Crash Course AI: Pet Adoption Decision

Jabril explores using AI to decide between adopting a cat or a dog. He builds a neural network model using Python and Google Colaboratory to predict pet happiness based on features like cuddliness and energy. However, he discovers bias in the model favoring dogs due to skewed survey data. Jabril analyzes the data, identifies the bias, and plans to correct it by collecting balanced data and refining features.

What was the main reason Jabril decided to use AI for his pet adoption decision?

• To save time on research

• To find the cheapest pet

• To avoid personal bias

• To impress his friends

Which tool did Jabril use to write his AI code?

• Jupyter Notebook

• Google Colaboratory

• Visual Studio Code

• Atom

What are the four features Jabril considered important for predicting pet happiness?

• Soft, friendly, quiet, playful

• Cuddly, soft, quiet, energetic

• Cuddly, playful, loyal, energetic

• Friendly, loyal, playful, quiet

What type of neural network did Jabril use for his AI model?

• Recurrent Neural Network

• Generative Adversarial Network

• Convolutional Neural Network

• Multi-Layer Perceptron

What was the accuracy of Jabril's AI model on the testing data?

• 90%

• 100%

• 85%

• 75%

What unexpected bias did Jabril notice in his AI model's predictions?

• Bias towards birds

• Bias towards turtles

• Bias towards dogs

• Bias towards cats

What was one of the main reasons for the bias in Jabril's AI model?

• Correlated features

• Limited data

• Too many features

• Incorrect coding

How did Jabril plan to fix the imbalance in his dataset?

• By using a different AI model

• By collecting more data from cat owners

• By changing the features

• By removing dog data

What feature did Jabril consider removing to avoid bias in his AI model?

• Cuddly

• Soft

• Quiet

• Energetic

What is a key takeaway from Jabril's experience with building AI systems?

• AI systems are always accurate

• AI systems require no human oversight

• AI systems are easy to build

• AI systems need careful design and iteration