Engineered Prompts in ChatGPT for Educational Assessment in Software Engineering and Computer Science

被引:1
作者
Diyab, Ayman [1 ]
Frost, Russell Morris [1 ]
Fedoruk, Benjamin David [2 ]
Diyab, Ahmad [3 ]
机构
[1] Lakehead Univ, Fac Engn, Thunder Bay, ON P7B 5E1, Canada
[2] Ontario Tech Univ, Mitch & Leslie Frazer Fac Educ, Oshawa, ON L1G 0C5, Canada
[3] Western Univ, Shad Alumni, London, ON N6A 3K7, Canada
关键词
generative AI; ChatGPT; educational assessment; prompt engineering; assessment aids; computer science; software engineering;
D O I
10.3390/educsci15020156
中图分类号
G40 [教育学];
学科分类号
040101 ; 120403 ;
摘要
AI Assess, a ChatGPT-based assessment system utilizing the ChatGPT platform by OpenAI, composed of four components, is proposed herein. The components are tested on the GPT model to determine to what extent they can grade various exam questions based on learning outcomes, generate relevant practice problems to improve content retention, identify student knowledge gaps, and provide instantaneous feedback to students. The assessment system has been explored using software engineering and computer science courses and is successful through testing and evaluation. AI Assess has demonstrated the ability to generate practice problems based on syllabus information and learning outcomes. The components have been shown to identify weak areas for students. Finally, it has been shown to provide different levels of feedback. The combined set of components, if incorporated into a complete software system and implemented in classrooms with proposed transparency mechanisms, has vast potential to reduce instructor workload, improve student understanding, and enhance the learning experience. The potential for GPT-powered chatbots in educational assessments is vast and must be embraced by the education sector.
引用
收藏
页数:33
相关论文
共 50 条
[1]   Automatic grading for Arabic short answer questions using optimized deep learning model [J].
Abdul Salam, Mustafa ;
Abd El-Fatah, Mohamed ;
Hassan, Naglaa Fathy .
PLOS ONE, 2022, 17 (08)
[2]  
Abdullah M, 2022, Soc Net Anal Mgmt Se, P223, DOI 10.1109/SNAMS58071.2022.10062688
[3]   The achievement of higher flexibility in multiple-choice-based tests using image classification techniques [J].
Afifi, Mahmoud ;
Hussain, Khaled F. .
INTERNATIONAL JOURNAL ON DOCUMENT ANALYSIS AND RECOGNITION, 2019, 22 (02) :127-142
[4]  
Ahmadi D., 2018, International Journal of Research in English Education, V3, P115, DOI DOI 10.29252/IJREE.3.2.115
[5]   Automated clinical pathway standardization using SNOMED CT- based semantic relatedness [J].
Alahmar, Ayman ;
AlMousa, Mohannad ;
Benlamri, Rachid .
DIGITAL HEALTH, 2022, 8
[6]   Optimizing Hospital Resources using Big Data Analytics with Standardized e-Clinical Pathways [J].
Alahmar, Ayman ;
Benlamri, Rachid .
2020 IEEE INTL CONF ON DEPENDABLE, AUTONOMIC AND SECURE COMPUTING, INTL CONF ON PERVASIVE INTELLIGENCE AND COMPUTING, INTL CONF ON CLOUD AND BIG DATA COMPUTING, INTL CONF ON CYBER SCIENCE AND TECHNOLOGY CONGRESS (DASC/PICOM/CBDCOM/CYBERSCITECH), 2020, :650-657
[7]   Ontological framework for standardizing and digitizing clinical pathways in healthcare information systems [J].
Alahmar, Ayman ;
Crupi, Matteo Ermando ;
Benlamri, Rachid .
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2020, 196
[8]  
Anderson J., 2018, Pew Research Center, V10
[9]   Social media-based COVID-19 sentiment classification model using Bi-LSTM [J].
Arbane, Mohamed ;
Benlamri, Rachid ;
Brik, Youcef ;
Alahmar, Ayman Diyab .
EXPERT SYSTEMS WITH APPLICATIONS, 2023, 212
[10]  
Astin A. W., 1984, Journal of College Student Development, V40, P529