So you’ve heard about ChatGPT and want to explore it but don’t know where to start? This is the second in a short series of blog posts giving some pointers on how to get going.
If you’ve not done so already, we’d recommend reading the first blog before you get started, where we cover some of the limitations of ChatGPT.
This post is for the free version of ChatGPT. You might have heard that GPT-4 has been announced, but at the moment is only available for the paid version. This will be updated when that changes!
We are now going to explore how ChatGPT can be used to generate ideas. We’ll focus on lesson and course planning, but the same approach can be applied to any activity.
Drafting ideas for lesson plans and other activities
To get the most out of ChatGPT you need to provide it with clear instructions on what you want to achieve. We’ll look at some of examples of this now.
A simple lesson plan
We’ll start with a simple example. I want a lesson plan for a session about chatbots. The additional information I’m going to provide in the prompt is:
- The length of the lesson (2 hours)
- The level of the students (‘introductory’).
- Additional information – I want a 30-minute group activity.
So the whole prompt is as follows:
Create an outline plan for a two hour lesson on chatbots, for students on an introductory course. Include one 30 minute group activity.
Here’s the lesson plan that ChatGPT created for me:
I. Introduction to Chatbots (15 minutes)
- Definition of chatbots and their purpose
- Types of chatbots (rule-based vs. AI-powered)
II. Chatbot Design and Development (30 minutes)
- Design considerations for chatbots (e.g. personality, user interface)
- Development tools and platforms (e.g. Dialogflow, Botpress)
III. Chatbot Use Cases (15 minutes)
- Examples of chatbots in various industries (e.g. customer service, healthcare, e-commerce)
IV. Chatbot Implementation (30 minutes)
- Integration with messaging platforms (e.g. Facebook Messenger, Slack)
- Testing and deployment strategies
V. Ethics of Chatbots (15 minutes)
- Discussion of ethical considerations in chatbot design and use (e.g. privacy, bias)
VI. Group Activity (30 minutes)
- Divide students into small groups
- Assign each group a different industry (e.g. healthcare, banking, education)
- Have each group brainstorm and design a chatbot for their assigned industry, considering design, development, implementation, and ethics
VII. Conclusion and Next Steps (15 minutes)
- Recap of key concepts covered in the lesson
- Suggestions for further learning and practice (e.g. creating a chatbot project)
Overall, this lesson plan includes a mix of theory and practical activities, including a 30-minute group activity where students can apply what they’ve learned.
Is this perfect and ready to use? Probably not – the timings look quite ambitious for a start though! Could it be useful to start planning though? Hopefully.
Also, don’t forget that we learnt in our first blog post that ChatGPT will sometimes get facts wrong, so make sure you check the output carefully before using it.
A course plan based on learning outcomes
We’ll now look at a more complex example – using learning outcomes to plan a course.
In our imaginary course, we have these as the learning outcomes:
After completing this module, students will be expected to be able to:
- Define what a chatbot is and explain the purposes and benefits of using chatbots.
- Design and develop a chatbot using appropriate tools and techniques.
- Create a chatbot that incorporates natural language processing (NLP) and machine learning (ML) algorithms to improve user experience.
- Implement and integrate a chatbot with a messaging platform (e.g. Facebook Messenger, Slack).
- Evaluate and optimize the performance of a chatbot using analytics and metrics (e.g. user engagement, response time, retention rates)
Again, to get the most out of ChatGPT I’m going to provide additional information:
- Length of course
- Length of lectures
- Additional information: I’d like a final assessment idea.
We can combine this information into a longer prompt.
This is the full prompt:
Create a 3 week course outline for a course for an introduction to chatbots. There is a two hour lecture per week and I’d like an idea for the final assessment. These are the learning outcomes:
After completing this module, students will be expected to be able to:
Define what a chatbot is and explain the purposes and benefits of using chatbots.
Design and develop a chatbot using appropriate tools and techniques.
Create a chatbot that incorporates natural language processing (NLP) and machine learning (ML) algorithms to improve user experience.
Implement and integrate a chatbot with a messaging platform (e.g. Facebook Messenger, Slack).
Evaluate and optimize the performance of a chatbot using analytics and metrics (e.g. user engagement, response time, retention rates).
Here’s the full course outline that ChatGPT has proposed:
Course: Introduction to Chatbots
Topic: Introduction to Chatbots
- Definition of chatbots and their applications
- Types of chatbots: rule-based vs. AI-based chatbots
- Benefits of using chatbots
Topic: Designing and Developing a Chatbot
- Designing a conversation flow for a chatbot
- Building a chatbot using appropriate tools (e.g. Dialogflow, Botpress)
- Best practices for developing chatbots
Topic: Natural Language Processing and Machine Learning
- Introduction to Natural Language Processing (NLP) and Machine Learning (ML)
- Incorporating NLP and ML algorithms in chatbots
- Improving the user experience through NLP and ML
Topic: Integrating a Chatbot with a Messaging Platform
- Integrating a chatbot with popular messaging platforms (e.g. Facebook Messenger, Slack)
- Configuring the chatbot for different platforms
- Best practices for integrating a chatbot with a messaging platform
Topic: Evaluating and Optimizing Chatbot Performance
- Understanding key performance metrics for chatbots (e.g. user engagement, response time, retention rates)
- Evaluating chatbot performance using analytics tools
- Optimizing chatbot performance based on user feedback and analytics
Topic: Final Project and Assessment
Students will develop and present their own chatbot project incorporating the concepts learned in the course
The assessment will include a presentation of the chatbot and a report on its performance metrics and optimization strategies
Note: This course outline can be adapted to fit a longer or shorter timeframe, depending on the level of depth and complexity desired.
Again, we are not suggesting that ChatGPT will generate a perfect outline, but it might be useful for coming up with initial ideas that you can then refine.
Hopefully that has given you some ideas. Try a similar approach for other activities, maybe writing an article or blog post.
Next we are going to look at creating multiple choice quizzes.
Find out more by visiting our National centre for AI page to view publications and resources, join us for events and discover what AI has to offer through our range of interactive online demos.
For regular updates from the NCAI sign up to our mailing list.
Get in touch with the team directly at NCAI@jisc.ac.uk