Introduction to Artificial Intelligence

Imagine that you work for a university that wants to use machine learning and Naive Bayes to predict which students might have difficulty graduating. So you create three predictors. These are financial hardship, grade point average and class attendance. In a meeting, a data scientist points out that you might not want to use class attendance and grade point average because they are strongly autocorrelated. If someone doesn't attend class, then they'll likely get a poor grade. How might you answer this question?

• Naive Bayes is Naive because it can classify even when predictors are autocorrelated

• class attendance and grade point average are not closely related

• it might be a good idea to change the predictors so that they are not correlated

• Naive Bayes is naive because it doesn't need class predictors to classify data

Why is it important to understand different machine learning algorithms?

• they are like kitchen knives , you can use one of them to solve all your problems

• they are tools to help you decide what reports you like to see and problem you like to solve

• it is important to see which algorithms won't work with other algorithms

• you can never change your algorithm so its important to decide everything up front

Imagine that you work for a health insurance company. Your company covers a lot of people who suffer from diabetes. The organization wants to research the characteristics of people with diabetes. That way the company can intervene and try to change the customers’ behavior to prevent future illness. How can your insurance company use K means clustering to help?

• Make sure the k in k means clustering is 6 , because you need lot of clusters.

• Try to keep the value of k is 1,that way you can work with just one big cluster

• research the character traits of the cluster with the highest number of diabetics

• use supervised learning to train the system to find diabetics

In supervised machine learning what's the difference between training data and test data?

• Training data is data machine uses to learn and test data is sample of data to test what is learned

• Training data is all the data that the system can gather , then the test data is small sample used for training

• Test data is where all the learnings take place before its trains on large sample of data

• There is not a difference , you can mix and match training data and test data

You're a product manager for a team that's using an artificial neural network to develop your product. One of the data scientists says that the back propagation of errors is correcting for guesses that have a steep gradient descent. What is that saying about the network?

• The network is making predictions that are turning out to be very wrong

• the network back propagation is not finding that many errors.

• the network is very close to find the correct prediction

• the network is making predictions that will have very low cost function.