About Me
My background is in electrical engineering, and I have experience as a patent examiner in the vehicle and navigation data processing field. I left the patent office to spend the next couple years traveling and volunteering in Asia and Latin America. After taking a machine learning class on Coursera, I became fascinated with big data and the wide range of applications of data science tools. I moved to San Francisco and enrolled at Zipfian Academy, where I expanded my data science skills by working with data to solve a variety of real world problems.
Projects I'm working on
How to get loans funded on kiva.org
I analyzed microfinance loans on kiva.org to predict which loans will get funded and identify the key features of successful loans. I used python and postgres SQL to process the data and perform feature engineering, transformed the text with TF-IDF, modeled the data with a weighted random forest, and plotted the results.
Housing market model
I was part of the second place team at the 2015 San Francisco Hack for Change hackathon. We built an agent based model to demonstrate the effects of adding housing supply on housing prices and displacement of renters and presented our findings at the end of the hackathon. We are now working on adding more complexity to the model to make it more accurately reflect San Francisco's housing market. It will be used by San Francisco Bay Area Renters Foundation and SF In Progress to help demonstrate the need for more housing.
Data Science Skills
- Machine Learning: Linear and Logistic Regression, Gradient Descent, Decision Trees, Random Forest, Boosting, SVM, Naive Bayes, K-means, Clustering, PCA, NLP, Recommenders
- Python: Pandas, Numpy, Scikit-learn, StatsModels, NLTK, Flask, Matplotlib
- Databases: Postgres SQL, MongoDB
- Distributed Computing: Map Reduce, GraphLab, Spark (beginner)
- Statistics: A/B testing, Multi-Armed Bandit, Bayesian Inference, Hypothesis testing
Links
I love traveling! This is me at Neuschwanstein Castle in Bavaria, Germany in May 2015.