University of British Columbia
Enhanced k-prototypes clustering for mixed data by bootstrap augmentation
Abstract
dc:descriptionThe k-prototypes algorithm is a popular approach for clustering mixed data, yet it faces challenges such as susceptibility to local optima and misclassification of boundary observations with no measure of uncertainty due to hard partitioning. Our proposal integrates bootstrap-augmented optimization with k-prototypes to address these issues: expanding the search space of the algorithm while simultaneously providing probabilistic estimates for cluster memberships. We demonstrate the utility of this approach through simulations and real-world data analyses.
Degree
thesis:*- Name thesis:degree_name
- Master of Science - MSc
- Level thesis:degree_level
- master's
- Discipline thesis:degree_discipline
- Mathematics
- Grantor dc:publisher
- University of British Columbia
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Chang, Fujia
- Contributors dc:contributor
-
- Andrews, Jeffrey L.
Rights
dc:rights- Statement dc:rights
-
- Attribution-NonCommercial-NoDerivatives 4.0 International
- Language dc:language
- eng
Identifiers
dc:identifier.*- Handle dc:identifier
- http://hdl.handle.net/2429/89326