人工智能赋能高等农林教育实验教学体系重构*——面向知农爱农新型人才培养的教学云平台开发

Reconstructing the experimental teaching system in higher agricultural and forestry education through artificial intelligence：Development of a teaching cloud platform for cultivating agriculture-oriented talents

关键词:  人工智能  高等农林教育  知农爱农  教学云平台  实验教学

Key words:  artificial intelligence  higher agricultural and forestry education  understanding and loving agriculture  teaching cloud platform  experimental teaching

摘要:

【目的】 在新农科和乡村振兴背景下，传统实验教学存在“三农”元素不足、实践环节缺乏连贯性等突出问题。文章依托“AI农数导师”教学云平台，聚焦实验教学体系重构，构建并实现三农场景融合、实验教学一体化和人机共教治理的教学云平台与实施路径，服务“知农爱农”新型人才培养。【方法】 以暨南大学经济统计学专业《统计软件》课程为牵引，整合农业统计、生物统计、农村调查等教材与项目数据，构建覆盖统计方法、软件实现路径、“三农”应用场景与决策指标的教育语料库和知识图谱；开发集智能备课、个性化练习、代码生成与调试于一体的教育智能体，嵌入农业大数据分析云平台，形成覆盖课前诊断与导学、课中实验实施、课后拓展与反思的一体化数智教学流程。通过学习行为数据、问卷与作业质量对成效进行评价。【结果】 平台提高了课程中“三农”案例比重和数据密度，学生R语言编程通过率、综合实验完成质量和项目化成果转化水平得到提升，统计思维、编程能力、数据素养与跨学科应用能力进一步增强；同时，学生对农业绿色发展、数字乡村建设和乡村振兴现实需求的理解明显加深，知农爱农的价值认同得到强化。【结论】 基于教育语料库、知识图谱与教育智能体的教学云平台，能够重构“理论—软件—数据—‘三农’场景”一体化实验教学体系，提升学生懂农业、懂经济、懂数据的综合素养，为综合性院校开展“人工智能+新农科”交叉育人提供可推广的实践范式。

Abstract:

[Purpose] Against the backdrop of new agricultural science and rural revitalization，traditional experimental teaching faces prominent issues such as insufficient elements relating to agriculture，rural areas，and rural residents，as well as a lack of continuity in practical segments. Relying on the "AI Agricultural Data Tutor" teaching cloud platform，this paper focuses on the reconstruction of the experimental teaching system. It constructs and implements a teaching cloud platform and an implementation path that integrate agricultural scenarios，unified experimental teaching，and human-machine collaborative governance，serving the cultivation of new-type talents who understand and love agriculture.[Method] Using the "Statistical Software" course in the major of Economic Statistics at Jinan University as the anchor，this study integrated teaching materials and project-based data from agricultural statistics，biostatistics，and rural surveys to construct an educational corpus and knowledge graph encompassing statistical methods，software implementation pathways，application scenarios concerning issues relating to agriculture，rural areas，and rural residents，and decision-making indicators. It further developed an educational intelligent agent that integrates intelligent lesson preparation，personalized exercises，code generation，and debugging，and embeded this agent into an agricultural big data analytics cloud platform，thereby forming an integrated digital-intelligent teaching process that covered pre-class diagnosis and guided learning，in-class experimental implementation，and post-class extension and reflection. The effectiveness of this approach was evaluated through learning behavior data，questionnaire responses，and the quality of student assignments.[Result] The platform increased the proportion of agriculture-related cases and the density of real data used in the course. Students′ pass rate in R programming，the quality of comprehensive experimental tasks，and the level of project-based achievement transformation were improved. Their statistical thinking，programming ability，data literacy，and interdisciplinary application ability were further enhanced. Meanwhile，students gained a significantly deeper understanding of agricultural green development，digital rural development，and the practical needs of rural revitalization，and their value identification with knowing and loving agriculture was strengthened.[Conclusion] The teaching cloud platform based on an educational corpus，a knowledge graph，and educational intelligent agents is able to reconstruct an integrated experimental teaching system linking "theory，software，data，agriculture-related scenarios"，enhance students′ comprehensive literacy in understanding agriculture，economics，and data，and provide a replicable practical paradigm for interdisciplinary talent cultivation that integrates "artificial intelligence + new agricultural sciences" in comprehensive universities.