Weekly Challenges

Solve the challenge, share your solution and summit the ranks of our Community!

Also available in | Français | Português | Español | 日本語
IDEAS WANTED

Want to get involved? We're always looking for ideas and content for Weekly Challenges.

SUBMIT YOUR IDEA

Challenge #157: An Expert Challenge

mhgarcia
8 - Asteroid

My attempt - this one was a fun exercise

mouna_belaid
8 - Asteroid

Please find below my solution

FrederikE
13 - Pulsar
Spoiler
FrederikE_0-1661490265487.png

 

grazitti_sapna
17 - Castor

That is so predictive!!

 

grazitti_sapna_0-1662006862281.png

 

Sapna Gupta
JamesCharnley
13 - Pulsar
Spoiler
JamesCharnley_0-1663077926257.png

 

Thanks to @LiuZhang  for a predictive refresher in Alteryx earlier in the week that unlocked this challenge for me. Was trying to compare random forest models for a while until learning that Logistic was needed. Really great practise and nice to see an example predictive Q

DawnDuong
13 - Pulsar
13 - Pulsar

Thank for posting this challenge. I learn a new tool and also correct my previous wrong conceptual understanding of the various tools.

Real respect for the Predictive Master Certified folks.

trung2403
8 - Asteroid

A very very tough one! I had to google around to find out where to get the Variable importance plot. Normally, I will use Association Analysis to select the most important variables, dont know that we can use Forest Model to select variables. Then how to compare 2 models based on the Chi-square is another challenging task. In the past, i only used Model Comparison or Lift Chart, never thought of using Nested Test. It will be a very long journey before i can take the Predictive Master Cert. Thanks for the challenge. Hope that we can have more similar challenges soon.

Spoiler
trung2403_0-1665884844259.png

 

acarter881
12 - Quasar

Here's my solution.

Spoiler
my_weekly_challenge_157_solution.png
martinding
13 - Pulsar
Spoiler
157.png

ARussell34
8 - Asteroid

Glad to know now that the Forest Model tool outputs a chart containing this information and that a Nested Test tool can compare two models.