Reflective Report: COVID-19

Data Dashboard
Abstract

This reflective report documents the creation and execution of an interactive data visualization
dashboard aimed at analyzing the global impact of COVID-19. Utilizing the Kaggle dataset from Our
World in Data, the project applied advanced visualization techniques to represent metrics such as new
cases, total cases, deaths, and vaccinations. Developed using Streamlit and Plotly Express, the
dashboard offers users dynamic and visually appealing tools to explore key pandemic trends across
continents and countries. This report discusses the systematic approach to dataset preprocessing,
visualization design, and interactivity enhancements, while also addressing technical challenges
encountered and overcome. It further reflects on the significant learning outcomes, emphasizing the
critical role of storytelling through data. Ultimately, this project highlights the power of modern data
visualization in enabling accessible and insightful analysis of large datasets, contributing to informed
decision-making during a global health crisis.
Acknowledgments

I would like to thank my group members for their invaluable contributions to this project. Their
collaboration and commitment ensured the successful development of the dashboard. Additionally, I
appreciate the guidance provided by the module leader, Nouh Elmitwally, whose insights helped
shape the project’s focus and execution.
Contents
Abstract.................................................................................................................................................2
Acknowledgments.................................................................................................................................3
Executive Summary...............................................................................................................................5
1. Introduction.......................................................................................................................................6
2. Project Process..................................................................................................................................6
2.1 Dataset Overview.........................................................................................................................6
2.2 Visualizations Implemented.........................................................................................................7
3. Challenges Faced...............................................................................................................................7
3.1 Data Complexity and Cleaning.....................................................................................................7
3.2 Optimization................................................................................................................................8
3.3 Visualization Choice.....................................................................................................................8
4. Key Learnings.....................................................................................................................................8
4.1 Technical Skills.............................................................................................................................8
4.2 Data Analysis................................................................................................................................8
4.3 Collaboration and Problem-Solving.............................................................................................8
4.4 Importance of Visual Storytelling.................................................................................................8
5. Improvements for Future Work.........................................................................................................8
6. Conclusion.........................................................................................................................................9
7. References.......................................................................................................................................11
Executive Summary

This project centered on developing an interactive COVID-19 data dashboard to present insights into
the pandemic using a comprehensive dataset sourced from Kaggle. The dashboard integrated multiple
visualizations to explore various metrics, including new cases, total cases, total deaths, and
vaccination rates, providing users with an engaging and informative experience. Core features
included line charts for temporal trends, bar charts for comparative analysis, and a choropleth
heatmap for geographic visualization.

A robust methodology was followed, beginning with data cleaning and aggregation to ensure the
dataset's usability. The project employed Streamlit for user interactivity and Plotly Express for
dynamic visualizations, enabling users to interact with filters and explore data by continents and
countries. Challenges such as handling incomplete data, optimizing performance, and designing
effective visualizations were tackled collaboratively within the group, leveraging technical solutions
and iterative design improvements.

Key takeaways from the project included enhanced technical skills in Python and data visualization
tools, improved teamwork and problem-solving abilities, and a deeper appreciation of the importance
of clear visual storytelling in data analysis. Recommendations for future work include incorporating
additional metrics like testing rates, further enhancing interactivity, and implementing advanced data
management techniques for scalability. Overall, the project succeeded in creating a valuable tool for
analyzing the global effects of COVID-19 while fostering significant academic and practical learning.
1. Introduction

The COVID-19 pandemic, unprecedented in scale and impact, has underscored the importance of
accessible and actionable data visualization for informed decision-making. This project aimed to
design and develop an interactive dashboard that provides meaningful insights into the pandemic
using a dataset sourced from Kaggle. The data, compiled by Our World in Data, encompasses
critical metrics such as daily and cumulative cases, deaths, and vaccination rates, segmented by
countries and continents.

The objective of the dashboard was twofold:

1. To leverage modern visualization techniques to analyze complex and voluminous data.

2. To create an intuitive platform that caters to both technical and non-technical audiences,
enabling them to explore key trends and disparities in COVID-19 metrics.

The dashboard was built using Streamlit, a Python-based framework known for its simplicity and
interactivity, and Plotly Express, a powerful library for creating dynamic and visually appealing
charts. Visualizations such as line charts, bar graphs, and a choropleth heatmap were employed to
represent temporal, geographical, and comparative data effectively. Interactivity was a key feature,
allowing users to filter data by regions and focus on specific metrics.

This report delves into the systematic approach undertaken for dataset preprocessing, visualization
design, and implementation of interactivity. It also reflects on the challenges faced during the project,
such as managing missing data and optimizing performance, and highlights the key learnings gained
from this experience. By successfully combining technical precision with user-centric design, the
project demonstrated the transformative power of data visualization in addressing complex global
phenomena like the COVID-19 pandemic.

2. Project Process

2.1 Dataset Overview

The Kaggle dataset contained extensive COVID-19 metrics, including:

 Daily new cases and deaths.

 Cumulative totals for cases, deaths, and vaccinations.

 Geographic data (e.g., continents, countries, and ISO codes).

 Temporal data spanning the pandemic timeline.

Data cleaning involved:

  Removing rows with missing critical metrics like new cases and deaths.

 Aggregating data for continent-level visualizations.

 Filtering data dynamically for heatmap visualization.

2.2 Visualizations Implemented

1. New COVID-19 Cases Over Time

o Purpose: To visualize the pandemic's progression on a daily basis.

o Implementation: A line chart mapped daily new cases against dates. Peaks in the
graph correlated with major pandemic waves.

2. Total Cases by Continent

o Purpose: To compare the pandemic’s impact across continents.

o Implementation: Aggregated total cases by continent and visualized them through a

bar chart.

3. Heatmap Visualization

o Purpose: To display geographic distribution of total cases interactively.

o Implementation: A choropleth map with a dropdown menu for continent filtering.

The iso_code and total_cases columns enabled location-based coloring.

4. Total Deaths Over Time

o Purpose: To analyze the mortality trend during the pandemic.

o Implementation: A line chart showing cumulative deaths, highlighting significant

mortality peaks during major outbreaks.

5. Total Vaccinations by Country

o Purpose: To showcase global vaccination progress.

o Implementation: A bar chart displaying maximum vaccination counts per country.

3. Challenges Faced

3.1 Data Complexity and Cleaning

The dataset contained incomplete and inconsistent records for certain metrics, such as missing
vaccination data for some countries. Resolving this required filtering out problematic rows and
interpolating missing values where feasible.

3.2 Optimization

Rendering interactive visualizations from large datasets posed performance challenges. For example,
the heatmap required substantial computation for global rendering. Streamlit's caching mechanism
and data aggregation helped optimize performance.

3.3 Visualization Choice

Selecting the best visualization type for different metrics was non-trivial. Multiple iterations were
required to ensure clarity, accessibility, and interpretability, particularly for users without technical
expertise.

4. Key Learnings

4.1 Technical Skills

 Gained hands-on experience with Streamlit for developing interactive dashboards.

 Enhanced skills in using Plotly Express for dynamic and visually appealing data
visualizations.

4.2 Data Analysis

 Developed proficiency in cleaning, transforming, and aggregating large datasets to derive

meaningful insights.

4.3 Collaboration and Problem-Solving

 Strengthened teamwork and communication skills while collaborating on task assignments

and reviewing visualizations iteratively.

4.4 Importance of Visual Storytelling

 Learned how to effectively present complex data through visual mediums that are accessible
to both technical and non-technical audiences.

5. Improvements for Future Work

1. Enhanced Interactivity:
Incorporating time sliders to dynamically adjust date ranges and additional filters for granular
exploration.
 2. Expanded Metrics:
Adding metrics such as testing rates and healthcare system capacity for a more
comprehensive view of the pandemic.

3. Performance Improvements:
Utilizing advanced data storage solutions like databases (e.g., SQLite) to enhance scalability
and responsiveness.

6. Conclusion

The COVID-19 Data Dashboard project successfully fulfilled its objective of providing a
comprehensive and interactive platform for exploring pandemic data. Through the integration of
modern tools like Streamlit and Plotly Express, the dashboard enabled users to analyze critical
metrics such as new cases, total cases, deaths, and vaccination rates across different regions and time
periods. The project showcased the value of data visualization in making complex datasets accessible
and actionable.

This project not only demonstrated the technical feasibility of creating dynamic dashboards but also
emphasized the importance of visual storytelling in data communication. The interactive features,
such as dropdown filters and continent-based heatmaps, added depth and functionality to the analysis,
making it a valuable resource for stakeholders ranging from policymakers to general users.

Beyond the technical achievements, the project provided significant learning opportunities. It
enhanced our proficiency in Python-based visualization tools, strengthened our ability to preprocess
and analyze large datasets, and fostered a collaborative approach to problem-solving. Challenges,
such as managing data inconsistencies and optimizing performance for interactivity, were overcome
through iterative design and technical innovation.

Looking ahead, there are several avenues for improving the dashboard. Incorporating additional
metrics, such as healthcare capacity or recovery rates, would provide a more holistic view of the
pandemic. Advanced interactivity, such as time sliders or drill-down features, could further enrich
user experience. Finally, integrating the dashboard with real-time data sources could transform it into
a live monitoring tool.

In conclusion, this project not only achieved its goals but also reinforced the critical role of data
visualization in understanding global phenomena. It demonstrated the power of combining
technology, creativity, and teamwork to deliver meaningful insights, paving the way for more
impactful applications in the future.
7. References

 Kaggle: Our World in Data - COVID-19 Dataset

 Plotly Documentation: https://plotly.com/

 Streamlit Documentation: https://streamlit.io/