SDM Methods via fpfs-Matrices: Benchmark of 35 Algorithms

35 soft decision-making (SDM) algorithms based on fuzzy parameterized fuzzy soft matrices (fpfs-matrices) were benchmarked across ten datasets from the UCI Machine Learning Repository. The evaluation, conducted by Ömer Karakoç and last updated in May 2026, used metrics like accuracy, precision, recall, specificity, and F1-score, with statistical significance confirmed via Friedman and Nemenyi tests. The dataset, shared under a CC-BY-4.0 license, identifies top-performing algorithms such as A19 and YHX14.

Use Cases

• Benchmarking classification algorithm performance based on the reported accuracy, precision, recall, and F1-scores.
• Selecting a soft decision-making method for a new project based on the ranked list of 35 algorithms.
• Studying the application of fuzzy parameterized fuzzy soft matrices (fpfs-matrices) in machine learning contexts.
• Conducting statistical comparison of algorithms using the Friedman and Nemenyi test methodology described.

Strengths

• Benchmarks a specific set of 35 algorithms, providing a direct numerical comparison.
• Evaluation uses multiple standard metrics (accuracy, precision, recall, specificity, F1-score) and statistical tests.
• Methods were tested on ten diverse datasets from the UCI Machine Learning Repository.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

UCI Machine Learning Repository (for benchmark datasets).

Collection Method

Algorithm evaluation and benchmarking using the FPFS-CMC classifier.

Freshness

Last updated 2026-05-13 17:44:26; freshness should be verified.