DEV Community

Cover image for No-Code Data Exploration: Use Data Painter to Explore Your Data Visually
The Open Coder
The Open Coder

Posted on • Updated on

No-Code Data Exploration: Use Data Painter to Explore Your Data Visually

Data Painter is a creative yet powerful feature of RATH. You can easily clean data, model data, and explore data using a Painting Tool, which turns the complex Exploratory Data Analysis process visual and simple.

Data painter is especially good at explaining complex visual patterns which can be difficult to describe by simple statistics indicators.

Prerequisites

The following data visualization from the "College" dataset shows the relationship between AverageFacultySalary and AverageCost. We want to study why the data is not evenly distributed, but rather split into two parts.

Edit with Data Painter

Click on the Data Painter button to explore this chart.

Explore data

Case 1.

On the Data Painter tab, we can paint the dots which represent the data we want to explore with your mouse cursor.

Click on the Search button to automatically analyze your data painting. RATH can automatically detect your intent and recommends a data visualization for a possible explanation.

Explore data with Data Painter

The auto-generated result explains why the data is distributed unevenly: it was caused by the FundingModel variable. The blue dots on top of the chart are private schools, while the red dots on the lower side of the chart are public schools.

Data Exploration Result with Data Painter

For more suggested results, you can click on the "Next" and "Last" buttons to navigate through all the generated results.

Case 2.

In the following case, we are about to explore an interesting pattern that is more complex.

Step 1. We are about to analyze a chart with a complicated pattern. The chart demonstrates the relationship between SOC and the Maximum battery voltage.
Data Exploration - patterns about battery SOC

Step 2. We can use the Data Painter to study the unusual pattern in the chart. Click on the Data Painter button under the chart, and color the pattern by dragging your mouse cursor.

Click on the Search button. RATH can automatically detect your intention. Here, RATH suggests the deciding factor is the Vehicle Condition. The pattern represents the vehicle engine is in a flameout condition.

Explore SOC patterns with Data Painter

Erase unwanted data

You can use Data Painter to remove unwanted parts of data, or remove anomalies.

To erase data in Data Painter, click on the Erase button, and wipe out the unwanted data.

Erase data with Data Painter

After removing the unwanted parts of the data, you can focus on the data that you want to anslysis.

Erase data with Data Painter

Explore patterns in data

You might spot some patterns in the data. However, there are no available variables or features that could accurately describe the patterns in the original data.

In this case, you can create new variables and features with Data Painter, by painting the data in different colors. Each color represents a different subset or group.

Explore patterns in data

You can also visualize the patterns in your data by building customized charts. Click on the Explore button and use a Tableau-like User Interface to build customized charts by dragging and dropping.

Build customized charts in Data Painter

Like my post?👾

I am also working on an Awesome Open Source project named: RATH.
Check it out on GitHub!

Top comments (0)