> For the complete documentation index, see [llms.txt](https://vikram-bajaj.gitbook.io/cs-gy-6313-information-visualization/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://vikram-bajaj.gitbook.io/cs-gy-6313-information-visualization/data-transformation.md).

# Data Transformation

As discussed earlier, the first step in creating a visualization is **data selection**. This refers to choosing from the data the attributes needed for the visualization.

Once we select the attributes, the next step is usually to perform **aggregation** or other **transformations** in order to be able to create appropriate visualizations, that communicate the required information effectively.

Common aggregation functions include SUM, MAX, MIN, AVERAGE, MEDIAN and STDDEV.

## Transformations Related to Temporal Attributes

These involve transformations at different levels/resolutions: seconds, minutes, hours; days, weeks, months, years etc.

## Transformations Related to Spatial Attributes

These involve transformations at different levels/resolutions: zip code, county, city, state, country etc.

**Geo coding/decoding** refers to going from the name of a place to its geo coordinates and vice versa.

## Binning

It is the process of transforming a **quantitative** attribute to an **ordinal** attribute. Ex. restaurant inspection scores from numbers to letters.

## Rescaling/Re-Expression

There are few ways to accomplish this:

### - Normalization

This refers to bringing all the values to a scale of say \[-1, 1].

### - Percentages

This refers to converting quantitative values to percentages, which makes it easier to make comparisons.