WE INTRODUCE A NEW APPROACH TO CLUSTERING CATEGORICAL DATA: CONDORCET CLUSTERING WITH A FIXED NUMBER OF GROUPS, DENOTED ?-CONDORCET. AS K-MODES, THIS APPROACH IS ESSENTIALLY BASED ON SIMILARITY AND DISSIMILARITY MEASURES. THE PAPER IS DIVIDED INTO THREE PARTS: FIRST, WE PROPOSE A NEW CONDORCET CRITERION, WITH A FIXED NUMBER OF GROUPS (TO SELECT CASES INTO CLUSTERS). IN THE SECOND PART, WE PROPOSE A HEURISTIC ALGORITHM TO CARRY OUT THE TASK. IN THE THIRD PART, WE COMPARE ?-CONDORCET CLUSTERING WITH K-MODES CLUSTERING. THE COMPARISON IS MADE WITH A QUALITY?S INDEX, ACCURACY OF A MEASUREMENT, AND A WITHIN-CLUSTER SUM-OF-SQUARES INDEX. OUR FINDINGS ARE ILLUSTRATED USING REAL DATASETS: THE FELINE DATASET AND THE US CENSUS 1990 DATASET.

WE INTRODUCE A NEW APPROACH TO CLUSTERING CATEGORICAL DATA: CONDORCET CLUSTERING WITH A FIXED NUMBER OF GROUPS, DENOTED A-CONDORCET. AS K-MODES, THIS APPROACH IS ESSENTIALLY BASED ON SIMILARITY AND DISSIMILARITY MEASURES. THE PAPER IS DIVIDED INTO THREE PARTS: FIRST, WE PROPOSE A NEW CONDORCET CRITERION, WITH A FIXED NUMBER OF GROUPS (TO SELECT CASES INTO CLUSTERS). IN THE SECOND PART, WE PROPOSE A HEURISTIC ALGORITHM TO CARRY OUT THE TASK. IN THE THIRD PART, WE COMPARE ?-CONDORCET CLUSTERING WITH K-MODES CLUSTERING. THE COMPARISON IS MADE WITH A QUALITY?S INDEX, ACCURACY OF A MEASUREMENT, AND A WITHIN-CLUSTER SUM-OF-SQUARES INDEX. OUR FINDINGS ARE ILLUSTRATED USING REAL DATASETS: THE FELINE DATASET AND THE US CENSUS 1990 DATASET.