Aggregation

Aggregation
Description	Aggregation of categorical data involves the gathering of information for statistical analysis and expressing it in a summarized form.
Why to use	Numerical Analysis – Data Preparation
When to use	When you want to collect specific information about particular groups based on specific variables.	When not to use	On textual data.
Prerequisites	It should be used on numerical and categorical data.
Input	Any dataset that contains categorical as well as numerical data.	Output	Aggregated numerical or categorical data.
Statistical Methods used	Sum Mean Mode Minimum Maximum Count Standard Deviation Variance	Limitations	Sometimes using only aggregation is not enough as it gives only single level analysis. You may need to use other methods to get accurate results.

Aggregation is located under Model Studio () in Data Preparation, in the task pane on the left. Use drag-and-drop method to use the algorithm in the canvas. Click the algorithm to view and select different properties for analysis. Refer to Properties of Aggregation.

Aggregation is a group-by algorithm in which a given data is grouped for a certain categorical data variable like name, date, color, educational level and so on. The data that is grouped is the numerical data and is called the Aggregate Function. You cannot use this algorithm unless you have selected the GroupBy function.

Properties of Aggregation

The available properties of Aggregation are as shown in the figure given below.

The table given below describes different fields present on the properties of Aggregation.

Field	Description	Remark
Task Name	It displays the name of the selected task.	You can click the text field to edit or modify the name of the task as required.
GroupBy	It allows you to select the function for which you want to group the data.	Multiple functions can be selected. You can group by numerical as well as categorical function.
Aggregate Function	It allows you to select the type of data that is to be aggregated.	Multiple functions can be selected. Same data can be grouped according to different statistical measures at the same time. You can aggregate numerical as well as categorical data. Numerical data is aggregated by Sum Mean Mode Minimum Maximum Count Standard Deviation Variance Categorical data is aggregated by Minimum Maximum Count

Interpretation from Aggregation

The aggregation algorithm can be used on either the numerical data or the categorical data. The table given below describes the results for Aggregation.

Aggregation Method	Result	Remark
Sum	It gives the aggregation of the given data by the sum of the values in that column.	It can be performed only on numerical data.
Mean	It gives the aggregation of the given data by mean of the values in that column.	It can be performed only on numerical data.
Mode	It gives the aggregation of the given data by mode of the values in that column.	It can be performed only on numerical data.
Minimum	It gives the aggregation of the given data by the minimum value of the variable in that column.	It can be performed on numerical as well as categorical data.
Maximum	It gives the aggregation of the given data by the maximum value of the variable in that column.	It can be performed on numerical as well as categorical data.
Count	It gives the aggregation of the given data by the count of the variable in that column.	It can be performed on numerical as well as categorical data.
Standard Deviation	It gives the aggregation of the given data by the standard deviation of the variable in that column.	It can be performed only on numerical data.
Variance	It gives the aggregation of the given data by the variance of the variable in that column.	It can be performed only on numerical data.

Example of Aggregation

The figure given below displays the output of aggregation performed on sample data. The data of the number of deaths (numerical data) in a US county is aggregated by sum, mean, standard deviation, and the maximum value of the number of deaths. The data is grouped by the name of the county and date (both categorical data).

Field	Result
county	It displays the name of the US county whose data corresponding to the number of deaths is aggregated.
deaths_Aggr_0	It displays the aggregate deaths in that county by the sum of deaths on a particular date.
date	It displays the date corresponding to which the data is aggregated.
deaths_Aggr_1	It displays the aggregate deaths in that county by the mean number of deaths on a particular date.
deaths_Aggr_2	It displays the aggregate deaths in that county by the standard deviation of the deaths on a particular date.
deaths_Aggr_3	It displays the aggregate deaths in that county by the maximum value of the number of deaths on a particular date.

Table of Contents