Ranking Single fpfs-Matrices: Benchmarking 35 Soft Decision-Making Algorithms

A benchmark study evaluates 35 soft decision-making algorithms based on fuzzy parameterized fuzzy soft matrices (fpfs-matrices). The algorithms were assessed using the FPFS-CMC classifier across ten datasets from the UCI Machine Learning Repository, with performance measured by accuracy, precision, recall, specificity, and F1-score. The research, authored by Ömer Karakoç and last updated in May 2026, identifies top-performing algorithms like A19 and YHX14.

Use Cases

• Selecting a soft decision-making algorithm for classification tasks based on benchmarked performance metrics.
• Comparing the effectiveness of different fpfs-matrix-based methods for modeling uncertainty in machine learning.
• Validating new decision-support algorithms against a structured benchmark of 35 existing methods.
• Applying top-ranked algorithms like A19 or YHX14 to complex classification challenges involving uncertain data.

Strengths

• Benchmarks a substantial set of 35 distinct algorithms.
• Evaluates performance across ten diverse datasets from the UCI repository.
• Uses multiple established metrics (accuracy, precision, recall, specificity, F1-score) and statistical significance tests (Friedman, Nemenyi).
• Identifies specific top-performing algorithms (e.g., A19 Rank 1, YHX14 Rank 2).

Limitations

• Dataset size is very small at 5.5 KB, indicating limited raw data scope.
• Row count and column-level documentation are unknown, limiting detailed assessment of data structure.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

UCI Machine Learning Repository (for the ten benchmark datasets).

Collection Method

Algorithm evaluation using the FPFS-CMC classifier.

Freshness

Last updated 2026-05-13 17:44:39