Answers are grounded in this paper and cite their sources.
Empirical study reveals class cardinality increases prediction uncertainty in e-commerce customer analytics, highlighting the need for probabilistic evaluation metrics.
Class imbalance and class cardinality both affect multiclass classification but their influence on probabilistic estimation has been less explored. This study examines these impacts using an e-commerce dataset. Multiclass classification tasks involving 4, 6, and 7 classes are evaluated using supervised classifiers like Support Vector Machine, Gaussian Naive Bayes, Logistic Regression, Random Forest and Decision Tree. Under a 5 fold cross-validation the performance of the tasks is assessed using macro F1 score, log-loss and accuracy. Results show macro F1 score and accuracy decrease as class cardinality increases causing greater classification difficulty. Tree based models like Random Forest exhibit more balanced performance across classes. Gaussian Naive Bayes obtains the lowest log-loss indicating more accurate and reliable probability estimations. Class cardinality effects is isolated by varying the number of classes as we keep the features and classifier fixed. Increasing class cardinality reduced posterior confidence and increased entropy and log-loss. Using real time categorical variables these trends are confirmed. Next experiments done showed that class imbalance primarily affects minority class performance, whereas class cardinality exerts a broader influence on probabilistic confidence and prediction uncertainty. The findings highlight the need to consider class cardinality, class imbalance, and probabilistic metrics when evaluating multiclass classification models.
No takes yet. Share an insight, caveat, or question.
Ghosh et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: