生成式人工智能技术在实验教学中的应用——以数据科学实验为例
Application of Generative Artificial Intelligence Technology in Experimental Teaching——A Case Study of Data Science Experiments
黄炜 1吴昀璟 2余辉 2张瑞2
作者信息
- 1. 湖北工业大学经济与管理学院,武汉 430068;武汉晴川学院商学院,武汉 430204
- 2. 湖北工业大学经济与管理学院,武汉 430068
- 折叠
摘要
开发了基于生成式人工智能(GAI)技术的数据科学实验框架.根据数据科学相关课程实验要求,依次在实验准备、探索性数据分析以及模型构建中融入GAI辅助实验.实验以UCI机器学习库公开数据集中白葡萄酒数据为例,利用GAI模型GPT-4 进行白葡萄酒质量的分类预测和模型优化,总体准确率达到 95.5%.实验表明,基于GAI技术的实验框架能够在保证实验准确率的前提下,降低实验者的学习门槛,提高学生运用大数据分析和解决现实问题的能力,提升数据科学实验教学的效果.
Abstract
This article designs and develops a data science experimental framework based on generative artificial intelligence(GAI)technology.According to the experimental steps of data science related courses,generative artificial intelligence is integrated into experimental preparation,exploratory data analysis and model construction.Taking the UCI machine learning library public dataset of white wine as an example,the experiment uses the generative artificial intelligence model GPT-4 to classify and optimize the quality of white wine.The overall accuracy of the experimental results reaches 95.5%.The experiment shows that the experimental framework based on generative artificial intelligence technology can reduce the learning threshold for experimenters,improve students'practical ability to use big data analysis and solve real-world problems,and enhance the effectiveness of data science experimental teaching while ensuring experimental accuracy.
关键词
数据科学/生成式人工智能/实验教学/人机共教Key words
data science/generative artificial intelligence/experimental teaching/human-machine co-teaching引用本文复制引用
基金项目
中国高校产学研创新基金项目(2022TX028)
出版年
2024
NETL
NSTL
万方数据