22.2.3 - Steps to Perform
Enroll to start learning
You’ve not yet enrolled in this course. Please enroll for free to listen to audio lessons, classroom podcasts and take practice test.
Interactive Audio Lesson
Listen to a student-teacher conversation explaining the topic in a relatable way.
Creating an AI Model
🔒 Unlock Audio Lesson
Sign up and enroll to listen to this audio lesson
Today, we are focusing on how to create an AI model! Can anyone tell me what the first step is?
Is it to choose the type of data we will use?
Exactly! We need to decide between images, text, or sound. Who can give me an example of what data we might use for an image classifier?
We could use images of happy and sad faces!
Great example! Once we have our data, the next step is to collect samples for different classes. What do you think collecting samples involves?
It means taking a variety of pictures or sounds that fit into our categories.
You are on the right track! After collecting your data, we'll train the model using our chosen tools like Teachable Machine. Can anyone think of what comes next?
We need to test and refine the model to get better results!
Exactly! Finally, you'll present your findings. Remember, this process helps you understand how machines learn patterns. Great job everyone!
Understanding Sustainable Development Goals
🔒 Unlock Audio Lesson
Sign up and enroll to listen to this audio lesson
Now, let's dive into the second project where we solve a problem related to the Sustainable Development Goals. Does anyone know what SDGs are?
They are goals set to tackle global challenges like poverty and climate change.
Exactly! So, we want to think about local problems we can solve. What might be some areas we could focus on?
Pollution, like air or water pollution!
Energy conservation is another big issue.
Great suggestions! After choosing a problem, we will create a 4Ws Problem Canvas. Can anyone remind us what the 4Ws stand for?
Who, What, Where, and Why!
Perfect! Once we complete the canvas, we will identify features that contribute to the problem and use data collection for our AI-enabled solutions. Exciting prospects!
Data Collection and Visualization
🔒 Unlock Audio Lesson
Sign up and enroll to listen to this audio lesson
Now that we have defined our problem, let's discuss data collection and visualization. Why is data important?
Data helps us understand the problem better!
Right! We can use spreadsheets to record our collected data. What types of visuals can help us present data effectively?
Bar charts and pie charts are great for showing comparisons!
And line graphs can show changes over time.
Excellent observations! Finding patterns in your data will be crucial for our AI solution. Can anyone suggest a potential AI-enabled solution based on our data?
Maybe we could create an app that sends pollution alerts based on air quality data!
Fantastic idea! Collecting and visualizing data can lead to powerful solutions that make a difference.
Introduction & Overview
Read summaries of the section's main ideas at different levels of detail.
Quick Overview
Standard
The section details the steps students should take to complete projects involving AI models and solutions for problems related to Sustainable Development Goals. Students will be guided through selecting data types, training models, and understanding problem-solving frameworks.
Detailed
Steps to Perform
This section provides a systematic approach for students to engage in projects that integrate Artificial Intelligence (AI) with real-world applications, emphasizing collaboration and critical thinking. The two primary projects involve building an AI model and addressing issues aligned with the Sustainable Development Goals (SDGs).
Project Overview
- Create an AI Model: Students will select data types (images, text, or sound), gather samples, train the model using user-friendly tools like Teachable Machine or Machine Learning for Kids, refine their results, and present their findings.
- Solve a Problem Related to SDGs: Students will identify a local problem (e.g., pollution, water wastage), employ the 4Ws Problem Canvas to analyze the issue, collect data, visualize results, and create an AI-enabled solution, like an app for pollution alerts.
Suggested Activities
- Field visits to observe AI in action.
- Maintaining a student portfolio to document their AI learning journey.
Benefits of These Projects
Understanding AI through practical activities enhances data literacy, promotes teamwork, fosters innovative solutions, and creates awareness about global challenges. The focus is on merging technology with a purposeful approach to education.
Audio Book
Dive deep into the subject with an immersive audiobook experience.
Choose a Type of Data
Chapter 1 of 5
🔒 Unlock Audio Chapter
Sign up and enroll to access the full audio experience
Chapter Content
- Choose a type of data (Images/Text/Sound).
Detailed Explanation
In this first step, you need to decide what kind of information you want your AI model to work with. You can choose images, text, or sounds. Each type of data will give you a different way to train your model. For example, if you choose images, you might train your model to recognize different objects or faces. If you opt for text, your model could learn to understand sentiments or categorize news articles.
Examples & Analogies
Think of it like a chef choosing the main ingredient for a recipe. Just as a chef selects chicken, vegetables, or fish based on the dish they want to prepare, you select images, text, or sound based on what task you want your AI model to perform.
Collect Sample Data
Chapter 2 of 5
🔒 Unlock Audio Chapter
Sign up and enroll to access the full audio experience
Chapter Content
- Collect sample data for different classes.
Detailed Explanation
Here, you gather examples of each category you've chosen to work with. For instance, if you're working with images to recognize happy and sad faces, you need to collect enough images of happy faces and sad faces. This collection of data is essential, as the model learns to identify and differentiate between these categories based on the examples you provide.
Examples & Analogies
Imagine building a library of books for a study class. Just like a student collects different books on various subjects to understand those topics better, you collect different examples to help your AI model learn to identify and classify the types of data accurately.
Train the Model
Chapter 3 of 5
🔒 Unlock Audio Chapter
Sign up and enroll to access the full audio experience
Chapter Content
- Train the model using the tool.
Detailed Explanation
In this step, you use the collected sample data to teach your model. Training means inputting the data into the tool you've selected, allowing the AI to learn the patterns associated with each category you provided. The AI will analyze the characteristics of the data and create algorithms that help it recognize these patterns when new data is introduced.
Examples & Analogies
It's like training a dog. Just as you would show your dog commands like 'sit' or 'stay' repeatedly until it understands and responds, you show your model the data numerous times so it learns to recognize the patterns within.
Test and Refine the Model
Chapter 4 of 5
🔒 Unlock Audio Chapter
Sign up and enroll to access the full audio experience
Chapter Content
- Test and refine the model.
Detailed Explanation
Once the model is trained, you need to test how well it performs. You input new data that the model has not seen before and see if it can accurately recognize or categorize it correctly. If the model doesn’t perform well, you may need to refine it by adjusting the data used, modifying the algorithm, or providing more training examples.
Examples & Analogies
Think of this like taking a practice exam before the real test. You can identify areas where you need more study or practice to improve your score, just as you adjust your AI model based on its testing performance.
Present the Output
Chapter 5 of 5
🔒 Unlock Audio Chapter
Sign up and enroll to access the full audio experience
Chapter Content
- Present the output and share your learning.
Detailed Explanation
After your AI model has been successfully trained and tested, the final step is to present its results. This may involve showcasing how the model works, what it can recognize, and sharing the process of your learning experience. This step is crucial in helping others understand the model's capabilities and the value of the AI project.
Examples & Analogies
It’s like giving a science presentation where you display your findings and explain your research process. Just as you would want your audience to understand what you discovered and learned, you want to communicate the outcomes of your AI project effectively.
Key Concepts
-
AI Model: A trained system that learns from data to perform specific tasks.
-
Sustainable Development Goals (SDG): Global targets aimed at creating a better world.
-
4Ws Canvas: A structured approach to deeply analyze problems.
-
Data Visualization: The process of representing data visually to extract insights.
Examples & Applications
An image classifier recognizing emotions based on facial expressions.
A mobile app providing alerts on local air quality.
Memory Aids
Interactive tools to help you remember key concepts
Rhymes
Data and AI, create and try, solve the world's problems, reach for the sky!
Stories
Imagine a world where pollution is controlled, thanks to an app created by students who learned AI. They used their knowledge to analyze data and develop a smart solution.
Memory Tools
Remember the 4Ws: 'Who, What, Where, Why' helps us to analyze problems clearly.
Acronyms
For AI model steps, remember 'C-T-T-P' for Collecting data, Training, Testing, Presenting.
Flash Cards
Glossary
- AI Model
A system designed to perform tasks autonomously by learning from data.
- Sustainable Development Goals (SDGs)
A set of 17 global goals aimed at addressing challenges to achieve a better and more sustainable future.
- 4Ws Problem Canvas
A framework used to analyze an issue by asking Who, What, Where, and Why.
- Data Visualization
The graphical representation of data to understand trends, patterns, and insights.
Reference links
Supplementary resources to enhance your learning experience.