Scholarlink

With three campuses and an open distance learning unit, the North-West University draws students from all over South Africa. Across campuses, the module, Introduction to Computing and Programming, is offered in the first semester of the first year of several Bachelor of Science and Commerce courses. Many of these students are new to programming. In addition, the mother tongue of a large portion of students is not English. In this paper an intervention where generative artificial intelligence, specifically the ChatGPT API calls combined with its multilingual abilities, is investigated in terms of its capabilities to guide novice programmers to code solutions to problems. In this case to the benefit of Sesotho speaking students—to act as a programming tutor in the Python programming language. This bot was designed to emulate a human Sesotho-speaking tutor, and care has been taken to guide a student to find a coding solution rather than supplying scripts-as-answers. This is crucial in aiding the learning to code journey of students. An added benefit is that this bot will supply answers around the clock, to allow students to work on coding problems when it suits their schedule. They can also tailor the intervals and number of queries according to their needs without being concerned about troubling a flesh-and-blood person. The study involves students from two campuses in using the bot. With the focus of this research being qualitative, the students using Sesotho were treated as a case to enable an analysis of the use of the bot. An open-ended questionnaire was utilised to allow the students to relate their experience and the value of the intervention at the end of the learning period. Enlightenment on questions such as—does the Sesotho translation add value in this scenario, how deep could student questions delve, and did the bot guide students to learn, were the focus. Although refinement, in part by refining prompts, as well as based on the development of this technology is necessary, the potential of the use of generative artificial intelligence in this context is vast.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The Value of a Sesotho Bot as Tutor for an Introductory Python Programming Class

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Imelda Smit,
  • Laurie Butgereit

摘要

With three campuses and an open distance learning unit, the North-West University draws students from all over South Africa. Across campuses, the module, Introduction to Computing and Programming, is offered in the first semester of the first year of several Bachelor of Science and Commerce courses. Many of these students are new to programming. In addition, the mother tongue of a large portion of students is not English. In this paper an intervention where generative artificial intelligence, specifically the ChatGPT API calls combined with its multilingual abilities, is investigated in terms of its capabilities to guide novice programmers to code solutions to problems. In this case to the benefit of Sesotho speaking students—to act as a programming tutor in the Python programming language. This bot was designed to emulate a human Sesotho-speaking tutor, and care has been taken to guide a student to find a coding solution rather than supplying scripts-as-answers. This is crucial in aiding the learning to code journey of students. An added benefit is that this bot will supply answers around the clock, to allow students to work on coding problems when it suits their schedule. They can also tailor the intervals and number of queries according to their needs without being concerned about troubling a flesh-and-blood person. The study involves students from two campuses in using the bot. With the focus of this research being qualitative, the students using Sesotho were treated as a case to enable an analysis of the use of the bot. An open-ended questionnaire was utilised to allow the students to relate their experience and the value of the intervention at the end of the learning period. Enlightenment on questions such as—does the Sesotho translation add value in this scenario, how deep could student questions delve, and did the bot guide students to learn, were the focus. Although refinement, in part by refining prompts, as well as based on the development of this technology is necessary, the potential of the use of generative artificial intelligence in this context is vast.