Efficient Data Cleaning Techniques in Dataiku?

Miasm1

Hi all,

How do you handle missing values and outliers in Dataiku? Any plugins or workflows you'd recommend for efficient data cleaning?

Thanks for your tips!

Alexandru

Hi @Miasm1
,

1) For missing values -> https://doc.dataiku.com/dss/latest/machine-learning/features-handling/index.html#features-handling you can handle as part of feature handling.
https://knowledge.dataiku.com/latest/ml-analytics/model-design/concept-feature-handling.html#handle-missing-values

2) For outlier on clustering models see : https://doc.dataiku.com/dss/latest/machine-learning/unsupervised/settings.html#outliers-detection

You can use code -> https://developer.dataiku.com/latest/tutorials/plugins/recipes-clipping-dataset/index.html

Or prepare recipe
https://doc.dataiku.com/dss/latest/preparation/processors/number-clipping.html

Turribeach

@AlexT
It's a bot.