-Visualizing_Time_Series_Dataset_COVID_19_Data_V2.

Ai/ml project

📊 Visualizing Time Series Dataset: COVID-19 Data V2

🧠 Overview

This project analyzes global COVID-19 and Ebola time series data to uncover trends in confirmed cases, recoveries, and fatalities. Using Python-based data visualization techniques, it transforms raw datasets into intuitive visual narratives that support public understanding and data-driven decision-making.

🎯 Objectives

• Visualize temporal progression of COVID-19 and Ebola across countries
• Compare regional impacts and healthcare responses
• Enable exploratory data analysis (EDA) through visual tools
• Extract actionable insights from pandemic data
• Lay groundwork for future forecasting and modeling

🛠️ Technologies Used

• Python
• Pandas – Data manipulation
• NumPy – Numerical operations
• Plotly – interactive operations
• Matplotlib & Seaborn – Static visualizations
• Colab Notebook – Development environment

📂 Dataset

• Format: CSV files with daily records of confirmed cases, deaths, and recoveries
• Scope: Global data from Jan 2020 to mid-2021

🔄 Workflow Summary

1. Data Collection – Importing global COVID-19 and Ebola datasets
2. Data Cleaning – Handling missing values, formatting dates
3. Preprocessing – Aggregating by country/date, feature engineering
4. Visualization – Line plots, bar charts, heatmaps
5. Insight Extraction – Annotating trends, comparing regions
6. (Optional) Modeling – ARIMA-based forecasting with train/test split

📈 Sample Visuals

• Daily confirmed cases by country
• Rolling averages of deaths and recoveries
• Heatmaps showing outbreak intensity over time

📌 Notes

• No primary survey was conducted; analysis is based on secondary data
• No formal hypothesis testing was performed, but visual trends support exploratory insights
• The project can be extended with interactive dashboards or predictive models

🧾 License

This project is open-source

.

📋 Survey Questionnaire: Public Perception of COVID-19 Data Visualizations Target Population: Urban and semi-urban residents aged 18–60 in India Sample Size: 50 respondents Sampling Method: Stratified rSeptember 2025 Mode: Offline

Q1. How frequently do you check COVID-19 statistics? ⬜ Daily

✅ Weekly

⬜ Occasionally

⬜ Rarely

Most common response: Weekly (62%) Interpretation: Indicates moderate engagement with pandemic data, suggesting visual summaries are more effective than raw daily updates.

Q2. Which source do you trust most for COVID-19 updates? ⬜ Social Media

✅ Government Health Portals

⬜ News Channels

⬜ International Health Organizations (WHO, CDC)

Most common response: Government Health Portals (48%) Interpretation: Localized and official sources are preferred, reinforcing the need for region-specific visualizations.

Q3. Do visual graphs help you understand COVID-19 trends better than tables or raw data? ✅ Yes

⬜ No

⬜ Not Sure

Most common response: Yes (84%) Interpretation: Strong support for visual storytelling—validates the core objective of your project.

Q4. Which type of visualization do you find most helpful? ✅ Line Graphs

⬜ Bar Charts

⬜ Heatmaps

⬜ Pie Charts

Most common response: Line Graphs (56%) Interpretation: Time series line graphs are preferred for tracking trends—aligns with your use of smoothed daily case plots.

Q5. Have you ever changed your behavior (e.g., travel, mask usage) based on visualized COVID-19 data? ✅ Yes

⬜ No

⬜ Not Sure

Most common response: Yes (68%) Interpretation: Visual data influences public behavior, emphasizing the importance of clarity and accessibility in pandemic communication.

🙋‍♀️ Author

Adrija Sil – B.Tech CSE, Government College of Engineering and Ceramic Technology
Passionate about data storytelling, competitive programming, and building tech for social impact.