Scribd
Upload
307 views8 pages

Worksheet Titanic Python PDF

This project uses machine learning to predict whether Titanic passengers survived based on their attributes like gender, age, class, etc. The student trains a model on historical passenger data, then tests it on characters from the movie Titanic, predicting correctly that Jack survived but Rose's survival matched the movie. The student has gained experience applying machine learning to a real-world problem and recognizing patterns in data.

Uploaded by

rashmimeh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
307 views8 pages

Worksheet Titanic Python PDF

This project uses machine learning to predict whether Titanic passengers survived based on their attributes like gender, age, class, etc. The student trains a model on historical passenger data, then tests it on characters from the movie Titanic, predicting correctly that Jack survived but Rose's survival matched the movie. The student has gained experience applying machine learning to a real-world problem and recognizing patterns in data.

Uploaded by

rashmimeh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

Titanic

The RMS Titanic sank in the North Atlantic Ocean in 1912 during her
maiden voyage from Southampton to New York. More than 1,500 of
the 2,220 passengers and crew died when the ship sank.

In this project you will make a Python program that can predict
whether a specific passenger survived.

You’ll use machine learning to train a predictive model using


information about hundreds of examples of the passengers.

This project worksheet is licensed under a Creative Commons Attribution Non-Commercial Share-Alike License
http://creativecommons.org/licenses/by-nc-sa/4.0/

Page 1 of 8 Last updated: 7 July 2019


1. Go to https://machinelearningforkids.co.uk/ in a web browser

2. Click on “Get started”

3. Click on “Log In” and type in your username and password


If you don’t have a username, ask your teacher to create one for you.
If you can’t remember your username or password, ask your teacher or
group leader to reset it for you.

4. Click on “Projects” on the top menu bar

5. Click the “Copy template” button.

6. Find the Titanic survivors project template and click on the “Import
Titanic survivors” button.
A description of the passenger data is displayed. Click “Import”.

7. Click on your new “Titanic survivors” project

Page 2 of 8 Last updated: 7 July 2019


8. Start by displaying the training data that you’ve just copied.
Click the “Train” button.

9. There are two training buckets.


“survived” contains examples of passengers/crew who survived.
“did_not_survive” bucket contains examples of people who did not.
Each example gives you some information about the person:
ticket class 1, 2, 3 Did they have a first class, second class, or third class ticket?
gender male, female Were they male or female?
age age in years How old were they?
sibl. sp.siblings or If they’re a child, how many brothers and sisters were also on board?
spouses If they’re an adult, 1 if they had a husband or wife with them, or 0 if not
par. ch. parents or If they’re a child, how many parents they had on board with them.
children If they’re an adult, how many children they had on board with them.
ticket fare cost in pounds How much did their ticket cost?
embarked Cherbourg, Where did they get on board? Cherbourg (France), Southampton
Southampton, (England), or Queenstown (Ireland)?
Queenstown

Page 3 of 8 Last updated: 7 July 2019


10. Review the training examples. Can you see any patterns in the data?
Are there any common differences between passengers that survived,
and passengers that didn’t?
For example, have you heard the phrase “Women and children first!”
If that was really how the passengers and crew were got to the lifeboats,
what sort of pattern would you expect to see in the training examples?
Can you find that sort of pattern in the data in your project?

What other patterns can you think of? For example, do you think
passengers with more expensive tickets would have a better or worse
chance of surviving the sinking?

Try to predict what sorts of patterns your machine learning model could
learn to recognize. And then look to see if you can find that sort of pattern
in your training data.

11. Click on the “< Back to project” link.

12. Click on the “Learn & Test” button.

13. Click on the “Train new machine learning model” button.


The example passenger data will be used to train a predictive model.

Page 4 of 8 Last updated: 7 July 2019


We’ll test your model using the characters from the 1997 movie, Titanic.

Jack Dawson (played by Leonardo DiCaprio)


Jack was a 20 year-old man.
He paid nothing for his third-class ticket, because he won it in a poker game.
He boarded RMS Titanic in Southampton, in England.
He had become an orphan when he was 15, and had no other family on board.

Rose DeWitt-Bukater (played by Kate Winslet)


Rose was a 17 year-old woman.
She had a first-class ticket. The film doesn’t mention exactly how much it cost,
but based on the room that she had, we can estimate her ticket cost £450.
She boarded the ship in Southampton, England.
She came on board with her mother and her fiancé. She didn’t have any
brothers or sisters on board.

14. Click the “< Back to project” link

15. Next, we’ll use Python. Click “Make”

16. Click on the Python button

Page 5 of 8 Last updated: 7 July 2019


17. Fill in the information about the Jack Dawson character in the fields
displayed on the left.
This will update the skeleton code on the right.

18. Copy the updated skeleton code displayed into a text editor.
Save it as a file called jack.py

19. Find the last line of the skeleton code:


print ("result: '%s' with %d%% confidence" % (label, confidence))

Change the last line to this:


print ("%d%% sure that Jack %s" % (confidence, label))

20. Run your program with the command: python jack.py


Your machine learning model will display its prediction for whether Jack
survived the sinking of the Titanic.
Does the prediction match what happened in the movie?

Page 6 of 8 Last updated: 7 July 2019


21. Try doing the same for Rose’s character in the movie.
Fill in the information about Rose’s character in the webpage boxes.
Save the updated skeleton code to a file called rose.py
Update the last line of code to say:
print ("%d%% sure that Rose %s" % (confidence, label))
Run the program using the command: python rose.py

Your machine learning model will display its prediction for whether Rose
survived the sinking of the Titanic.
Does the prediction match what happened in the movie?

What have you done?

You’ve used a historical set of information about passengers and crew of


the RMS Titanic to train a predictive model. That predictive model used
the patterns in your historical training data to learn to recognize patterns
in what sort of passengers survived or not.

When you looked for patterns in the training data, you hopefully thought
about possible reasons to explain it. For example, younger passengers
were more likely to survive because children would’ve been prioritized
when boarding lifeboats. Computers won’t have done that last bit. The
computer will hopefully have seen the pattern between age and survival
but won’t have tried to explain it. A machine learning model will learn to
recognize patterns in the data, but they won’t care about why.

Finally, you made Python programs to test your predictive model, using
information about fictional Titanic passengers from the movie, Titanic.

Page 7 of 8 Last updated: 7 July 2019


Ideas and Extensions

Now that you’ve finished, why not give one of these ideas a try?

Or come up with one of your own?

Try testing with other passengers

Can you find any information about other Titanic passengers


or crew that you can test your predictive model on?

Invent your own predictive model

Can you think of another topic that you could collect


numerical or multiple-choice values about?

Page 8 of 8 Last updated: 7 July 2019

You might also like