CanvasXpress Data Wrangling

Introduction to Data Wrangling

Data wrangling is a crucial step in any data analysis workflow, involving the process of cleaning, structuring, and enriching raw data into a more usable format. In CanvasXpress, this process is streamlined with a powerful set of functionalities that allow you to manipulate your data directly within the browser, without the need for server-side processing.

This section provides a comprehensive guide to the various data wrangling methods available in CanvasXpress, enabling you to prepare your datasets for effective visualization and analysis. These functions can be applied programmatically through the API or interactively via the user interface.

Data Wrangling in CanvasXpress

Data wrangling, which involves restructuring, enriching, and transforming data into a more valuable format, is important for various purposes, including analytics. CanvasXpress applies a data wrangling process across data sets, integrates information from various data sources to solve problems, and achieve analytical goals.

The CanvasXpress library has a unique set of functionalities that support data-wrangling without the need to contact a server. This result-oriented function empowers our end users to effectively access their data without any hassle. These functions can be called programmatically or used directly through one of the UIs.

Consider the following dataset to showcase some of these methods.

Grouping Data

Grouping can be done through the CanvasXpress configuration

new CanvasXpress(targetId, data, {
  graphType: "Boxplot",
  // ... other configurations ...
  groupingFactors: ["Gender"]
});

or a function call after the visualization has been rendered

var cX = new CanvasXpress(targetId, data, config);
cX.groupSamples(["Gender"]);

Facets

Facets can be done through the CanvasXpress configuration

new CanvasXpress(targetId, data, {
  graphType: "Bar",
  ...,
  segregateSamplesBy: ["Excercise"]
});

or a function call after the visualization has been rendered

var cX = new CanvasXpress(targetId, data, config);
cX.segregateSamples(["Excercise"]);

Sort Data

Sorting can be done through the CanvasXpress configuration

// Sort data by Age
new CanvasXpress(
  targetId,
  data, {
  graphType: "Bar",
  sortData: [['var', 'smp', 'Age']]
});

//Sort data by Height
new CanvasXpress(
  targetId,
  data, {
  graphType: "Bar",
  sortData: [['cat', 'smp', 'Height']]
});

or a function call after the visualization has been rendered

var cX = new CanvasXpress(targetId, data, config);

// Sort by Age
cX.modifySort(['var', 'smp', 'Age']);

// Sort by Height
cX.modifySort(['cat', 'smp', 'Height']);

Correlating Data

Data can be correlated to its meta-data

Correlation can be done through a function call

var cX = new CanvasXpress(targetId, data, config);
cX.correlateData(true, "Hip");

Histogram Data

Histograms can be created through the CanvasXpress configuration

new CanvasXpress(
  targetId,
  data, {
  graphType: "Bar",
  showHistogram: true
});

or a function call after the visualization has been rendered

var cX = new CanvasXpress(targetId, data, config);
cX.createHistogram();

Clustering Data

Cluster data can be done through the CanvasXpress configuration

new CanvasXpress(
  targetId,
  data, {
  graphType: "Bar",
  samplesClustered: true
});

or a function call after the visualization has been rendered

var cX = new CanvasXpress(targetId, data, config);
cX.clusterSamples();

Transform Data

Data transformation can be done through a funciton call

var cX = new CanvasXpress(targetId, data, config);
cX.transform("log2");

Transpose Data

Transposing data can be done through the CanvasXpress configuration

new CanvasXpress(
  targetId,
  data, {
  graphType: "Bar",
  transposeData: true
});

or a function call after the visualization has been rendered

var cX = new CanvasXpress(targetId, data, config);
cX.transpose();

Pivot Data

Pivot data can be done through a funciton call

var cX = new CanvasXpress(targetId, data, config);
cX.pivotX("Gender");