EVALUATE Classification Models using TRUE or FALSE and POSITIVE or NEGATIVE

Humans have an amazing natural capability to take meaningful decisions which makes them superior to machines

Decision is Power. Power Decides .

Hello , Hope you are doing good.

From the bottom of my heart, I would like to Thank the Intelligent Inventors & Developers of the Open Source Softwares & the supportive knowledge-sharing Open Source Community

There are many times when we get confused about doing THIS OR THAT ?

Waking up or Snoozing the alarm ?

Is my Decision right or Wrong ?

Choosing a or b or c or d in MCQ (Multiple Choice Questions)

I am sure most of you would have come across True or False type of questions in exams where we get confused. I have had many shocking heart breaking moments when I had scored 99 out of 100 just because of writing False in the place of True or writing True in the place of False. To Err is Human

Just like humans makes mistakes, even machines make mistakes/ Errors.

Classification especially requires the decision making capability which depends on a number of factors/features.

Classification means grouping into separate categories/labels.

Sometimes machines classify correctly. Other times machines classify wrongly.

There are 4 terms that help us to understand if our classification model works correctly or not.

1. TRUE POSITIVE

2. TRUE NEGATIVE

3. FALSE POSITIVE

4. FALSE NEGATIVE

All the above terms consist of two words.

The first word is either True or False & it describes the model.

The Second word is either Positive or Negative & it is a category.

BONUS TIP:

The first word indicates if our model is good or Bad

True means our model is working correctly

False means our model has Errors

EXPLANATION:

1. True Positive means the model predicted the value as POSITIVE AND the actual value is ALSO POSITIVE

2. True Negative means the model predicted the value as NEGATIVE AND the actual value is ALSO NEGATIVE

3. False Positive means the model predicted the value as POSITIVE but the actual value is NEGATIVE

4. True Negative means the model predicted the value as NEGATIVE but the actual value is POSITIVE

BONUS TABLE:

I have a created a simple , easy to understand table for you all as a Bonus. Enjoy learning!

Though Coronavirus has changed life upside down for many, Coronavirus has taught many lessons to human beings like the importance of Family & Agriculture, Protecting Nature & other living organisms, etc.

I hope you might have come across the news classifying the Coronavirus testing Kits are good or bad

Some Coronavirus testing Kits are good which means the test results are right .

Some testing Kits are not upto the standard which means the test results are wrong .

Let us use this Coronavirus testing Analogy to understand the 4 important terms:

1. If a person has Coronavirus(+) & test results are (+) , then it is True Positive & the testing kit is good

2. If a person doesn’t have Coronavirus(-) & test results are (-) , then it is True Negative & the testing kit is good

3. If a person has Coronavirus(+) & test results are (-) , then it is False Positive & the testing kit is Bad

4. If a person doesn’t have Coronavirus(+) & test results are (-), then it is False Negative & the testing kit is Bad

BONUS TABLE:

I have a created another simple , easy to understand table for you all as a Bonus. Enjoy learning!

I hope this article helped you in understanding the concepts in a clear way. These concepts may look easy from outside, but while working with models, we might get easily confused. Just remember this Coronavirus Testing Analogy whenever you get confused & your confusion will disappear like wind.

Thank you for spending your precious time to read this article. Happy Learning !

I warmly welcome your questions, doubts,suggestions & feedbacks.

Thanks in advance !

If you wish, you can connect with me through Gmail,Twitter, Github,Linkedin