In this project, we aim at finding the optimal location for building solar farm in the state of California considering
annual solar radiance, topography (slope and elevation), land use restriction and minimum distance to the nearest city
(project submitted for data science boot camp by the Erdős institute here)
This project evaluates the performance of four pre-trained language models—BERT-base (110M), GPT2-XL (1.5B), Qwen2.5 (1.5B), and LLaMA-3.2 (1B)—on multi-class emotion classification using subsets of the GoEmotions dataset.
In this project, we predict crash severity for accidents occured in Baton Rouge, LA from 2016-2021 based on multiple factors, including road conditions, vehicle conditions, traffic conditions, and driver attributes.
This interactive dashboard provides in-depth analysis of global weather patterns, environmental factors, and geographical variations using the World Weather Dataset.
This project aims to conduct an Exploratory Data Analysis (EDA) using SQL on the publicly available Olist e-commerce dataset from Kaggle. The dataset provides comprehensive information about orders, customers, sellers, products, payments, and reviews within the context of a Brazilian e-commerce platform. The primary objective is to extract actionable insights related to customer behavior, sales performance, product popularity, and order dynamics.
This project explores factors influencing celebrity divorces using machine learning. We scraped and analyzed data from Wikipedia, identifying key predictors like marriage age, career diversity, and spouse profession.