Visualize-Data-with-QuickSightREADME (1).md
Project Link: View Project
Author: Nima Weatherly
Email: nima.weatherly@gmail.com
In this project, I will demonstrate how to use Amazon QuickSight to analyze Netflix data and generate visualizations and insights. I'm doing this project to learn how to use data services. I am doing this project to learn how to use cloud data services for data analysis.
Services I used were AWS S3 and Quicksight. This project demonstrates key concepts in cloud integration, data preparation, and visualization. Using S3 and QuickSight, I connected and refreshed a dataset, created calculated fields, and built visualizations showing both counts and percentages of movies vs TV shows by release year. By organizing visuals into a dashboard and documenting the process, I combined technical execution with clear data storytelling.
This project took me approximately 45 minutes to complete. The most challenging part was creating the visualizations. It was most rewarding to actually get to see the visualizations demonstrating the mariad ways data can be presented.
After this project, I plan to work on a CI/CD that will connect Visual Studio Code to GitHub and AWS. I will start this project on 27 Sept.
S3 is used in this project to store two files, which are manifest.json and netflix_titles.csv.
I edited the manifest.json file by adding the actual path to my netflix_titles.csv file. Manifest tells the format of the data and netflix_titles.csv file is the raw data. It’s important to edit this file because unless it matches the actual path Quicksight will not be able to find the data set file.
Creating a QuickSight account cost $0 as long as you make sure to uncheck the offer at the end of the setup process. The account comes with 30 days of free access.
Creating an account took me about 3 minutes.
I connected the S3 bucket to QuickSight by visiting Datasets page. I then selecte S3.
The manifest.json file was important in this step because it tells Quicksight how to understand the data and show it in charts or graphs.
To create a visualization in Amazon QuickSight, first connect to your dataset (e.g., from S3, RDS, or uploaded files). In an Analysis, drag fields into the field wells: assign dimensions to axes and measures to values. Choose a chart type (bar, line, pie, etc.) from the Visual types panel. You can apply filters, groups, or calculated fields to refine insights. Customize colors, labels, and formatting to improve readability. Multiple visuals can be combined into a dashboard, which you can then share or publish for others to explore interactively.
The chart/graph shown here is a breakdown of records in the netflix file by release year.
I created this graph by dragging and dropping release year to the visuals field into the small multiples section.
Filters are useful for centering data to answer a specific question and presenting the data in a form that helps to understand what is being presented.
This visualization is a breakdown of Titles by year divided into categories and presented as one bar with color distinction.
As a finishing touch, I added descriptive titles to each visualization and an overall title for the group of visualizations.
Did you know you could export your dashboard as PDFs too? In QuickSight, open the dashboard, click the Share dropdown, and select Export to PDF. You can export the current view or all sheets, and include filters or parameters applied. QuickSight then generates a downloadable PDF file of the dashboard for sharing or documentation.
In this project's extension, I downloaded fresh data that's different from my original dataset because the data was incomplete. Analysing incomplete data brings the risk of a misunderstanding of what the story the data is telling.
Once I downloaded new data, I had to update my S3 bucket because I needed to give S3 access to the new file. I also uploaded a new manifest.json file that tells Quicksight how the data is organized and which file the associated date is housed within.
I initally couldn't see my updated data in QuickSight, so I had to refresh the page, go to data, datasets, click the 3 dots and select edit. Then on the left lower half of the screen I clicked Refresh now. On the next page I selected Refresh, OK, OK and then got a banner saying that the update was complete. I then selected publish & visualize. When I went back to the original visualization I saw that it had been updated with the countries that had previously been missing now filled in.
