BIO

April 2019

By designing a clear database for automatically storing and easily retrieving the data, our work will help the company keep track of basic transactions, provide information on borrowers and investors, and generate insights through various analytical procedures that will help the company run the business more efficiently, help managers and employees make better decisions, mitigate investment risks, and at last improve the whole decision-making process.

The goal of this project is to examine whether President Trump’s tweets is a source of influence for the stocks market in the United States. The effect of tweets can be measured by coinciding changes in the stock market or changes in the share prices of certain companies targeted in the President’s Tweets. Of the roughly 34,000 tweets used in this research between include were examined using keywords sentiment analysis to find which the overall tweet sentiment. Once determined, we selected the tweets representing the strongest positive and negative emphasis to use as our designed intervention point. We will also use the Causal Impact Timeseries Algorithm developed by Google to validate the pre-conceived relationship between Trump’s twitter attacks/praises directed at a company and its resulting market performance in the stocks.

PREDICTING AIRBNB PRICE IN NEW YORK CITY (RANK TOP 2%)

December 2018

Pricing a rental property on Airbnb is a challenging task for the owner. On the other hand, customers have to evaluate an offered price with minimal knowledge of an optimal value for the property. This project aims to develop a price prediction model using a range of methods from linear regression to random forest, , XGBoost and support vector machine (SVM) to tackle this challenge.

Developed model to detect drivers’ dangerous behaviors with Convolutional Neural Network and deployed the model with AWS cloud technology