Abstract
Synthetic data is essential for assessing clustering techniques, complementing and extending real data, and allowing for a more complete coverage of a given problem's space. In turn, synthetic data generators have the potential of creating vast amounts of data -- a crucial activity when real-world data is at premium -- while providing a well-understood generation procedure and an interpretable instrument for methodically investigating cluster analysis algorithms. Here, we present \textit{Clugen}, a modular procedure for synthetic data generation, capable of creating multidimensional clusters supported by line segments using arbitrary distributions. \textit{Clugen} is open source, 100\% unit tested and fully documented, and is available for the Python, R, Julia and MATLAB/Octave ecosystems. We demonstrate that our proposal is able to produce rich and varied results in various dimensions, is fit for use in the assessment of clustering algorithms, and has the potential to be a widely used framework in diverse clustering-related research tasks.
Abstract (translated)
合成数据对于评估聚类技术、补充和扩展真实数据、以及更完整地覆盖给定问题的空间至关重要。另一方面,合成数据生成器有潜力生成大量数据 -- 在真实数据稀缺的情况下,这是一个重要的活动 -- 同时提供一种易于理解的生成程序和一个可解释的工具,用于深入研究聚类分析算法。在这里,我们介绍了 extit{Clugen},它是一个模块化的程序,用于合成数据生成,能够使用任意分布支持多维聚类,它是开源的、100%单元测试并全面文档化的,适用于Python、R、Julia和MATLAB/Octave生态系统。我们证明,我们的提议能够在各种维度上产生丰富和多样化的结果,适用于评估聚类算法,并且有潜力成为各种聚类相关研究任务中广泛使用的框架。
URL
https://arxiv.org/abs/2301.10327