in Classification, Data Analysis, and Knowledge Organization (pp. 31–38). Kluwer Academic Publishers. https://doi.org/10.1007/978-3-540-70981-7_4
Mendeley helps you to discover research relevant for your work.