$1\mathrm{.}1.$A$.$ Preliminary data exploration, understanding and minor corrections in Excel file:

The starting point for this subtask is the dataset itself, including the data dictionary and preparation

tasks$/$notes in the second worksheet $($Raw$\_$data$\_$dict$).$

The next suggested step is to filter each column in the raw data to see whether there are blanks

$($missing values$)$ or obvious data entry errors; the latter should be corrected in the Excel spreadsheet

already before loading. The anomaly in the column $$Dependents$$ should also be corrected in the Excel

spreadsheet first, making reasonable assumptions. Finally, add and generate the variables

Income$\_$total and Loan$\_$Income$\_$ratio as explained in the data dictionary tab.

$1\mathrm{.}1.$B$.$ Data preparation for multiple analysis tasks:

After the initial error removal in the source file $(1.$A$.),$ you now have to perform the following data

preparation steps in that order:

$$ Missing value handling$/$transformation;

$$ Data transformation for statistical analysis $($e$.$g$.$ string$/$cat to numeric, one to many$)$;

$$ Outlier identification and treatment.

For this sub$$task$,$ you have essentially three options:

$1)$ Use KNIME workflows for each analysis task or all tasks in one single workflow;

$2)$ Use only EXCEL or any other tool and load the data for each analysis task;

$3)$ Use a combination of the two options above $($e$.$g$.$ create a KNIME workflow to prepare the data

and write or copy them to Excel files after outlier correction, and then use this for analysis tasks in

Excel and KNIME; OR: treat missing variables and convert category variables already in Excel do

outlier analysis only in KNIME.