SCPP: A Unified Python Library for Soft Clustering

2026-07-21

AI Summary

This work addresses the challenges of reproducibility, comparability, and practical deployment of soft clustering methods, which have been hindered by the absence of a unified interface and standardized evaluation framework. To bridge this gap, we introduce SCPP—an open-source Python library fully compatible with scikit-learn—that provides, for the first time, standardized interfaces for 40 soft clustering algorithms spanning fuzzy, probabilistic, graph-based, matrix factorization, and deep learning approaches. SCPP supports consistent training, prediction, membership output, and comprehensive evaluation. Built with object-oriented design, the library integrates automated testing, extensive documentation, and interoperability with mainstream scientific computing libraries. Furthermore, it includes a complete benchmarking platform featuring diverse datasets, quality metrics, and analyses of runtime, memory usage, and scalability, thereby significantly enhancing algorithm reproducibility and extensibility.

Abstract

In this paper, we present SCPP (Soft Clustering Python Package), an open-source Python framework for soft clustering. SCPP establishes a canonical, scikit-learn-compatible estimator interface that standardizes model training, prediction, membership representation, evaluation, and benchmarking across heterogeneous soft clustering methods, including fuzzy, probabilistic, graph-based, matrix factorization, and deep learning methods. The framework currently integrates 40 representative algorithms together with a comprehensive benchmarking comprising datasets, clustering quality metrics, and standardized runtime, memory, and scalability evaluation. SCPP further provides extensive documentation, practical examples, automated testing, and seamless integration with the scientific Python ecosystem, enabling reproducible experimentation and straightforward extension with new algorithms. The source code is publicly available at https://github.com/soft-clustering/soft-clustering.

Problem

• soft clustering
• unified framework
• standardization
• benchmarking
• heterogeneous methods

Innovation

• soft clustering
• unified framework
• scikit-learn compatibility
• benchmarking
• open-source library