社交电商监管资源调度算法研究

doi:10.3778/j.issn.1002-8331.2310-0222

摘要/Abstract

摘要： 社交电商监管资源调度是一个具有挑战性的课题。随着社交平台及用户数量的迅速增长，商品质量的良莠不齐成为一个突出的问题。然而，全面细致的监测是一项艰巨的任务，而有限的监管人员难以主动巡查并发现违法违规行为。为了解决在众多电商平台之间优化分配监管资源的问题，提出了基于置信区间上界的三种调度算法，它们能够在利用历史数据选择最优巡查策略的基础上，展开新的探索以加强对欺诈行为的理解。算法采用了巡查行动组合编码和特征相似性简约方法，以提高监测性能。实验分析通过不同的行为模型、应用环境变量和内置参数的比对，验证了综合算法在性能指标上具有较大的优势，对于提升社交电商平台的监管效果具有重要的学术价值和实践指导意义。

关键词: 资源调度, 社交电商, 置信区间上界, 优化算法

Abstract: The issue of regulating resource scheduling on social e-commerce platforms is a challenging task. With the rapid growth of social e-commerce platforms and the number of users, the uneven quality of goods has become a prominent problem. However, comprehensive and meticulous monitoring is a daunting task, and limited regulatory personnel find it difficult to actively inspect and detect illegal activities. In order to optimize the allocation of regulatory resources among numerous e-commerce platforms, three scheduling algorithms based on the upper confidence bound have been proposed. These algorithms can select the optimal inspection strategy based on historical data and then explore new approaches to enhance understanding of fraudulent behavior. The algorithms utilize inspection action combination encoding and feature similarity simplification methods to improve monitoring performance. Through comparison of different behavioral models, application environment variables, and built-in parameters in the experimental analysis, the significant advantages of the comprehensive algorithm in performance indicators have been verified.This holds important academic value and practical guidance significance for enhancing the regulatory effectiveness of social e-commerce platforms.

Key words: resource scheduling, social e-commerce, upper confidence bound, optimization algorithm

彭可欣, 曾曦葶, 何灏. 社交电商监管资源调度算法研究[J]. 计算机工程与应用, 2025, 61(8): 315-323.

PENG Kexin, ZENG Xiting, HE Hao. Research on Resource Scheduling Algorithms for Social E-Commerce Regulation[J]. Computer Engineering and Applications, 2025, 61(8): 315-323.

参考文献

[1] SU Y, MA Z, YIN X, et al. Research on the impact of live commerce on starbucks consumer satisfaction-based on the moderating effect of supervision strength[J]. Journal of Asian Business Strategy, 2022, 12(2): 115-138.
[2] BO X. Supervision status and countermeasures of false propaganda of food in live streaming of e-commerce in China[J]. Journal of Food Science and Technology, 2021, 39(6): 22-27.
[3] HAN L, ZHANG Z, LI W, et al. Game analysis on supervision of quality problems of live goods with short video platform participation[J]. Open Journal of Business and Management, 2022, 10(3): 1274-1289.
[4] JIANG J, XUE P. Research on the supervision mechanism in live-stream selling market: based on evolutionary game[C]//Proceedings of the 4th International Conference on Economic Management and Cultural Industry, 2022: 1178-1192.
[5] ZHAO X, HUO Y. The evolutionary game analysis and simulation research on e-commerce live broadcast counterfeit sales with the participation of a third-party supervisor[C]//Proceedings of the 2nd International Conference on Education, Information Management and Service Science, 2022: 85-96.
[6] SHEN L, CHEN Y, FAN R, et al. Government supervision on explosive enterprises’ immoral behaviors in e-commerce enterprises: an evolutionary game analysis[J]. Complexity, 2021, 20(2): 1-11.
[7] ZHANG R, FANG L, HE X, et al. Supervision and optimization mechanism of e-commerce security management[M]. Cham: Springer, 2023: 281-295.
[8] DAI Z, CHEN Y, YU H, et al. On provably robust meta-Bayesian optimization[C]//Proceedings of the Uncertainty in Artificial Intelligence, 2022: 475-485.
[9] XIE G, ZHANG W, HU Z, et al. Upper confident bound advantage function proximal policy optimization[J]. Cluster Computing, 2023, 26(3): 2001-2010.
[10] SOUISSI B, GHORBEL A. Upper confidence bound integrated genetic algorithm optimized long short term memory network for click through rate prediction[J]. Applied Stochastic Models in Business and Industry, 2022, 38(3): 475-496.
[11] KHAJONCHOTPANYA N, XUE Y, RUJEERAPAIBOON N. A revised approach for risk-averse multi-armed bandits under CVAR criterion[J]. Operations Research Letters, 2021, 49(4): 465-472.
[12] BOGUNOVIC I, KRAUSE A. Misspecified gaussian process bandit optimization[C]//Proceedings of the 35th International Conference on Neural Information Processing Systems, 2021: 3004-3015.
[13] YAN C, HAN H, ZHANG Y, et al. Dynamic clustering based contextual combinatorial multi-armed bandit for online recommendation[J]. Knowledge-Based Systems, 2022, 257(8): 10-27.
[14] RUAN Y, YEKKEHKHANY A, ETESAMI S R. Online learning for job scheduling on heterogeneous machines[C]//Proceedings of the 59th IEEE Conference on Decision and Control, 2020: 591-596.
[15] BIAN C, WANG X, SHAO W, et al. Adaptive confidence bound based Bayesian optimization via potentially optimal Lipschitz conditions[J]. Engineering Optimization, 2022, 45(11): 1-19.
[16] LIU Q, CHENG R, JIN Y, et al. Reference vector-assisted adaptive model management for surrogate-assisted many-objective optimization[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2022, 52(12): 7760-7773.
[17] QIU S, WANG L, BAI C, et al. Contrastive UCB: provably efficient contrastive self-supervised learning in online reinforcement learning[C]//Proceedings of the International Conference on Machine Learning, 2022: 18168-18210.
[18] LIANG Y, HUANG C, BAO X, et al. Sequential dynamic event recommendation in event-based social networks: an upper confidence bound approach[J]. Information Sciences, 2021, 542(1): 1-23.
[19] WüTHRICH M, SCH?LKOPF B, KRAUSE A. Regret bounds for Gaussian-process optimization in large domains[C]//Proceedings of the 35th International Conference on Neural Information Processing Systems, 2021: 7385-7396.
[20] SHEN B, GNANASAMBANDAM R, WANG R, et al. Multi-task Gaussian process upper confidence bound for hyperparameter tuning and its application for simulation studies of additive manufacturing[J]. IISE Transactions, 2023, 55(5): 496-508.