One of the common challenges of Data Analysts is missing values in the datasets.
In this thread, we will see how to deal with missing values in real-world datasets.
A thread π§΅π
In this thread, we will see how to deal with missing values in real-world datasets.
A thread π§΅π
There are mainly ππ΅πΏπ²π² approaches to deal with missing values in the datasets.
πΈ Drop Columns with Missing Values
πΈ Imputation
πΈ An Extension To Imputation
Details about each approach π
2/6
πΈ Drop Columns with Missing Values
πΈ Imputation
πΈ An Extension To Imputation
Details about each approach π
2/6
π‘ Drop columns with missing values: This is the simplest solution to deal with missing values in datasets.
3/6
3/6
π‘ Imputation: This is a better approach than simple drop columns approach. In imputation, missing values will be filled with some number. For example, mean value will be filled for every missing one.
4/6
4/6
π‘ Extension to Imputation: Imputation works well, but imputated values may be above or below their actual values. To handle this case, we keep the track of missing values by adding extra column. π
5/6
5/6
That's all about handling missing values in thee datasets. In the next thread π§΅, I will be sharing the source code for each approach. Staty tuned. If you like this content
πΈ Follow me @itsafiz
πΈ Retweet the first tweet.
6/6
πΈ Follow me @itsafiz
πΈ Retweet the first tweet.
6/6
Loading suggestions...