Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses

Zhang Hai; Chen Xu; Xie Jiaojiao; Tang Wei

doi:doi:10.11648/j.si.20261406.17

Research Article |

| Peer-Reviewed

Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses

Zhang Hai^*, Chen Xu, Xie Jiaojiao, Tang Wei

Published in Science Innovation (Volume 14, Issue 6)

Received: 7 October 2026 Accepted: 26 October 2026 Published: 10 November 2026

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

Electromagnetic information-related courses serve as a crucial foundational component for majors in electronic information. These courses are characterized by abstract theories, complex formulas, and significant practical relevance. However, traditional teaching methods exhibit notable shortcomings in terms of intelligence, personalization, and interactivity. To address these challenges, this paper presents an artificial intelligence-assisted teaching platform developed on the "Coze" platform, integrating technologies such as large language models, enhanced knowledge base retrieval, database storage, and visual drag-and-drop functionalities. The platform is designed with a three-layer architecture comprising a presentation layer, a business logic layer, and a data layer. It encompasses nine distinct functions, including AI companions, practice-assistant tools, question bank collection, one-click test generation, one-click grading, one-click PPT generation, knowledge graph creation, online classrooms, and translation conversion. Furthermore, it mitigates the issue of AI hallucination through a dual mechanism that combines enhanced knowledge base retrieval with prompt word constraints. Deployment and testing results demonstrate that the platform is functionally complete, operates stably, and is user-friendly. It effectively addresses challenges such as the abstract nature of electromagnetic information knowledge, inadequate support for personalized learning, and the excessive repetitive workload faced by educators. This system provides a low-cost and high-efficiency development paradigm that serves as a reference for teachers lacking a computer background to leverage AI tools in constructing an auxiliary teaching system.

Published in	Science Innovation (Volume 14, Issue 6)
DOI	10.11648/j.si.20261406.17
Page(s)	269-277
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2026. Published by Science Publishing Group

Keywords

Electromagnetic Information-related Courses, Coze Programming, AI-assisted Teaching, Natural Language Processing, Multi-user Terminal Design

1．引言

在高等教育数字化转型的背景下，人工智能技术与教育教学的深度融合已成为推动教育改革、提升教学质量的重要引擎。电磁信息类课程作为电子信息类专业的重要基础课程，涵盖电磁场与电磁波、天线原理、微波技术、电磁兼容技术等核心课程，具有知识点抽象难懂、实践关联性强、教学难度大等特点。课程涉及大量复杂公式符号，学生难以直观理解；课程内容与工程实践结合紧密，传统课堂教学难以实现理论与实践的有效衔接；教师在备课、出卷、批改等环节工作量与重复性大，而学生课后自主学习又缺乏针对性指导。Long Y等将人工智能赋能学习归纳为内容驱动创作、协作学习与人工智能辅助、个性化学习系统、智能辅助系统四个方面

[1]

。扣子编程平台作为一款基于自然语言的低代码开发平台，具有操作便捷、无需复杂编程基础的优势，能够快速实现功能模块的开发与部署，为人工智能辅助教学平台的开发提供了便捷的技术支撑

[2]

。本文结合电磁信息类课程的教学痛点与扣子编程的技术优势，设计并实现了一款电磁信息类课程专用的人工智能辅助教学平台。主要聚焦电磁信息类这一特定类型课程，探索人工智能技术与专业课程深度融合的路径与方法，弥补了现有研究中针对该类课程AI教学平台相对不足的空白。同时基于“扣子编程”低代码平台的开发实践，验证了非计算机专业背景的教育工作者亦可利用新一代AI开发工具构建实用型教学系统，为教育技术领域提供了低成本、高效率的开发范式参考

[3]

。

2．技术概述与课程分析

2.1．技术概述

2.1.1．人工智能辅助教学

人工智能辅助教学是指将人工智能技术应用于教育教学过程，通过智能交互、个性化推荐、智能答疑等功能，辅助教师开展教学活动、辅助学生进行自主学习，实现教学过程智能化、个性化与高效化的一种教学模式，具有智能化、个性化、高效化三个核心特征

[4]

。梁迎丽和刘陈将人工智能教育应用的五大典型特征和四大具体应用形态总结为“人工智能与教育的融合创新发展体系”

[5]

。张治指出生成式人工智能正从底层逻辑上重塑教育的目标、内容与方式

[6]

。同时，在专业课程教学中，AI辅助教学能通过知识点可视化、智能答疑、个性化练习等有效改善专业课程抽象、复杂、教学难度大的问题，帮助学生理解复杂知识点、巩固学习成果，同时减轻教师教学负担

[7]

。

2.1.2．扣子编程平台

扣子编程（Coze）是字节跳动推出的一站式智能体开发平台，支持用户通过自然语言描述需求，快速搭建AI应用，无需掌握复杂的编程语言与底层技术。平台提供丰富的插件与模板，可无缝对接多种编程语言并搭建数据库，高效实现自然语言交互、数据处理、第三方工具集成等功能，具备开发高效、交互便捷、可扩展性强等特点，为本文平台的开发提供了技术支持

[8]

。

2.1.3．其他支撑技术

前端方面，平台基于Next.js 16的App Router开发，使用React 19进行UI渲染，全栈使用TypeScript 5保证类型安全性，采用Tailwind CSS 4进行响应式设计，利用shadcn/ui构建标准交互组件，引入KaTeX和react-markdown渲染数学公式，使用ReactFlow实现知识图谱的可视化拖拽展示，使用pptxgenjs在浏览器端生成并下载PPT。后端方面，基于Node.js 24与Next.js API Routes RESTful接口，数据存储采用PostgreSQL关系型数据库并配合Drizzle ORM进行类型安全操作，用户鉴权使用JWT＋bcryptjs，AI能力由coze-coding-dev-sdk提供（含大语言模型对话、图片生成、嵌入向量等），大文件上传使用busboy流式解析以防内存溢出。数据库遵循关系型范式，以references设置外键约束并配置onDelete级联，在热点查询列建立B-Tree索引，并通过drizzle-zod完成运行时数据校验与TypeScript类型推断，保证前后端数据交互时类型一致。

另外，自然语言处理（NLP）技术是实现人机语言交互的核心支撑，赵铁军等对自然语言处理技术进行了系统综述

[9]

；刘小冬从计算机处理角度归纳了自然语言理解“形式化—算法化—程序化”三个过程

[10]

。将生成式AI赋能的混合教学辅助平台总体架构分为用户层、应用层和数据层三层，这种分层化设计分工明确、便于维护，为本文平台搭建提供了参考思路

[11]

。

2.2．电磁信息类课程特点

本文聚焦的电磁信息类课程包括电磁场与电磁波、天线原理、微波技术、电磁兼容技术四门课程，其共同特点如下：

（1）理论性强：课程涉及大量公式推导与物理概念，如麦克斯韦方程组、电磁波传播规律、天线辐射原理等，知识点抽象，学生理解与记忆难度大，需依托知识图谱、AI答疑等技术辅助理解。麦克斯韦方程组的两种形式集中体现了课程的抽象性。

（2）关联性强：四门课程知识点相互关联，电磁场与电磁波是天线原理、微波技术课程的基础，电磁兼容技术则融合了前三门课程的相关内容。平台需通过数据库整合知识点数据，借助知识图谱技术构建关联网络，帮助学生建立完整的知识体系。

（3）教学需求多样：教师需高效完成出卷、批改、教学资源推送等工作，学生需要个性化学习指导、答疑解惑、练习题训练等，管理员需要对平台用户、资源、权限进行统一管理并维护平台稳定。

3．平台需求分析与总体设计

3.1．需求分析

3.1.1．用户需求分析

平台目标用户为教师、学生和管理员，其功能需求围绕九大核心功能模块展开

[12]

。教师核心需求包括：一键出卷（解决出卷耗时长、知识点不全面、题型不合理问题，无需人工挑选排列题目即可快速生成覆盖重难点的试卷）；一键批阅（解决批卷工作量大、效率低及难以及时了解学生不足的问题，实现客观题自动批阅、主观题提供参考建议）；一键生成PPT（解决制作课件耗时长、公式排版复杂、知识点罗列混乱问题，减少备课工作量）；线上课堂（解决线下课堂时空局限大、课后复习困难、师生交流较少问题，实现线上线下结合教学）。学生核心需求包括：AI学伴（全天候随时答疑，针对不同学生提出个性化学习建议，实现因材施教）；练题助手（按课程或知识点查找题目，支持错题归纳与专项练习并提供详细解析）；知识图谱（清晰表示知识点间关系，帮助学生建立完整知识框架）；翻译转换（解决阅读英文文献与专业术语的困扰，提供便捷翻译与内容存储）；收藏题库（整理收藏学习资料，便于温习与考前突击）。管理员需求包括用户管理（账号增删改、权限分配、批量导入导出、异常账号处理）、问题反馈处理、平台公告发布等。

3.1.2．非功能需求分析

易用性方面，要求页面简洁明了、流程简单、无需过多学习即可上手；支持自然语言对话方式下达任务；页面美观大方、功能按键明确可见。安全性方面，要求保障用户账号安全（采用加密技术存储密码，防止账号被盗）与数据安全（对平台各类数据加密存储，防止泄露、篡改、丢失）等

[13]

。

3.2．平台总体架构设计

结合需求分析结果与设计原则（实用性、易用性、准确性、可扩展性、安全性），平台采用分层化架构设计思想，总体分为表现层、业务逻辑层、数据层三层。三层之间遵循自左向右的单向依赖关系：表现层依赖业务逻辑层，业务逻辑层依赖数据层，数据层不依赖任何上层模块。这种单向依赖避免了层间循环调用，降低了系统复杂度与模块间耦合度，有利于梳理设计思路、提高模块运作稳定性，便于后续功能扩展与维护。平台三层架构设计如图1所示。

表现层由用户界面和自然语言交互接口构成：用户界面采用响应式布局、简洁大方、按钮位置合理；自然语言交互接口负责接收用户自然语言命令，并将其转化为业务逻辑层可理解的请求，同时将结果以自然语言形式呈现。业务逻辑层是平台核心处理层，实现用户权限检查、自然语言理解、功能模块逻辑操作、数据统计与分析等，并将九大核心功能模块封装为独立的业务模块，各模块间通过接口传递配合，方便后期修改、更新与扩展。数据层是平台数据支撑层，使用PostgreSQL数据库保存并管理用户数据、课程数据、题库数据三类核心数据，为业务逻辑层提供数据访问能力。

Download: Download full-size image

Figure 1. 图1 平台三层架构设计图。

3.3．数据库设计

结合平台的功能需求与数据类型，本文设计了用户数据表、收藏题库数据表、试卷数据表、练习记录数据表等多个数据表，确保数据规范性与一致性，核心数据表设计内容如表1所示。

表1 平台核心数据表设计。

数据表	主要字段	用途
用户数据表	id、账号、密码、角色、权限	存储教师、学生、管理员的账户数据，实现账号管理与权限控制
收藏题库数据表	id、userId、question、knowledgePoint、difficulty、courseType、answer	存储练题助手生成的题目数据，为后续功能提供题目支撑
试卷数据表	id、userId、title、course、knowledgePoints、difficulty、questionCount、questions	存储出卷配置参数与生成试卷题目，为一键出卷提供数据支撑

4．平台核心功能的设计与实现

4.1．开发环境与技术选型

平台开发运行于云端沙箱，该环境预配置Node.js 24运行时、pnpm包管理器、PostgreSQL数据库及各类系统编译工具，开发者通过浏览器即可编写、调试和预览。平台使用Coze CLI作为开发脚手架，通过coze init创建项目、coze dev启动开发服务、coze build构建生产版本，代码修改后自动触发热更新。核心技术选型如表2所示。

表2 平台前后端技术选型表。

类别	技术	用途
前端	Next.js 16 + React 19	全栈框架与UI渲染核心，采用App Router模式
前端	TypeScript 5	全栈类型安全，减少运行时错误
前端	Tailwind CSS 4 + shadcn/ui	响应式界面与标准化交互组件
前端	KaTeX + react-markdown	渲染数学公式与普通文字内容
前端	ReactFlow	知识图谱的可视化拖拽与关联展示
前端	pptxgenjs	浏览器端生成并下载PPT文件
后端	Node.js 24 + Next.js API Routes	服务端运行时与RESTful接口层
后端	PostgreSQL + Drizzle ORM	关系型数据库与类型安全操作
后端	JWT + bcryptjs	用户认证与密码加密
后端	coze-coding-dev-sdk	AI能力集成（对话、图片生成、对象存储等）
后端	busboy	流式处理大文件上传，避免内存溢出

4.2．九大核心功能模块

平台涵盖九大功能于一体，用户登录后可在首页按需选择功能模块使用，平台首页功能导航如图2所示。

4.2.1．学习辅导类模块

（1）AI学伴模块：基于大语言模型的对话系统，通过设定“电磁信息类课程专家”身份提示词提升回答的专业性与准确性；设计知识库，用户可上传教学资料，AI在回答前优先检索知识库内容再结合大模型作答，有效缓解AI幻觉；每次提问结合上下文对话历史，采用打字机式流式输出，回答后自动追问是否需深入、是否需例题等，通过引导式思考提升学生自主构建知识的能力。AI学伴功能模块界面如图3所示。

Download: Download full-size image

Figure 2. 图2 平台首页九大功能模块界面。

（2）练题助手模块：用户选择知识点、题型、难度与出题数量后，系统将条件写成提示词发给AI生成题目，题目可筛选收藏至题库；设计作答区域，支持文字或图片作答，AI先生成参考答案及解析，再识别用户作答并对比批改，给出批改结果与学习建议。为保障题目与解析质量，向AI提供电磁信息类课程公式符号书写规范并内置输出逻辑，确保题目正确渲染。练题助手功能模块界面如图4所示。

（3）知识图谱模块：使用ReactFlow可视化库将知识点渲染为关系图，节点表示知识点、连线表示关联，支持拖拽节点、缩放画布、增删编辑节点与连线，点击节点可查看基础概念与重要公式，视图状态可保存，为课程知识体系的构建提供了可视化支撑。知识图谱模块界面如图5所示。

（4）翻译转换模块：用户输入文字或上传图片后，系统将内容发给AI并指定目标语言，图片会先经AI识别文字再翻译，满足用户阅读英文文献的需求。

Download: Download full-size image

Figure 3. 图3 AI学伴功能模块界面。

4.2.2．教学考核类模块

（1）一键出卷模块：用户设置试卷标题、考察知识点范围、各题型数量后生成试卷，题目可由收藏题库自选或随机抽取，生成试卷的同时生成参考答案及解析，并支持连接打印机打印，实现了从需求到纸质试卷的完整闭环。

Download: Download full-size image

Figure 4. 图4 练题助手功能模块界面。

（2）一键批改模块：用户上传试卷图片或文件，AI分析题干内容并生成参考答案与解析，随后用户上传作答图片或文件，AI识别作答结果后与参考答案比较分析，得出批改结果及学习建议。

（3）一键生成PPT模块：严格按“封面→概念→物理图像→公式→例题→解析→知识点总结”的7页结构生成PPT，例题从收藏题库对应知识点中选取；为弥补AI对背景缺乏审美判断，模块内置PPT背景图片资源库供用户选择，系统将文字与图片组装成.pptx文件，可直接下载并在PowerPoint中打开。

Download: Download full-size image

Figure 5. 图5 知识图谱模块界面。

4.2.3．资源管理类模块

（1）收藏题库模块：用户收藏的题目作为数据存入PostgreSQL数据库，支持按知识点、难度、题型、课程分类筛选、搜索与删除，本质为增删改查数据库操作。

（2）线上课堂模块：教师端可上传教学视频及学习网站链接，视频统一保存至服务器指定目录并录入数据库归档；学生端可在线播放或下载视频、跳转第三方学习平台，满足多元化的线上自主学习需求。

此外，平台还实现用户权限管理、系统公告发布、用户反馈收集等辅助功能，管理员端可进行用户管理、发布公告、处理反馈。用户的登录账号密码统一由管理员授予或更改，管理员账号在后台设置不可被随意注册，确保平台安全稳定运行。

4.3．关键实现技术与界面优化

4.3.1．AI幻觉缓解机制

针对AI幻觉这一内生性缺陷，平台采用“知识库检索增强＋提示词约束”双重机制：一是知识库检索增强，用户上传教学资料后，AI回答前优先检索知识库内容再结合大模型作答，降低因数据知识边界与质量参差导致的幻觉；二是提示词约束，为AI设定身份提示词、公式符号书写规范、输出格式要求等，约束生成内容的专业性与准确性，缓解复杂解析中公式乱码等问题

[14]

。

4.3.2．界面设计与优化

界面设计遵循简洁直观、专业风格、一致性与规范性原则，以黑色为主色、低饱和度渐变背景，辅以线条图标与飘散颗粒营造稳重高端的视觉效果

[15]

。优化细节包括：通过不断增加提示词优化KaTeX渲染引擎，降低公式符号渲染错误；对图片、视频进行压缩并优化数据加载方式，加快页面加载速度；增加按钮点击反馈、加载动画及操作成功/失败提示，防止误操作。

5．平台测试与优化

5.1．测试方案

测试目的在于检验平台功能有效性、运行稳定性与用户体验友好性，测试类型覆盖功能测试、兼容性测试、性能测试、易用性测试、安全性测试五类。结合平台功能需求设计的核心模块典型测试用例如表3所示。

表3 核心功能模块测试用例。

模块	测试场景与预期结果
一键出卷	按要求生成试卷及参考答案与解析，题目数量内容准确，公式无乱码，可正确连接打印机
一键批改	正确识别题目并生成参考答案与解析，准确识别作答，给出合理批改结果与学习建议
AI学伴	专业易懂地解答电磁类问题，公式渲染正常，能够引导用户思考
线上课堂	视频可成功下载播放无卡顿，网站链接跳转顺畅
练题助手	题目正常生成无公式错误，批改结果正确合理，题目可成功收藏至题库
翻译转换	多语言翻译准确无误，无乱码
一键生成PPT	下载导出的PPT美观准确，公式符号格式正确，无渲染错误

5.2．测试结果与分析

测试实施采用单元测试、集成测试、系统测试与用户使用测试相结合的方式。结果显示：功能测试中，一键批改功能在生成较复杂解析时偶现公式符号乱码，其余功能均正常达到预期效果；兼容性测试中各功能在不同主流浏览器上运行良好，无页面错位或功能错误；性能测试中多名用户同时使用时页面与数据加载正常，无服务器崩溃或响应超时；安全性测试中无越权访问或数据泄漏，权限管理与数据加密措施有效；易用性测试中用户认为操作简便流畅、界面简洁美观。

5.3．测试不足与优化

针对测试发现的两处不足实施优化：一是针对一键批改复杂解析的公式乱码问题，向AI提出更规范的公式符号显示要求，提供提示词严格规定输出格式，并采用更优秀的AI大模型，基本解决了乱码问题；二是针对知识图谱页面色调较深影响视觉体验的问题，将背景色由黑色改为白色，增强图形与线条反差，改善用户浏览体验。

6．结论

本文针对电磁信息类课程教学中教师备课与批改工作量大、学生课后学习缺乏针对性辅导、教与学互动不足等痛点，基于扣子平台设计并实现了一套多功能人工智能辅助教学平台，完成了需求分析、总体设计、详细设计与实现、测试与优化全过程，实现了预期研究目标。主要成果如下：

（1）明确了平台需求与设计方案。通过对电磁信息类课程师生的调研，明确了功能需求与非功能需求，基于“实用性、易用性、准确性、可扩展性、安全性”五大设计原则，设计了“表现层—业务逻辑层—数据层”三层架构，完成了数据库设计与功能模块总体逻辑设计。

（2）完成了平台核心功能的开发与实现。基于扣子编程平台实现了一键出卷、一键批改、AI学伴、线上课堂、知识图谱、翻译转换、一键生成PPT、练题助手、收藏题库九大核心功能模块，同时实现用户权限管理、系统公告发布、用户反馈收集等辅助功能。

（3）通过多维度测试验证了平台的实用性与稳定性。测试覆盖功能完备性、兼容性、性能、易用性与安全性，结果显示平台可正常运行，在一定程度上解决了电磁信息类课程教学中的实际问题，提高了教与学的实际效果。

（4）验证了低代码技术应用于教学平台开发的可行性与推广价值。利用扣子编程低代码平台，无需编写大量代码即可完成教学平台开发及上线，降低了教学平台开发难度，对非计算机专业教师开发教学辅助系统具有重要参考价值。

基金项目

本文是“华侨大学2025年本科教育教学改革研究项目（HQJGYB2523）：人工智能背景下电磁信息类课程群综合改革与实践”的研究成果之一。

References

[1]	Long Y, Zhu Y, Zhang Y. Learning and teaching in AI-empowered higher education [M]//Wen W, Zhang Y. The global development of AI-empowered higher education: Beyond the horizon. Singapore: Springer, 2025. https://doi.org/10.1007/978-981-96-9005-3
[2]	胡佳健, 汪维富, 钟志贤. 国内外生成式人工智能教育现状分析与热点探究 [J]. 数字教育, 2024, 10(06): 84-92.
[3]	叶维裕, 陈景. AI时代教育人工智能辅助教学现状及研究 [J]. 科技风, 2025, (04): 68-70. https://doi.org/10.19392/j.cnki.1671-7341.202504023
[4]	罗玮琪. 人工智能在教育领域中的应用 [J]. 汉字文化, 2025, (08): 184-186. https://doi.org/10.14014/j.cnki.cn11-2597/g2.2025.08.024
[5]	梁迎丽, 刘陈. 人工智能教育应用的现状分析、典型特征与发展趋势 [J]. 中国电化教育, 2018, (03): 24-30. https://doi.org/10.3969/j.issn.1006-9860.2018.03.004
[6]	张治. ChatGPT/生成式人工智能重塑教育的底层逻辑和可能路径 [J]. 华东师范大学学报(教育科学版), 2023, 41(07): 131-142. https://doi.org/10.16382/j.cnki.1000-5560.2023.07.012
[7]	朱永新, 杨帆. ChatGPT/生成式人工智能与教育创新：机遇、挑战以及未来 [J]. 华东师范大学学报(教育科学版), 2023, 41(07): 1-14. https://doi.org/10.16382/j.cnki.1000-5560.2023.07.001
[8]	韦玉荣, 熊圣昊, 黄思颖. 基于 Coze 平台与 DeepSeek 大模型的智能辅助教学智能体 [J]. 软件工程与应用, 2025, 14 (06): 1246-1257. https://doi.org/10.12677/sea.2025.146110
[9]	赵铁军, 许木璠, 陈安东. 自然语言处理研究综述 [J]. 新疆师范大学学报(哲学社会科学版), 2025, 46(02): 89-111. https://doi.org/10.14100/j.cnki.65-1039/g4.20230804.001
[10]	刘小冬. 自然语言理解综述 [J]. 统计与信息论坛, 2007, 22(02): 5-12.
[11]	Malik A, Khan L M, Hussain K, et al. AI in higher education: unveiling academicians’ perspectives on teaching, research, and ethics in the age of ChatGPT [J]. Interactive Learning Environments, 2025, 33(3): 2390-2406. https://doi.org/10.1080/10494820.2024.2409407
[12]	裴壮, 田秀霞, 李冰雪. 知识图谱赋能的面向对象程序设计C++教学改革与实践 [J]. 华东师范大学学报(自然科学版), 2024, (05): 104-113. https://doi.org/10.3969/j.issn.1000-5641.2024.05.010
[13]	张缨斌, 吴若乔, 何雨轩, 等. 感知情境与人在回路的智能教育——《人工智能与教学的未来：见解与提议》要点与反思 [J]. 开放教育研究, 2023, 29(04): 11-20. https://doi.org/10.13966/j.cnki.kfjyyj.2023.04.002
[14]	郭全中, 张磊, 韦薇. AI幻觉的生成机理与敏捷治理 [J]. 新闻爱好者, 2026, (02): 36-41. https://doi.org/10.16017/j.cnki.xwahz.20251205.001
[15]	管幸生, 邹明杰, 冯涛. 基于 QFD 的线上教学软件界面设计研究 [J]. 福建工程学院学报, 2021, 19(04): 385-390. https://doi.org/10.3969/j.issn.1672-4348.2021.04.013

Cite This Article

Plain Text BibTeX RIS

APA Style

Hai, Z., Xu, C., Jiaojiao, X., Wei, T. (2026). Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses. Science Innovation, 14(6), 269-277. https://doi.org/10.11648/j.si.20261406.17

Copy | Download

ACS Style

Hai, Z.; Xu, C.; Jiaojiao, X.; Wei, T. Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses. Sci. Innov. 2026, 14(6), 269-277. doi: 10.11648/j.si.20261406.17

Copy | Download

AMA Style

Hai Z, Xu C, Jiaojiao X, Wei T. Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses. Sci Innov. 2026;14(6):269-277. doi: 10.11648/j.si.20261406.17

Copy | Download

@article{10.11648/j.si.20261406.17,
  author = {Zhang Hai and Chen Xu and Xie Jiaojiao and Tang Wei},
  title = {Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses},
  journal = {Science Innovation},
  volume = {14},
  number = {6},
  pages = {269-277},
  doi = {10.11648/j.si.20261406.17},
  url = {https://doi.org/10.11648/j.si.20261406.17},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.si.20261406.17},
  abstract = {Electromagnetic information-related courses serve as a crucial foundational component for majors in electronic information. These courses are characterized by abstract theories, complex formulas, and significant practical relevance. However, traditional teaching methods exhibit notable shortcomings in terms of intelligence, personalization, and interactivity. To address these challenges, this paper presents an artificial intelligence-assisted teaching platform developed on the "Coze" platform, integrating technologies such as large language models, enhanced knowledge base retrieval, database storage, and visual drag-and-drop functionalities. The platform is designed with a three-layer architecture comprising a presentation layer, a business logic layer, and a data layer. It encompasses nine distinct functions, including AI companions, practice-assistant tools, question bank collection, one-click test generation, one-click grading, one-click PPT generation, knowledge graph creation, online classrooms, and translation conversion. Furthermore, it mitigates the issue of AI hallucination through a dual mechanism that combines enhanced knowledge base retrieval with prompt word constraints. Deployment and testing results demonstrate that the platform is functionally complete, operates stably, and is user-friendly. It effectively addresses challenges such as the abstract nature of electromagnetic information knowledge, inadequate support for personalized learning, and the excessive repetitive workload faced by educators. This system provides a low-cost and high-efficiency development paradigm that serves as a reference for teachers lacking a computer background to leverage AI tools in constructing an auxiliary teaching system.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses
AU  - Zhang Hai
AU  - Chen Xu
AU  - Xie Jiaojiao
AU  - Tang Wei
Y1  - 2026/11/10
PY  - 2026
N1  - https://doi.org/10.11648/j.si.20261406.17
DO  - 10.11648/j.si.20261406.17
T2  - Science Innovation
JF  - Science Innovation
JO  - Science Innovation
SP  - 269
EP  - 277
PB  - Science Publishing Group
SN  - 2328-787X
UR  - https://doi.org/10.11648/j.si.20261406.17
AB  - Electromagnetic information-related courses serve as a crucial foundational component for majors in electronic information. These courses are characterized by abstract theories, complex formulas, and significant practical relevance. However, traditional teaching methods exhibit notable shortcomings in terms of intelligence, personalization, and interactivity. To address these challenges, this paper presents an artificial intelligence-assisted teaching platform developed on the "Coze" platform, integrating technologies such as large language models, enhanced knowledge base retrieval, database storage, and visual drag-and-drop functionalities. The platform is designed with a three-layer architecture comprising a presentation layer, a business logic layer, and a data layer. It encompasses nine distinct functions, including AI companions, practice-assistant tools, question bank collection, one-click test generation, one-click grading, one-click PPT generation, knowledge graph creation, online classrooms, and translation conversion. Furthermore, it mitigates the issue of AI hallucination through a dual mechanism that combines enhanced knowledge base retrieval with prompt word constraints. Deployment and testing results demonstrate that the platform is functionally complete, operates stably, and is user-friendly. It effectively addresses challenges such as the abstract nature of electromagnetic information knowledge, inadequate support for personalized learning, and the excessive repetitive workload faced by educators. This system provides a low-cost and high-efficiency development paradigm that serves as a reference for teachers lacking a computer background to leverage AI tools in constructing an auxiliary teaching system.
VL  - 14
IS  - 6
ER  -

Copy | Download

Author Information

Zhang Hai

College of Information Science and Engineering, Huaqiao University, Xiamen, China

Contact Email
Chen Xu

College of Information Science and Engineering, Huaqiao University, Xiamen, China
Xie Jiaojiao

College of Information Science and Engineering, Huaqiao University, Xiamen, China
Tang Wei

College of Information Science and Engineering, Huaqiao University, Xiamen, China

Download PDF

Submit an Article

Figure 1

Figure 1. 图 1 平台三层架构设计图。
Figure 2

Figure 2. 图 2 平台首页九大功能模块界面。
Figure 3

Figure 3. 图 3 AI学伴功能模块界面。
Figure 4

Figure 4. 图 4 练题助手功能模块界面。
Figure 5

Figure 5. 图 5 知识图谱模块界面。

Table 1

表1 平台核心数据表设计。
Table 2

表2 平台前后端技术选型表。
Table 3

表3 核心功能模块测试用例。

Plain Text BibTeX RIS

APA Style

Hai, Z., Xu, C., Jiaojiao, X., Wei, T. (2026). Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses. Science Innovation, 14(6), 269-277. https://doi.org/10.11648/j.si.20261406.17

Copy | Download

ACS Style

Hai, Z.; Xu, C.; Jiaojiao, X.; Wei, T. Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses. Sci. Innov. 2026, 14(6), 269-277. doi: 10.11648/j.si.20261406.17

Copy | Download

AMA Style

Hai Z, Xu C, Jiaojiao X, Wei T. Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses. Sci Innov. 2026;14(6):269-277. doi: 10.11648/j.si.20261406.17

Copy | Download

@article{10.11648/j.si.20261406.17,
  author = {Zhang Hai and Chen Xu and Xie Jiaojiao and Tang Wei},
  title = {Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses},
  journal = {Science Innovation},
  volume = {14},
  number = {6},
  pages = {269-277},
  doi = {10.11648/j.si.20261406.17},
  url = {https://doi.org/10.11648/j.si.20261406.17},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.si.20261406.17},
  abstract = {Electromagnetic information-related courses serve as a crucial foundational component for majors in electronic information. These courses are characterized by abstract theories, complex formulas, and significant practical relevance. However, traditional teaching methods exhibit notable shortcomings in terms of intelligence, personalization, and interactivity. To address these challenges, this paper presents an artificial intelligence-assisted teaching platform developed on the "Coze" platform, integrating technologies such as large language models, enhanced knowledge base retrieval, database storage, and visual drag-and-drop functionalities. The platform is designed with a three-layer architecture comprising a presentation layer, a business logic layer, and a data layer. It encompasses nine distinct functions, including AI companions, practice-assistant tools, question bank collection, one-click test generation, one-click grading, one-click PPT generation, knowledge graph creation, online classrooms, and translation conversion. Furthermore, it mitigates the issue of AI hallucination through a dual mechanism that combines enhanced knowledge base retrieval with prompt word constraints. Deployment and testing results demonstrate that the platform is functionally complete, operates stably, and is user-friendly. It effectively addresses challenges such as the abstract nature of electromagnetic information knowledge, inadequate support for personalized learning, and the excessive repetitive workload faced by educators. This system provides a low-cost and high-efficiency development paradigm that serves as a reference for teachers lacking a computer background to leverage AI tools in constructing an auxiliary teaching system.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Design and Implementation of an AI-Assisted Teaching Platform for Electromagnetic Information Courses
AU  - Zhang Hai
AU  - Chen Xu
AU  - Xie Jiaojiao
AU  - Tang Wei
Y1  - 2026/11/10
PY  - 2026
N1  - https://doi.org/10.11648/j.si.20261406.17
DO  - 10.11648/j.si.20261406.17
T2  - Science Innovation
JF  - Science Innovation
JO  - Science Innovation
SP  - 269
EP  - 277
PB  - Science Publishing Group
SN  - 2328-787X
UR  - https://doi.org/10.11648/j.si.20261406.17
AB  - Electromagnetic information-related courses serve as a crucial foundational component for majors in electronic information. These courses are characterized by abstract theories, complex formulas, and significant practical relevance. However, traditional teaching methods exhibit notable shortcomings in terms of intelligence, personalization, and interactivity. To address these challenges, this paper presents an artificial intelligence-assisted teaching platform developed on the "Coze" platform, integrating technologies such as large language models, enhanced knowledge base retrieval, database storage, and visual drag-and-drop functionalities. The platform is designed with a three-layer architecture comprising a presentation layer, a business logic layer, and a data layer. It encompasses nine distinct functions, including AI companions, practice-assistant tools, question bank collection, one-click test generation, one-click grading, one-click PPT generation, knowledge graph creation, online classrooms, and translation conversion. Furthermore, it mitigates the issue of AI hallucination through a dual mechanism that combines enhanced knowledge base retrieval with prompt word constraints. Deployment and testing results demonstrate that the platform is functionally complete, operates stably, and is user-friendly. It effectively addresses challenges such as the abstract nature of electromagnetic information knowledge, inadequate support for personalized learning, and the excessive repetitive workload faced by educators. This system provides a low-cost and high-efficiency development paradigm that serves as a reference for teachers lacking a computer background to leverage AI tools in constructing an auxiliary teaching system.
VL  - 14
IS  - 6
ER  -

Copy | Download