Mapathon Project

Electric Vehicle Adoption Map | QGIS, Data Acquisition, GIS Analysis

As part of the IIT Bombay FOSSEE Mapathon, I created an insightful map that showcases India's efforts to promote electric vehicle adoption. This project integrated reliable data sources such as:

• Data from the Ministry of Heavy Industries: Acquired from official government releases and processed into structured Excel sheets for further analysis.
• State Boundaries from Bhuvan: Leveraged official boundary data to ensure geographical accuracy.

Using QGIS, I developed four maps that help users:

• Identify the best states for purchasing electric vehicles.
• Determine optimal inter-state travel routes based on charger density.
• Analyze the ongoing shift in the Indian automobile industry towards electric vehicles.

This project enhances decision-making in electric vehicle adoption by providing practical insights through comprehensive spatial data.

Freelance Project

Efficient Billing Management System for a Law Firm | Flask, SQLite, Python

I developed a robust Flask-based web application for a law firm, streamlining their billing processes and significantly reducing the time spent on managing financial records. Leveraging the Jinja template engine and Flask-SQL Alchemy, I connected the application to a local SQLite database, enabling seamless data management and retrieval.

Key features include:

• Bill Logging: Automatically generates Google documents for bill details and collects additional expense data.
• Database Viewing and Sorting: Allows users to view and sort billing records by bank name or bill generation time.
• Comprehensive Bill Generation: Compiles bills from selected banks, calculates grand totals, and enables easy download for records.
• User-Friendly Execution: Simplified application launch through .bat files, making it accessible and easy to use for non-technical staff.

This project empowered the law firm to efficiently manage and access their billing data, saving valuable time and resources.

Research Project

Landslide Susceptibility Mapping Algorithm | Google Earth Engine, Machine Learning, Python

I developed an advanced Landslide Susceptibility Mapping algorithm designed to minimize creator subjectivity, providing more accurate and reliable susceptible area identification. The project integrated multiple cutting-edge technologies and methodologies:

• Google Earth Engine: Utilized for efficient storage and management of extensive geospatial datasets.
• Hypothesis Testing: Applied during the data pre-processing phase to ensure statistical robustness.
• Machine Learning (Python): Implemented to analyze data and predict landslide susceptibility with high precision.
• Map Creation: Generated detailed susceptibility maps using Google Earth Engine, based on the refined data and machine learning results.

This research was recognized and presented at the National Conference of Machine Learning and Artificial Intelligence in October 2023, underscoring its significance and impact in the field.