融合大语言模型的区块链开源项目主题挖掘与演化分析

doi:10.3969/j.issn.0255-8297.2026.05.001

摘要/Abstract

摘要： 本文基于2021—2024年GitHub区块链开源项目，提出融合大语言模型与BERTopic （bidirectional encoder representations for topic）模型的开源项目主题识别框架：依据大语言模型提示词工程对开源项目进行特定数据预处理与文本精炼，并通过词嵌入、降维UMAP（uniform manifold approximation and projection）以及基于层次密度的空间聚类算法（带噪声）构建BERTopic模型以实现区块链主题识别，然后采用结合“大语言模型主题表示和命名”以及“专家评判与优化”的方式生成区块链开源项目主题标签。在得到主题识别结果的基础上，构建“主题强度-技术影响力-再创活跃度”三维指标体系进行主题评价和技术主题演化特征分析。研究表明，区块链开源生态呈现出“基础巩固-应用扩展-智能融合”的演化路径。除了识别出围绕智能合约、去中心化金融、非同质化代币等方面展开的常见技术主题外，还发现“AI+区块链”相关技术主题正成为新的增长点。多维指标评估显示，开发者社区在追随技术和市场热点的同时，持续夯实安全基础与工具链建设。

关键词: 区块链, 主题挖掘, 主题演化, 开源项目, GitHub, 大语言模型

Abstract: Based on GitHub blockchain open-source projects from 2021 to 2024, this paper proposed a topic identification framework for open-source projects that integrates a large language model with bidirectional encoder representations for topic (BERTopic). The framework first applied large language model prompt engineering to perform specific data preprocessing and text refinement for open-source projects. It then constructed a BERTopic model through word embedding, dimensionality reduction using uniform manifold approximation and projection (UMAP), and hierarchical density-based spatial clustering of applications with noise (HDBSCAN) to identify blockchain topics. Finally, topic labels for blockchain open-source projects were generated by combining “topic representation and naming using a large language model” with “expert evaluation and optimization”. Based on the topic identification results, a three-dimensional index system comprising “topic intensity-technological influence-secondary development activity” was constructed to evaluate the identified topics and analyze their evolutionary characteristics. The research shows that the blockchain open-source ecosystem presents an evolutionary path of “foundation consolidation-application expansion-intelligent integration”. In addition to identifying common technical topics around smart contracts, decentralized finance (DeFi), non-fungible tokens (NFTs), etc., the study also finds that “AI + blockchain”-related technical topics are becoming new growth points. The multidimensional index evaluation shows that the developer community continues to strengthen security foundations and toolchain development while keeping pace with technological and market trends.

Key words: blockchain, topic mining, topic evolution, open-source projects, GitHub, large language model

中图分类号:

TP391

饶芮瑞, 朱莉, 颜嘉麒. 融合大语言模型的区块链开源项目主题挖掘与演化分析[J]. 应用科学学报, 2026, 44(5): 701-715.

RAO Ruirui, ZHU Li, YAN Jiaqi. Topic Mining and Evolution Analysis of Blockchain Open-Source Projects Integrating Large Language Model[J]. Journal of Applied Sciences, 2026, 44(5): 701-715.

[1] Nakamoto S. Bitcoin: a peer-to-peer electronic cash system [EB/OL]. [2026-02-23]. https://bitcoin.org/bitcoin.pdf, 2008.
[2] 代闯闯, 栾海晶, 杨雪莹, 等. 区块链技术研究综述[J]. 计算机科学, 2021, 48(增刊2): 500-508. Dai C C, Luan H J, Yang X Y, et al. A review of blockchain technology research [J]. Computer Science, 2021, 48(Suppl.2): 500-508(in Chinese).
[3] Chen J E, Bao F, Li C, et al. The application and ethics of artificial intelligence in blockchain: a bibliometric-content analysis [J]. Journal of Global Information Management, 2023, 31(7): 1-32.
[4] Kalliamvakou E, Gousios G, Blincoe K, et al. An in-depth study of the promises and perils of mining GitHub [J]. Empirical Software Engineering, 2016, 21(5): 2035-2071.
[5] Park Y, Lim S, Gu C, et al. Forecasting topic trends of blockchain utilizing topic modeling and deep learning-based time-series prediction on different document types [J]. Journal of Informetrics, 2025, 19(2): 101639.
[6] Mafrur R. Blockchain data analytics: review and challenges [PP/OL]. V1. (2025-03-12) [2026- 02-23]. https://doi.org/10.48550/arXiv.2503.09165.
[7] Vaccargiu M, Aufiero S, Bartolucci S, et al. Sustainability in blockchain development: a BERTbased analysis of Ethereum developer discussions [C]//28th International Conference on Evaluation and Assessment in Software Engineering, 2024: 381-386.
[8] Ibba G, Vaccargiu M. Analysis of users’ most discussed topics and trends on blockchain technologies and smart contracts [C]//2023 IEEE International Conference on Software Analysis, Evolution and Reengineering, 2023: 865-873.
[9] 汪园, 王学东, 李金鑫. 基于文献计量的我国区块链研究的知识网络与结构分析[J]. 现代情报, 2018, 38(1): 147-153. Wang Y, Wang X D, Li J X. Knowledge network and structure analysis of blockchain research in China based on bibliometrics [J]. Journal of Modern Information, 2018, 38(1): 147-153. (in Chinese).
[10] 罗棋, 闵超, 颜嘉麒, 等. 国际区块链研究主题挖掘及演化分析[J]. 现代情报, 2021, 41(9): 157-166. Luo Q, Min C, Yan J Q, et al. Mining and evolution analysis of international blockchain research topics [J]. Journal of Modern Information, 2021, 41(9): 157-166. (in Chinese).
[11] Grootendorst M. BERTopic: neural topic modeling with a class-based TF-IDF procedure [PP/OL]. V1(2022-03-11) [2026-02-23]. https://doi.org/10.48550/arXiv.2203.05794.
[12] Abdelrazek A, Eid Y, Gawish E, et al. Topic modeling algorithms and applications: a survey [J]. Information Systems, 2023, 112: 102131.
[13] 温晓波, 姚长青, 高影繁. 大语言模型增强的科技文献主题识别[J]. 情报理论与实践, 2025, 48(9): 185-193. Wen X B, Yao C Q, Gao Y F. Topic identification of scientific and technological literature enhanced by large language models [J]. Information Studies: Theory & Application, 2025, 48(9): 185-193. (in Chinese)
[14] Brown T, Mann B, Ryder N, et al. Language models are few-shot learners [J]. Advances in Neural Information Processing Systems, 2020, 33: 1877-1901.
[15] Han S, Shin M, Park S, et al. Unified neural topic model via contrastive learning and term weighting [C]//17th Conference of the European Chapter of the Association for Computational Linguistics. 2023: 1802-1817.
[16] Yang X, Zhao H, Xu W, et al. Neural topic modeling with large language models in the loop [C]// 63rd Annual Meeting of the Association for Computational Linguistics, 2025: 1377-1401.
[17] 张凯, 杨敏纳, 隗玲. 融合Finetuned-BERTopic和大模型的技术主题识别方法研究[J]. 情报理论与实践, 2025, 48(3): 189-198. Zhang K, Yang M N, Wei L. Research on technology topic identification method integrating Finetuned-BERTopic and large model [J]. Information Studies: Theory & Application, 2025, 48(3): 189-198. (in Chinese)
[18] Sharma A, Thung F, Kochhar P S, et al. Cataloging GitHub repositories [C]// 21st International Conference on Evaluation and Assessment in Software Engineering, 2017: 314-319.
[19] 李舟军, 范宇, 吴贤杰. 面向自然语言处理的预训练技术研究综述[J]. 计算机科学, 2020, 47(3): 162-173. Li Z J, Fan Y, Wu X J. A review of pre-training technologies for natural language processing [J]. Computer Science, 2020, 47(3): 162-173. (in Chinese)
[20] Mcinnes L, Healy J, Astels S. HDBSCAN: hierarchical density based clustering [J]. Journal of Open Source Software, 2017, 2(11): 205.
[21] 蔡鑫, 靳军宝, 王政媛, 等. 融合主题语义与转移概率的“科学—技术” 主题演化路径识别研究[J]. 情报理论与实践, 2026, 49(1): 138-149. Cai X, Jin J B, Wang Z Y, et al. Research on identification of “science-technology” topic evolution path integrating topic semantics and transition probability [J]. Information Studies: Theory & Application, 2026, 49(1): 138-149. (in Chinese)
[22] Zhang Y, Chen M, Liu L. A review on text mining [C]//20156th IEEE International Conference on Software Engineering and Service Science, 2015: 681-685.
[23] Reuter A, Khadka B, Thielmann A, et al. Gptopic: dynamic and interactive topic representations [PP/OL]. V1(2024-03-06) [2026-02-23]. https://doi.org/10.48550/arXiv.2403.03628.
[24] 许海云, 龚兵营, 杨俊浩, 等. 新兴研究主题识别方法研究进展与前瞻[J]. 图书情报工作, 2025, 69(3): 135-150. Xu H Y, Gong B Y, Yang J H, et al. Research progress and prospect of emerging research topic identification methods [J]. Library and Information Service, 2025, 69(3): 135-150. (in Chinese)
[25] Github Documentation. Saving repositories with stars [EB/OL]. (2016-10-19) [2026-02-23]. https://help.github.com/articles/about-stars/.
[26] Bhatia A, Eghan E E, Grichi M, et al. Towards a change taxonomy for machine learning systems [PP/OL]. V1(2022-03-21) [2026-02-23]. https://doi.org/10.48550/arXiv.2203.11365.
[27] 马海群, 崔文波, 张涛. 我国数据安全政策文本主题挖掘及其演化分析[J]. 现代情报, 2024, 44(8): 28-38. Ma H Q, Cui W B, Zhang T. Topic mining and evolution analysis of China’s data security policy texts [J]. Journal of Modern Information, 2024, 44(8): 28-38. (in Chinese)
[28] 秦晓慧, 乐小虬. 基于LDA主题关联过滤的领域主题演化研究[J]. 现代图书情报技术, 2015, (3): 18-25. Qin X H, Le X Q. Research on domain topic evolution based on LDA topic correlation filtering [J]. New Technology of Library and Information Service, 2015(3): 18-25. (in Chinese)
[29] Izadi M, Heydarnoori A, Gousios G. Topic recommendation for software repositories using multilabel classification algorithms [J]. Empirical Software Engineering, 2021, 26(5): 93.

融合大语言模型的区块链开源项目主题挖掘与演化分析

Topic Mining and Evolution Analysis of Blockchain Open-Source Projects Integrating Large Language Model

PDF

可视化

摘要/Abstract

引用本文

使用本文

相关文章 15

编辑推荐

Metrics

本文评价

[1]	云健, 刘涛, 高新茹, 张海玉. 一种基于区块链的个性化联邦边缘学习保护框架[J]. 应用科学学报, 2026, 44(5): 731-751.
[2]	郭丽, 张文杰, 潘恒, 陆思奇, 邢颖, 李坤阳, 张博伟, 斯雪明. LauHunter：一种快速高效的洗钱检测工具[J]. 应用科学学报, 2026, 44(5): 716-730.
[3]	孟令洋, 李鹏, 孙丽, 杜剑. 基于深度强化学习的分片区块链节点多维评估和智能动态重构方案[J]. 应用科学学报, 2026, 44(5): 752-770.
[4]	李熙, 文斌, 王怡蓓. 基于零知识证明的可信数据质量评估[J]. 应用科学学报, 2026, 44(5): 771-786.
[5]	彭锦鸿, 陈茂霖, 薛梅, 李汝峰. 森林点云的三维场景图表达与LLM应用[J]. 应用科学学报, 2026, 44(4): 644-656.
[6]	周飞宇, 张星. 融合视觉场景理解与常识推理的零样本目标导航[J]. 应用科学学报, 2026, 44(4): 615-625.
[7]	杨松涛, 高盛祥, 欧阳飞, 余正涛. 上下文感知与模态对齐增强的语音翻译方法[J]. 应用科学学报, 2026, 44(4): 571-584.
[8]	李守伟, 蒋易珉, 张嘉政. 多源异质数据资产分链交易体系设计[J]. 应用科学学报, 2026, 44(3): 465-485.
[9]	斯雪明, 马硕森, 姚中原. 适用于物联网环境的多维并行区块链[J]. 应用科学学报, 2026, 44(2): 282-296.
[10]	陈强斌, 姚中原, 田浩, 斯雪明. BalChain：一种基于信誉与负载的分片区块链系统[J]. 应用科学学报, 2025, 43(5): 863-876.
[11]	何旭荣, 方有轩, 郑旭晓, 辛艳双. 基于预言机组件技术的跨链系统研究与实现[J]. 应用科学学报, 2025, 43(4): 630-642.
[12]	潘璇, 张抗抗, 程澳. 一种基于区块链的数据要素精准授权机制[J]. 应用科学学报, 2025, 43(4): 600-616.
[13]	张壮, 邹义林, 林泽鹏, 刘家圆, 訾宗青, 谭良, 佘堃. 通用区块链跨链交易形式化定义及实例验证分析[J]. 应用科学学报, 2025, 43(4): 559-585.
[14]	刘丽丽, 时忆杰, 秦素娟. 基于机器学习的智能合约漏洞检测技术[J]. 应用科学学报, 2025, 43(4): 541-558.
[15]	兰亚杰, 马自强, 苗莉, 胡富森. 基于区块链的图像数字版权保护系统[J]. 应用科学学报, 2025, 43(2): 315-333.