Bi-Partitioned Feature-Weighted K-Means Clustering for Detecting Insurance Fraud Claim Patterns
Bibliographic record
Abstract
The weighted K-means clustering algorithm is widely recognized for its ability to assign varying importance to features in clustering tasks. This paper introduces an enhanced version of the algorithm, incorporating a bi-partitioning strategy to segregate feature sets, thus improving its adaptability to high-dimensional and heterogeneous datasets. The proposed bi-partition weighted K-means (BPW K-means) clustering approach is tailored to address challenges in identifying patterns within datasets with distinct feature subspaces, such as those in insurance claim fraud detection. Experimental evaluations on real-world insurance datasets highlight significant improvements in both clustering accuracy and interpretability compared to the classical K-means, achieving an accuracy of approximately 91%, representing an improvement of about 38% over the classical K-means algorithm. Moreover, the method’s ability to uncover meaningful fraud-related clusters underscores its potential as a robust tool for fraud detection. Beyond insurance, the proposed framework applies to diverse domains where data heterogeneity demands refined clustering solutions. The application of the BPW K-means method to multiple real-world datasets highlights its clear superiority over the classical K-means algorithm.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".