5004CMD Big Data Analysis of Mobility Patterns: A Comparative Study of Sequential and Parallel Processing

Overview

This project analyses mobility data from the Bureau of Transportation Statistics (BTS) to understand travel behaviour in the United States. It focuses on how many people stay at home, how far people travel, and how travel patterns change over time.

The project also compares sequential and parallel processing to evaluate performance differences.

Datasets

Trips_by_Distance.csv
Trips_Full Data.csv

These datasets contain information on travel frequency, distance ranges, and population movement.

Tasks Covered

Weekly analysis of people staying at home
Distance-based travel analysis
Comparison of high-volume trips (10–25 vs 50–100 miles)
Linear regression model to predict travel frequency
Visualisation of travel patterns
Sequential vs parallel processing comparison

Technologies Used

Python
Pandas
NumPy
Matplotlib
Scikit-learn
Multiprocessing

How to Run

Install required libraries: pip install pandas numpy matplotlib scikit-learn
Run the script: python mobility_analysis_5004cmd.py

Output

Plots are generated for each question
Execution time comparison between sequential and parallel processing
Model evaluation results (RMSE and R²)

Notes

Analysis is based on National-level data
Parallel processing was tested with 10 and 20 workers
Results show that parallel processing is not always faster for smaller workloads

Author

Khushboo Zope Student ID: 15501082

5004CMD-2526JANMAY/CW_Big_Dta_KZ

About

Resources

Stars

Watchers

Forks

Releases

Languages