Python
Experienced in Python for thorough exploratory data analysis (EDA) and meticulous data cleaning using Pandas. Proficient in harnessing Matplotlib and Seaborn for insightful visualization, while employing Scikit-Learn for advanced modeling techniques. Skilled in utilizing these libraries for in-depth regression analysis, trend identification, and extracting valuable insights from complex datasets.
Australian Weather Prediction - Machine Learning
Analyzed a vast dataset of 145,460 data points to predict Australian rainfall with an 85% accuracy. Employed diverse machine learning models like Logistic Regression and DecisionTree, showcasing a comprehensive approach. Used Python for efficient data preprocessing and model implementation, ensuring a robust predictive analysis.
Insurance Premium Prediction - Linear Regression
Analyzed a dataset of 1338 insurance clients, focusing on customer attributes and premiums, notably uncovering the significant impact of smoking on charges. Leveraged Python's Scikit-Learn, Pandas, and custom functions to build a precise linear regression model for premium predictions, reducing error rates. The findings highlighted smokers paying nearly double the insurance costs, crucial for pricing and risk assessment in the industry.
Spotify Analysis
Analyzed a large Spotify music dataset, leveraging Python, Pandas, Matplotlib, and Seaborn for preprocessing and visualization. Identified top artists and trends using regression line correlation maps, showcasing expertise in complex data analysis and extracting valuable insights from intricate datasets.