Add Support for Survival Forests, XGBoost Survival Analysis and Kaplain Meier Curves

Dataiku does a great job handling classification and regression models today. It would be really nice to have a similar experience for time to event modeling (survival analysis). There are lots of use-cases like churn and clinical research where whether or not someone takes an action isn't the only point of interest. The time to event also matters. Survival models solve this issue.

 

It would be great to have an experience similar to the classification model experience for Survival analysis.

6 Comments

@M1-Nick Thanks for suggesting these.

--Tom

@M1-Nick Thanks for suggesting these.

seconded!

seconded!

MaximeA
Dataiker

Hello @M1-Nick @importthepandas @tgb417 

Thanks for raising and upvoting this ! This idea has entered our backlog. 

We will let you know if we have any updates

Status changed to: In the Backlog

Hello @M1-Nick @importthepandas @tgb417 

Thanks for raising and upvoting this ! This idea has entered our backlog. 

We will let you know if we have any updates

M1-Nick
Level 3

Awesome!

Awesome!

Excellent. Looking forward to hearing more. Glad to share some further feedback if that would be helpful.  

--Tom

Excellent. Looking forward to hearing more. Glad to share some further feedback if that would be helpful.  

josurriola
Level 2

Hey there, I was wondering if lifetimes type CLV models are being considered in this regard.

Hey there, I was wondering if lifetimes type CLV models are being considered in this regard.