| 90 | 0 | 23 |
| 下载次数 | 被引频次 | 阅读次数 |
【背景和目的】大户控制小户套购卷烟,关系到烟草专卖专营渠道管理,是烟草专卖监管研究的重点方向之一。本研究引入社会网络分析,旨在从异常关联网络中精准定位套购团伙及其关键主体,为从结构上打击地下违法卷烟交易网络提供新视角。【方法】以零售户使用相同ID订货为基础构建零售户异常关联网络,采用Louvain算法开展社团检测,并结合网络拓扑结构与社团属性,以识别关联网络中的核心节点、中介节点及潜在团伙。【结果】通过历史涉案数据对比验证分析,两类节点的涉案率分别达14.3%和20.7%,显著高于随机节点。节点移除实验表明,核心节点移除可使社团密度最大降低36.53%;中介节点移除导致社团最多碎裂为12个孤立子图。【结论】实验证明,本研究提出的通过Louvain分析算法,能有效揭示和瓦解大户控制小户套购卷烟网络结构,为烟草专卖监督提供了高效、可量化的理论和技术支持。
Abstract:[Background] The manipulation of small retailers by large retailers to scalp cigarettes is closely related to the management of tobacco monopoly and exclusive distribution channels, and it is one of the key research directions in tobacco monopoly supervision. This study introduces social network analysis, aiming to accurately locate cigarette scalping syndicates and their key entities from abnormal association networks, and provide a new perspective for structurally combating underground illegal cigarette trading networks. [Methods] An abnormal association network of retailers was constructed based on retailers using the same ID for cigarette ordering. The Louvain algorithm was employed to conduct community detection. By integrating network topology and community attributes, core nodes, betweenness nodes, and potential gangs within the association network were identified. [Results] Through comparative verification and analysis of historical case-related data, the case involvement rates of the two types of nodes reached 14.3% and 20.7% respectively, which were significantly higher than those of randomly selected nodes. The node removal experiments demonstrated that removal of core nodes could reduce the community density by a maximum of 36.53%; the removal of betweenness nodes led to the fragmentation of the community into a maximum of 12 isolated subgraphs. [Conclusion] Experiments confirm that the method based on the Louvain analysis algorithm proposed in this study can effectively reveal and disrupt the network structure where large retailers control small retailers to scalp cigarettes, providing efficient and quantifiable theoretical and technical support for tobacco monopoly supervision.
[1] 陈超. 新形势下我国烟草行业管理体制创新研究[J]. 财经界, 2024, (16):54-56. CHEN Chao. Research on the innovation of the management system of China's tobacco industry under the new situation[J]. Money China, 2024, (16):54–56.
[2] 唐启孝. 浅论真烟非法流通的成因及对策[J]. 中国烟草学报, 2015, 21(S1): 72–76.
[3] 马运泉. 新形势下真烟异常流动的治理策略研究[J]. 产业与科技论坛, 2024, 23(22): 230–232.
[4] 于闽, 何维, 余瑶, 等. 基于行为数据的零售户卷烟流动风险预警分析[J]. 数字技术与应用, 2025, 43(08): 174–176.
[5] 章惠民. 福建烟草商业系统专卖管理“数字化”监管体系构建[J]. 中国烟草学报, 2020, 26(05): 123–128.
[6] Shashank S S, Samya M, Shivansh M, et al. Social network analysis: a survey on process, tools, and application[J]. ACM Computing Surveys, 2024, 56(8): 1–39.
[7] 李立耀, 孙鲁敬, 杨家海. 社交网络研究综述[J]. 计算机科学, 2015, 42(11): 8–21+42.
[8] Khatoon M, Banu W A. A survey on community detection methods in social networks[J]. International Journal of Education & Management Engineering, 2015, 5(1): 8–18.
[9] 马方. 犯罪网络分析:社会网络分析在有组织犯罪研究中的应用[J]. 西南政法大学学报, 2012, 14(02): 34–43.
[10] 谢秦川. 非法交易犯罪团伙的社会网络分析研究[J]. 信息网络安全, 2014, (06): 88–91.
[11] 牛惊雷, 王艳红. 社会网络分析在互联网洗钱犯罪侦查中的运用[J]. 公安学刊(浙江警察学院学报), 2020, (06): 41–45.
[12] Blondel V D, Guillaume J L, Lambiotte R, et al. Fast unfolding of communities in large networks[J]. Journal of Statistical Mechanics Theory & Experiment, 2008.
基本信息:
中图分类号:TP18;F721
引用信息:
[1]陈秉恒,陈星融,吴洵洵,等.基于 Louvain 算法对大户控制小户套购卷烟异常关联网络分析[J].中国烟草学报().
基金信息:
广东烟草内部专卖管理监督智能化升级及实践(2023440000200123)
2026-04-16
2026-04-16
2026-04-16