Open Source Projects

The objective of this study is to apply several statistical learning methods to carefully analyze and predict home prices from home demographics. The data I used for this project is from the Kaggle data science competition in which I participated. After the competition, I continued making multiple changes, trying different algorithms, and eventually improving my machine learning modeling techniques and prediction power. The project and the detailed report are available on my GitHub. Please use Name the Price: Analysis and Prediction of House Prices with Machine Learning to access the project.

The primary goal of this project is to perform Natural Language Processing and Sentiment Analysis on the most discussed topic in the current economy. In this project, I extracted and used original Twitter data to quantify and visualize the general sentiment. In particular, sentiment regarding recession is analyzed and discussed. I employed several natural language processing techniques on 10K tweets to investigate and visualize the current sentiment of recession in the economy. The data are gathered from Twitter using API, advanced techniques are used to clean the data and study the sentiment. The reproducible code can be used by anyone with minor changes. The project and the detailed report are available on my GitHub at Natural Language Processing and Sentiment Analysis.

This project aims to analyze several key socioeconomic indicators related to poverty, education, unemployment rate, and population distribution across the United States. I used publicly available data from ERS to clean and visualize the socioeconomic indicators in the United States. Using several data cleaning techniques and interactive data visualizations, a shiny app was built and deployed on the web for use. The app is hosted on the shiny server and is available at Socioeconomic Indicators - United States - Shiny App. The project code is available on my GitHub at Socioeconomic Indicators - United States - Github Source.

In this project, I showcase commodity production in the United States over the years. I use publicly available data to develop an interactive data visualization Shiny application. Anyone can use this app by simply navigating through it. The app can be used to study how the production of various commodity crops in the United States has changed over time. It can also be used to view commodity production by the state as well as major producing states. The app is hosted on the shiny server and is available at Commodity Production - United States - Shiny App. The project code is available on my GitHub at Commodity Production - United States - Github Source.