BIP! Ranker可大规模计算论文影响力多维度指标
BIP! Ranker: A Software Library for Citation-Based Impact Indicators on Large-Scale Graphs
- 基于Spark构建，支持百亿级引用图计算
- 同时产出总影响、当前热度、早期势头等多维指标
- 开源工具，适合学术分析与科研评价研究者使用
科学影响力具有多维性：整体影响、当前热度、早期引用势头以及领域内相对表现各自反映出版物影响力的特定方面。然而在实践中，这些维度常被简化为单一指标（如引用次数）。目前针对大型引文网络的多维互补影响力指标计算工具仍稀缺，尤其难以处理像主要学术数据库提供的那种包含数十亿条引用、数亿篇文献的复杂引文图。我们提出 BIP! Ranker，一个基于 Spark 的开源软件库，可在大规模图上高效计算多种基于引用的影响力指标，支持处理包含百亿级引用、数亿篇论文的引文网络。
原文摘要 · Abstract (English)
Scientific impact is multidimensional: overall influence, current popularity, early citation momentum, and field-relative performance each capture a distinct facet of a publication's impact. Yet, in practice, these dimensions are often reduced to a single metric, such as citation count. Open solutions for computing multiple complementary impact indicators at scale remain scarce, particularly for citation graphs as large as those provided by major scholarly databases. We introduce BIP! Ranker, an open-source, Spark-based library for computing citation-based impact indicators at scale, capable of processing citation networks with billions of citations among hundreds of millions of publications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。