Irlan Dos Santos

Posted on Dec 19

Visualize Data with AWS QuickSight

Overview

Amazon QuickSight empowers users to analyze datasets and create insightful visualizations with ease. This guide walks you through the process of analyzing Netflix's extensive dataset of shows and movies to create a dynamic dashboard brimming with actionable insights.

Project Objectives

Here’s the roadmap to transform raw data into a polished dashboard:

Upload the Dataset: Store data in an Amazon S3 bucket.
Create a QuickSight Account: Set up and access Amazon QuickSight.
Connect Data: Link the S3-hosted dataset with QuickSight.
Visualize Data: Build graphs, charts, and perform analyses.
Publish Dashboard: Share your findings via a professionally styled dashboard.

Instructions

Dataset Preparation

Download Required Files:
- netflix_titles.csv: Contains information about Netflix's titles.
- manifest.json: A configuration file to guide QuickSight in reading the dataset.

Store Dataset in Amazon S3

Log in to your AWS Management Console.
Navigate to the S3 service and create a new bucket:
- Name the bucket (e.g., quicksight-project-dataset).
- Choose a region near you for optimal performance.
Upload netflix_titles.csv.
Copy the S3 URL of the uploaded dataset for later use.

Open the manifest.json file and replace the placeholder URL with the Amazon S3 URL of your dataset.

Set Up Amazon QuickSight

Search for Amazon QuickSight in your AWS console.

Sign up for the Enterprise Free Trial (ensure optional features like paginated reports are unchecked to avoid charges).

Enter your details for your QuickSight account. Make sure the email you use matches your AWS account email to avoid errors.

Select Select S3 buckets and look for the bucket you created earlier.

Make sure you UNCHECK ANY OPTIONAL ADD-ON options.

Stop and double-check:
Did you uncheck "ANY OPTIONAL ADD-ON options" checkbox? Make sure to do this so your AWS account doesn't get charged!

Click Finish.

Connect S3 Bucket to QuickSight

Navigate to the Datasets section in QuickSight and select New Dataset.

Choose S3 as the data source and enter a name (e.g., Netflix_Data).
Copy the S3 URL of manifest.json, paste it into the required field, and connect.

You will see this screen; click Visualize.

On this screen, click X.

Any Errors?

No stress! Errors can happen, especially when you're connecting your dataset with QuickSight. Here are the top errors and solutions:

Common Errors and Fixes

QuickSight doesn't have permissions to access the database
- Cause: QuickSight might not be connected to your S3 bucket.
- Solution:
  1. Select the profile icon on the top-right of the QuickSight dashboard.
  2. Go to Manage QuickSight and then Security & permissions.
  3. Under QuickSight access to AWS services, select Manage and add your S3 bucket.
Manifest file cannot be parsed as valid JSON
- Cause: The manifest.json file is incorrectly formatted or doesn't match the S3 region.
- Solution:
  1. Ensure your QuickSight account is in the same region as your S3 bucket.
  2. Check your S3 bucket's region in the AWS Management Console.
  3. Compare it with the region displayed in your QuickSight dashboard (profile icon > region details).
  4. Verify that manifest.json is formatted correctly:
  5. Ensure no brackets are missing.
  6. Confirm the S3 bucket URI in the file is accurate.
Unable to connect to the S3 bucket
- Cause: The S3 bucket may not be associated with your AWS account, or your QuickSight email may not match.
- Solution:
  - Ensure your QuickSight email matches your AWS account email.
  - If they differ, delete your QuickSight account and recreate it using the same email as your AWS account.
Not enough SPICE capacity
- Cause: SPICE (QuickSight's in-memory engine) capacity is region-specific.
- Solution:
  - Use the same region where you created your QuickSight account.
  - Free SPICE capacity is only available in the region where QuickSight was initially set up.

Tips

Double-check JSON formatting before uploading.
Always ensure QuickSight and S3 regions match.
Regularly verify permissions for AWS services in Manage QuickSight.

By addressing these common issues, you'll be well-prepared to connect your dataset successfully!

Create Your Visualizations

Some Explanation about the Fields:

Data: The head of a column, where information is cataloged.
Visuals: The type of graphs you can choose.
Add Data, Y-Axis Value, Group/Color: Attributes you can edit to adjust how data appears in graphs.

Explore the dataset and view available fields. -Just play combining the DATA groups, and change the type of graphs:

Click on release_year to the Y-axis to visualize title releases over time.
Experiment with chart types like bar charts, pie charts, and donut charts.
Group data by categories like type (TV shows vs. movies) for deeper insights.
Resize and rearrange charts for clarity and aesthetics.

Advanced Analysis

Create a horizontal 100% stacked bar chart to show proportions of TV shows vs. movies by release year.
Present the number of titles per year in a detailed table.
Identify the date with the highest number of new titles (e.g., January 1, 2020).
Filter for categories like Action & Adventure, TV Comedies, and Thrillers.
Filter to show titles released in or after 2015 for selected genres.

Polishing Your Dashboard

Add clear and descriptive titles to all charts:
- "Titles by Release Year"
- "Proportion of Movies vs. TV Shows"
- "Genre Distribution Over Time"
Publish the dashboard and export it as a PDF for sharing.

Wrap-Up

Congratulations! You’ve created a comprehensive QuickSight dashboard. Here’s a summary of what you achieved:

Uploaded data to Amazon S3.
Connected S3 to QuickSight.
Built visualizations that uncover meaningful trends.
Published and exported a polished dashboard.

Don’t forget to clean up your AWS resources to avoid unnecessary charges:

Delete your QuickSight account. -On the dashboard, go to top right corner, click on the user icon, and then Manage QuickSight.

Select Account Settings from the left hand navigation panel, and then Manage at the bottom of the page.

Empty and delete your S3 bucket. -If you prefer, you can use the cloudshell and run the command:


bash
aws s3 rb s3://YourBucketName --force

Good Job!!!

DEV Community