This ShinyApp is designed to be an unemployment rate comparison tool. It allows the user to view the unemployment rate for all states before and after covid-19. To view the app (best viewed on desktop) in action, please visit:
This project seeks to understand why a local restaurant has a one-star rating by using text and sentiment analysis.
Data exploration
An overview of ratings, which includes using a bell curve to visualize the distribution of ratings
Text Analysis
Categorize emotions (positive/negative), graph the most frequently used words, and create word clouds.
Sentiment Analysis
Tested customer sentiment using five packages: Syuzhet, Bing, AFINN, NRC, and Sentimentr.
Analysis that utilizes the OLS and Poisson regressions to determine whether more gun laws lead to more mass shootings
We test the Phillips Curve Theory using time series data to examine the short and long-run relationship between the inflation and unemployment rates.
Using the Pima Indian Diabetes data set from the Kaggle website, we built a supervised model that predicts whether a patient has diabetes with 83% accuracy. We used Logistic Regression, PLSDA Regression, SVM Radial, and Random Forest to accomplish this.