Generative AI Tools For Education – Gemini AI (2)

Co-designing a Prompting Framework with Gemini AI

Building on part one of this post series, consistent use of generative AI improves your prompting skills. By iteratively using and refining prompts based on results, you will develop personalised frameworks for specific needs. Alternatively, many generative AI tools provide built-in guidance or tutorials for prompting. You can use these to guide your prompting for optimal result. In this post, we can now explore co-designing a prompt framework with Gemini AI and then use the framework to create and generate week one contents of the course outline we generated in part one of this post. This demonstrates how educators can leverage Gemini AI’s capabilities to create an effective framework for generating educational content. We start by using a simple natural language prompt as already been suggested in part one of this post.

Prompt:

Can you give me a concise framework, acronyms with brief explanation for each on how to prompt you for the best educational output?

Output:

Absolutely! Here’s a concise framework using an acronym to remember how to prompt me for the best educational output: CRISP

1. C – Clarity:

• • State your educational goal clearly. What do you want your students to learn?
  • Avoid ambiguity and overly complex language.

2. R – Relevance:

• • Provide context about the subject matter and learning environment (e.g., course level, topic).
  • This helps me tailor the response to your specific needs.

3. I – Inquiry:

• • Use open-ended questions that encourage critical thinking and exploration.
  • Instead of “What is photosynthesis?” ask “How does light influence the process of photosynthesis?”

4. S – Specificity:

• • Be specific about the type of output you desire.
  • Do you want factual information, analysis of a concept, or creative prompts?

5. P – Polish:

• • Refine your prompt based on my initial response.
  • This iterative approach ensures you get the most out of our interaction.

By following the CRISP framework, you can effectively communicate your educational goals and get the best possible learning outcomes from me.

Gemini generated the above framework with brief explanations, you can always refine the prompt to give different output from this depending on what you want the framework to represent in terms of the acronyms. Besides, generative AI like Gemini will always produce different output from the previous one each time you prompt it to generate an output for the same prompt or a refined one. In this case, we leave  the first output without regenerating.

Sample prompt template based on the CRISP framework for Gemini AI

Once we have a framework (CRISP) as given above, we can go ahead and ask Gemini to create a template based on the Framework that we can edit each time we need to generate educational contents based on the framework. You can always create your own template based on the framework by following explanation for each acronym in the framework. But here we get Gemini AI to do the heavy lifting.

Earlier in the first part of the post, we have generated an outline for a “14 week Software Engineering with python” course based on the template we borrowed from Alchemy. Now, we are asking Gemini AI to create a prompt template based on the CRISP framework to generate contents for each module in the course outline. Note that we are asking for a template that we can repeatedly used to generate contents for the topics in the course outline. It is also important to note that in the prompt use, we have referred to the CRISP framework, without stating what it is and have also referred to the “Software Engineering Course with Python” without stating the outline. It is simply because we are still using the same session on Gemini, otherwise, we  have to state clearly what the framework entails as well as the outline for Gemini to understand.

Prompt:

Using the CRISP framework as you have suggested above, create a prompt template to generate content for each module in the software engineering course with python you outlined earlier?

Output :

CRISP Prompt Template for Software Engineering with Python Modules

Course Name: Software Engineering with Python (Replace with your specific course name)

Learning Management System: Online Modules (Replace if using a different format)

Module Topic: [Insert specific topic for the week (e.g., Functions & Code Reusability)]

CRISP Prompt:

C – Clarity:

The goal of this module is for students to understand the concept of functions in Python and their role in code reusability.

R – Relevance:

This module is part of a 14-week college-level Software Engineering with Python course. Students have already covered basic Python syntax, data types, and control flow statements.

I – Inquiry:

How can I design prompts and activities within this module that encourage students to explore the benefits of using functions in Python?

What kind of coding exercises or projects would effectively demonstrate the concept of code reusability through functions?

S – Specificity:

I would like the generated content for this module to include:

• A short introduction explaining functions and their advantages in Python.
• Interactive coding exercises that guide students through defining and calling functions with parameters and return values.
• A mini-project where students can apply their understanding of functions to create a program demonstrating code reusability.

P – Polish:

I can review the initial content generated and provide further feedback to refine the module content based on the specific learning objectives.

From the above output, Gemini AI created the prompt template with the CRISP framework using example module topic for clarity.  All you need to do is just replace the place-holders with the specific information of the course you need and follow the sample guide include the CRISP details.

Still engaging Gemini AI,  we requested Gemini AI to create a sample prompt with the template for week 1 module of the course in the outline “Welcome to software Engineering”. Note that we continue on the same session with previous prompts and so Gemini AI can still understand the context. Out of context, we will have to state clearly the framework to use.

Prompt:

Give me a sample prompt to create content for week 1 module “Welcome to Software Engineering with Python!”

Output:

CRISP Prompt: Week 1 – Welcome to Software Engineering with Python!

C – Clarity: The goal of this module is to introduce students to the course, establish a positive learning environment, and set expectations.

R – Relevance: This is the first module of a 14-week college-level Software Engineering with Python course. Students may have varying levels of programming experience.

I – Inquiry: How can I create an engaging introduction that sparks students’ interest in software engineering and learning Python? What activities or resources can I incorporate to help students get familiar with the learning management system and course structure?

S – Specificity: I would like the generated content for this module to include:

• A welcome message that highlights the importance and applications of software engineering in today’s world.
• A brief overview of the course syllabus, key topics to be covered, and learning objectives.
• An interactive icebreaker activity to introduce students to each other and gauge their prior programming experience (if any).
• A short quiz or assignment to help students set up their Python development environment and explore basic functionalities.

P – Polish:

I can review the initial content and ensure it strikes a balance between being informative and engaging for students at the beginning of the course.

Generating course content using the CRISP framework

With the sample prompt based on the template created above and the CRISP framework, we use the exact prompt generated to prompt Gemini AI for the course content.

Prompt:

CRISP Prompt: Week 1 – Welcome to Software Engineering with Python!

C – Clarity: The goal of this module is to introduce students to the course, establish a positive learning environment, and set expectations.

R – Relevance: This is the first module of a 14-week college-level Software Engineering with Python course. Students may have varying levels of programming experience.

I – Inquiry: How can I create an engaging introduction that sparks students’ interest in software engineering and learning Python? What activities or resources can I incorporate to help students get  familiar with the learning management system and course structure?

S – Specificity: I would like the generated content for this module to include:

• A welcome message that highlights the importance and applications of software engineering in today’s world.
• A brief overview of the course syllabus, key topics to be covered, and learning objectives.
• An interactive icebreaker activity to introduce students to each other and gauge their prior programming experience (if any).
• A short quiz or assignment to help students set up their Python development environment and explore basic functionalities.

Output:

Week 1: Welcome to Software Engineering with Python!

Following your CRISP prompt, here’s a sample content for your first-week module:

Module Title: Welcome to the Exciting World of Software Engineering with Python!

Introduction:

Welcome to this Software Engineering with Python course! Are you ready to unlock the power of coding and create amazing things? In this journey, you’ll not only learn Python, a versatile and in-demand programming language, but also delve into the core principles of software engineering. Software engineering is all about applying problem-solving skills and logical thinking to design, develop, and maintain software applications that we use every day. From the websites you browse to the apps on your phone, software engineering plays a crucial role in shaping our digital world!

Course Overview:

This 14-week course will equip you with the fundamental knowledge and skills to navigate the exciting world of software engineering. We’ll cover various topics like:

• Python Programming Fundamentals: Learn the building blocks of Python, including variables, data types, control flow, and functions.
• Object-Oriented Programming (OOP): Explore how to structure your code for better organization and reusability using OOP concepts like classes and objects.
• Data Structures and Algorithms: Understand different ways to organize data efficiently and explore problem-solving techniques using algorithms.
• Exception Handling and Testing: Learn how to make your code robust by handling errors gracefully and writing tests to ensure its functionality.
• Software Development Practices: Discover essential practices like version control and collaboration tools used by professional software engineers.

By the end of this course, you’ll be able to design, develop, and test your own Python applications, laying a strong foundation for your software engineering journey.

Icebreaker Activity:

Let’s get to know each other! In the discussion forum, introduce yourself and share:

• Your name and a fun fact about yourself.
• What motivates you to learn software engineering and Python?
• Do you have any prior programming experience? (Optional)

Getting Started:

To kick things off, let’s ensure everyone is on the same page.

• [Insert instructions and resources for setting up a Python development environment (e.g., downloading and installing Python, using online platforms).
• [Optional] Take a short quiz or complete a small coding exercise to familiarize yourself with basic Python functionalities (e.g., printing text, performing simple calculations).

We’ll be diving deeper into Python in the next module, but this initial setup will help you explore on your own and get comfortable with the environment.

Looking Ahead:

Throughout this course, we’ll combine lectures, hands-on coding exercises, projects, and discussions to create a dynamic and engaging learning experience. Feel free to ask questions, participate actively in discussions, and seek help whenever needed. I’m here to guide you on your path to becoming a skilled software engineer!

This module provides a welcoming introduction, establishes expectations, and gives students a taste of what’s to come. Remember to replace bracketed information with specific instructions and resources relevant to your course.

This content as generated by Gemini AI based on the CRISP framework is not a finished product, this is a starter content upon which you can build your course contents to the required standard in length and breadth, using the generative AI as an assistant in modifying and refining the prompts as needed. For instance where it says to “insert instruction and resources for setting up a python development environment”, if you already have this resource, you can add this to the content or provide a link to the resource. Alternatively, you may want to prompt Gemini AI to generate the instruction and link to the resources. Note that generative AI can hallucinate and also give you a dead link, this is where your verification and validation comes it. Ensure to verify all information given are correct and up-to-date and consult a subject matter expert for validation is you are not one.

The exercise above shows how we can easily use Gemini AI or any other generative AI without special knowledge on prompting. Simply by natural language and refining prompt based on the output, you can use generative AI to create Educational contents for your purpose. You can develop your own prompting framework with assistant from the generative AI as well as create a template for generating contents to suit your purpose.

Prompt to generate Course Contents for week 2 from the Course Outline

The following prompt is created from the template generated based on the CRISP Framework and used above to generate week 1 of the course content. Note that the P-Polish is missing, this is a directive for the user to modify the prompt based on the output if necessary till the required output is generated. You can try out this prompt as it is and see what output you get and take it from there. Alternatively, just replace with you specific course details and try out the template.

Course Name: Software Engineering with Python

Learning Management System: Online Modules Module

Topic: Python Fundamentals – Variables, Data Types, & Operators

CRISP Prompt:

C – Clarity: The goal of this Module is intended for students to be able to:

• Create and manipulate variables of different data types (in-class coding exercise).
• Apply various arithmetic, comparison, and logical operators to evaluate expressions (coding quiz).

R – Relevance: This module is part of a 14-week college-level Software Engineering with Python course. Students have done the introductory session where they:

• Define software engineering principles and their application in problem-solving (written assignment).
• Set up a Python development environment and run basic code examples (in-class activity).

I – Inquiry:

• How can I design prompts and activities within this module that encourage students to learn and understand the different data types?
• What kind exercise can I use to help them understand the different arithmetic and logical operations in python language and evaluating expressions?

S – Specificity: I would like the generated content for this module to include:

• A short introduction explaining variables, arithmetic and logical operations.
• Interactive coding exercises for students to evaluating expression with variable, arithmetic and logical operation.

In summary we:

• Prompted Gemini AI for a guide on how to prompt it for the best output in the context of Education.
• Looked into two different fameworks SCRIBE and Five “S” framework learnt from AI for Education summit. You can use these framework to guide your prompting in Education context.
• Generated course outline for a “14-week college-level Software Engineering with Python course” using a prompt template from Alchemy.
• Co-designed a prompt framework (CRISP) to guide in prompting Gemini AI for the best educational output.
• Co-Created a prompt template based on the CRISP framework.
• Using the template generated based on the CRISP framework, generated a course content  for the week one topic of course outline previously generated.
• Create a prompt from the template to generate course contents for week 2 topic from the course outline.

Various Educational tasks where generative AI can be used

Generative AI lends itself to many educational tasks and activities as a versatile tool in the hands of educators and learners alike.

Content Creation

• Personalised learning Materials: Generate quizzes, exercises, and practice problems that adapt to the individual learner’s understanding. AI can analyse student performance and create personalised learning pathways with targeted difficulty levels.
• Interactive Simulations: Develop engaging simulations that allow students to experiment with concepts in a safe and controlled environment. Imagine exploring historical events through simulations or practicing scientific experiments virtually.
• Creative Prompts: Generate writing prompts, story ideas, or artistic prompts related to a particular topic. This can spark creativity and encourage learners to explore concepts from different angles.

Assessment and Feedback

• Automated Grading: AI can automate the grading of certain types of assessments, like multiple-choice quizzes or short answer questions. This frees up teachers’ time for more personalised feedback on open-ended assignments.
• Personalised Feedback Generation: AI can analyse student work and provide individualised feedback that highlights strengths and weaknesses.

Practice and Reinforcement

• Dialogue-based Learning: Develop chatbots or virtual tutors that can answer student questions, clarify doubts, and guide them through problem-solving steps. This can provide students with on-demand support and personalised learning experiences.
• Interactive Practice Scenarios: Generate role-playing simulations or case studies where students can apply their knowledge in practical situations. This can be particularly beneficial for subjects like language learning or business communication.
• Gamification: Integrate game mechanics like points, badges, and leaderboards into learning activities. AI can personalise these elements based on student progress, keeping them motivated and engaged.

Open Ended Exploration

• Brainstorming Ideas: Use generative AI to brainstorm research topics, creative writing themes, or project ideas related to a specific subject area. This can help students overcome writer’s block and explore new avenues of learning.
• Concept Mapping and Exploration: Generate visual representations of concepts and their relationships. This can help students visualise complex ideas and identify connections between different topics.

Considerations in using generative AI as a tool in Education

• Clearly defined learning objectives: What specific skills or knowledge do you want students to gain from the activity?
• Target audience: Tailor the task or activity to the age, learning style, and prior knowledge of your students.
• Integration with existing curriculum: Ensure the AI-generated content aligns with your overall course goals and complements other learning materials.
• Teacher oversight and guidance: While AI can be a valuable tool, human educators are still essential for providing context, guiding discussions, and offering personalised support.

Ethical consideration for generative AI

The rise of the generative AI also come along with challenges concerning ethical use and development, especially when it comes to education. There are concerns surrounding issues such as plagiarism, security and data privacy, bias and deepening of the digital divide to mention a few. There are currently no standard approach to addressing these issues but UNESCO recommends a human-centred approach to AI guided by human rights principles, and the need to protect human dignity and the cultural diversity that defines the knowledge commons. Different organisation including government are developing or reviewing their policies to ensure ethics applies when it comes to AI. Education Institutions are also advised to have policies in place to guide the use of AI by Educators and learners alike.

There are 7 Principles of Responsible AI in place at Google, which reflects the current discuss on ethics and responsible use and development of AI

1. Be socially responsible
2. Avoid creating or reinforce unfair bias
3. Be build and tested safely
4. Accountable to the people
5. Incorporate privacy design principle
6. Uphold high standard or scientific excellence
7. Be made available for uses that accord with these principles

Generally, Generative AI are work in progress and will continue to evolve in different directions based on usage, needs and technology evolution. We all can contribute to the evolution of Generative AI by actively using it responsibly in the Education process. “AI can never replace the expertise, knowledge, or creativity of an educator – but it can be a helpful tool to enhance and enrich teaching and learning experiences” – A guide to AI in Education

At Webhostville Services, we actively embrace and continue to explore generative AI for both education and professional purposes. If you are interested in leveraging the latest, tried and tested technology to transform teaching and learning experience for better outcome in Education, we are only a click away, get in touch and look forward to the next iteration of this post series as we explore AI tools for Education.

In line with our workflow process, we have researched, edited and refined some of the contents with a combination of generative AI tools, Gemini AI and Ms copilot.

Links for references and more information

1. Google – Gemini AI
2. Google – Google One AI Premium
3. Google – Bard becomes Gemini: Try Ultra 1.0 and a new mobile app today
4. Google – What are AI hallucination
5. Ms Copilot – Website
6. Achemy – ChatGPT prompts for Higher Education
7. AIforEducation – Prompt Library
8. Google – Bard becomes Gemini: Try Ultra 1.0 and a new mobile app today
9. Google: 7 Principle of Responsible AI
10. Achemy – ChatGPT prompts for Higher Education
11. JISC – Generative AI Primer
12. UNESCO – Guidance for generative AI in education and research
13. UNESCO – Ethics of Artificial Intelligence
14. UNESCO – Generative Artificial Intelligence in education: Think piece by Stefania Giannini
15. WEF – With Generative AI we can reimagine education and the sky is the limit

16. WEF – Generative AI has disrupted education. Here’s how it can be used for good – UNESCO

17. Educause – Integrating Generative AI into Higher Education: Considerations

18. Oxford – Generative AI: the potential of AI in education