ParClusterers Benchmark Suite (PCBS)
收藏资源简介:
ParClusterers Benchmark Suite (PCBS) 是一个用于评估和比较可扩展图聚类算法的高质量基准套件。该数据集由麻省理工学院和Google的研究人员创建,包含多种图聚类算法和工具,适用于社区检测、分类和密集子图挖掘等任务。PCBS不仅提供了多种图聚类算法的实现,还支持与其他图聚类框架的集成,便于研究人员进行系统性的性能评估。数据集的创建过程包括从SNAP库和新的空间及嵌入数据集中生成图数据,旨在解决大规模图聚类算法的性能和质量评估问题。
ParClusterers Benchmark Suite (PCBS) is a high-quality benchmark suite for evaluating and comparing scalable graph clustering algorithms. Developed by researchers from the Massachusetts Institute of Technology (MIT) and Google, PCBS includes a variety of graph clustering algorithms and tools applicable to tasks such as community detection, classification, and dense subgraph mining. PCBS not only provides implementations of multiple graph clustering algorithms but also supports integration with other graph clustering frameworks, enabling researchers to conduct systematic performance evaluations. The creation process of this benchmark suite generates graph data from the SNAP library and newly developed spatial and embedding datasets, aiming to address the performance and quality evaluation issues of large-scale graph clustering algorithms.

- 1The ParClusterers Benchmark Suite (PCBS): A Fine-Grained Analysis of Scalable Graph Clustering麻省理工学院 · 2024年



