Vibhav Misra

Developed custom Python modules to compute Eigenvalues, PCA, and SVD, and compared their runtime and accuracy against NumPy and Scikit-learn implementations.

Performed time series analysis on Amazon reviews (2002–2014) and forecasted ratings using ARIMA, Prophet, and LSTM models, with LSTM delivering the best predictive performance.

Performed detailed EDA on telecom customer churn data using Numpy and Pandas to identify key factors influencing churn, uncover data patterns, and support churn prediction strategies.

Analyzed electric vehicle population data from Washington State to explore trends in EV adoption, electric range variations, and manufacturer distribution.

Performed exploratory data analysis, model training, hyperparameter tuning (GridSearchCV, RandomizedSearchCV), and performance evaluation using accuracy, MSE, R², and confusion matrices.

Designed a responsive web platform that uses ML models to predict diseases from user input. Implemented the backend with Flask and integrated model APIs using Pickle and Pandas.

Developed a machine learning-based prediction model for identifying depression in individuals. Focused on model implementation and analyzing the most influential features.

Conducted EDA to analyze feature distributions and correlations. Built and benchmarked multiple classification models to assess accuracy, precision, and recall in predicting depression.

Built a Flask-based web app to detect phishing websites using a Gradient Boosting Classifier. Engineered 30+ URL-based features for accurate prediction and developed a real-time user interface.

Trained a Logistic Regression model to predict the likelihood of forest fires based on environmental features. Used NumPy, Pandas, and scikit-learn for preprocessing and model training, and deployed the model using Pickle.

Developed a machine learning model to predict Falcon 9 first stage landing success. This end-to-end project included web scraping, data cleaning, EDA, feature engineering, and classification with models like Logistic Regression, Decision Trees, and SVMs.