Scribd
Upload
32 views3 pages

Unsupervised and Supervised Learning Problems: Problem 1

The document discusses four problems and whether they are examples of supervised or unsupervised learning. Problem 1 and 3 are supervised as they involve predicting labels. Problem 2 and 4 are unsupervised as they involve discovering patterns or groups without labels.

Uploaded by

smorshed03
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
32 views3 pages

Unsupervised and Supervised Learning Problems: Problem 1

The document discusses four problems and whether they are examples of supervised or unsupervised learning. Problem 1 and 3 are supervised as they involve predicting labels. Problem 2 and 4 are unsupervised as they involve discovering patterns or groups without labels.

Uploaded by

smorshed03
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/ 3

Unsupervised and Supervised Learning Problems

Let’s look at some Data analysis problems and state whether they are supervised or
unsupervised problems:

Problem 1:
Suppose we have measured the concentration in the air of a certain type of pollution.
We have measured the amount of pollution at different locations, under different
weather conditions at different times of the day and for different days of the week.

Now we would like to predict the air


pollution concentration at a new location
where weather and date are given.

This is an example of Supervised Learning.


Here the data points are the collection of
information about the surroundings like
Location, Weather, Date & time. And the
[email protected]
JQ36809TC1 labels are air pollution concentration which is a numerical value.

Problem 2:
When Frank Simpson was a Ph.D. student in the 60’s he spent some time in a
Monastery. He followed the novices’ training there and recorded their interactions with
each other.
Now we want to discover what the friendships or social groups among the novice
monks are.

This file is meant for personal use by [email protected] only.


Sharing or publishing the contents in part or full is liable for legal action.
This is an Unsupervised Learning problem. As here, we don’t have any labels that tell
us who belongs to which group and are there any social groups that exist.
Here we have to find the pattern in the data in an unsupervised way.

We have some similar kinds of problems like this one. For example: Detecting
communities or groups in large social networks like Facebook or Twitter.

Problem 3:
NASA uses rocket boosters to launch spacecraft into orbit. It’s expensive to build a
whole new rocket booster for every slight tweak and it’s difficult to recreate real
atmospheric conditions in a wind tunnel for testing these out. So, NASA uses computer
simulations, but it takes a long time to run the computer simulation for different rocket
boosters specifications.

[email protected]
JQ36809TC1

So, NASA would like to measure the simulated lift, for different rocket booster
specifications and predict the simulated lift under other given specifications.

This problem is a Supervised Learning Problem. Here the data points are the rocket
booster’s specifications and the labels are the simulated lift.

This file is meant for personal use by [email protected] only.


Sharing or publishing the contents in part or full is liable for legal action.
Problem 4:
This problem is a little different than the previous problems.

Suppose, we have small cameras that we have installed to monitor bird’s nests in New
England. But, we don’t have a lot of Bandwidth to send the video. So, we have to
compress the recorded video before sending it.

We can think of this as an Unsupervised Learning Problem. If we consider the pixels at


[email protected]
JQ36809TC1 each timestamp as a data point, then we can group together the pixels which are
similar in color and similar in time to achieve the compression.

This file is meant for personal use by [email protected] only.


Sharing or publishing the contents in part or full is liable for legal action.

You might also like