生成式AI赋能通信大数据课程建设研究

Research on the Construction of Communication Big Data Courses Empowered by Generative AI

摘要: 5G、5G-A及云网融合的规模化落地，使通信网络大数据呈现出多源、海量、强关联等特征。然而，当前高职通信大数据课程长期存在通信与大数据教学失衡问题：或偏重通用大数据技术而脱离通信行业，或侧重通信业务而弱化数据分析能力培养，导致复合型网规网优人才培养效果不佳。立足职业教育导向，本文以移动通信网络规划优化岗位的真实工作流程为脉络，引入生成式AI对课程进行系统性重构。研究围绕课程目标、教学内容、授课模式、教学资源、考核评价及教师专业能力六大维度搭建改革方案，确立了“通信场景驱动、业务流程可视、项目任务贯穿、生成式AI深度赋能”的课程架构；构建了覆盖数据采集、数据治理、指标分析、可视化展示、问题排查及优化方案制定的完整项目教学体系；并提出了AI助教、AI辅助编程、动态资源智能生成及全过程多元评价的实施路径。教学实践表明，生成式AI能有效提升资源建设效率、个性化辅导精准度与教学评价效能。为避免技术滥用引发学习思考退化与评价失真，AI教学应用须坚守四大原则：教师主导、涉密数据脱敏、AI输出人工核验及教学全流程留痕。

Abstract: With the large-scale deployment of 5G, 5G-A and cloud-network convergence, big data of communication networks features multi-source access, massive volume and strong correlation. Nevertheless, vocational colleges have long faced an imbalance between communication expertise and big data knowledge in their Communication Big Data courses. Some courses overemphasize general big data technologies divorced from the communication industry, while others focus excessively on communication services with insufficient training in data analysis capabilities, resulting in unsatisfactory cultivation of interdisciplinary talents specialized in mobile network planning and optimization. Centered on the orientation of vocational education, this paper takes the actual workflow of mobile communication network planning and optimization posts as the main thread and adopts generative AI to restructure the curriculum systematically. The reform scheme is developed from six dimensions: curriculum objectives, teaching content, teaching modes, teaching resources, assessment and evaluation, as well as teachers’ professional competency. A curriculum framework is established following the principles of communication scenario-driven design, visualized business processes, run-through project tasks and in-depth empowerment by generative AI. A comprehensive project-based teaching system covering data collection, data governance, indicator analysis, visual presentation, fault troubleshooting and optimization scheme formulation is constructed. Furthermore, implementation approaches are proposed, including AI teaching assistants, AI-aided programming, intelligent generation of dynamic resources and whole-process diversified evaluation. Teaching practices verify that generative AI can significantly improve the efficiency of resource development, the accuracy of personalized tutoring and the effectiveness of teaching evaluation. To prevent degraded critical thinking and distorted evaluation outcomes caused by inappropriate overuse of technology, the application of AI in teaching must abide by four core principles: teacher-led instruction, desensitization of confidential data, manual verification of AI outputs, and full traceability of all teaching activities.