Maximizing Direct Marketing Campaign Effectiveness

Utilizing Logistic Regression approaches with 2nd degree interactions between variables, I predict the probability a customer will buy in an upcoming marketing campaign. While the original dataset only contains 11% "subscribed" outcomes, we observe almost 50% of records in the top 500 likely...

Mining Spatial & Text Data for Adoptable Dogs Dataset

By integrating county-level location data with a dataset of adoptable dogs, I not only cleaned web-scraped data for adoptable dogs, but visualized their location at the nation-wide and state level for the US.

Time Series Forecasting w/ Self-Optimizing ARIMA Models

Utilizing WDI (World Development Index) data, I go step-by-step through the creation of ARIMA models for time-series forecasting, and utilize machine learning to create an optimized model using the BIC & AIC. This optimally fits an ARIMA model to the data for GDP growth in France, the UK, and India.

Predicting Data Scientist Salary

This model utilizes ML to predict Data Scientist Salary within 5k of actual salary based on Skills, Location, Company Name, Company Rating (Glassdoor) and more. See dataset description and information at this link: https://www.kaggle.com/nikhilbhathi/data-scientist-salary-us-glassdoor

How do Holiday's Impact COVID-19 Cases?

By leveraging a variety of time series techniques and frameworks, I develop time series models that fit and predict the trend of COVID case numbers by leveraging Facebook Prophet & ARIMA models, generating time-series cross-validation performance metrics and an analysis of the impact of holidays.

Scientific Paper Recommendation Engine with Content Similarity & Document Clustering

Utilizing TF-IDF, Doc2Vec, PCA, & K-Means, I develop a series of content-based recommendations for scientific articles from the openly published Arxiv dataset, which includes articles for various STEM fields. I apply t-SNE for cluster visualization and cosine similarity/linear kernel for similarity.