Eric Kim's Projects
Churn Modeling using Neural Networks and Hyperparameter Tuning (TensorFlow and Keras)
We turn our attention to classification. Classification tries to predict, which of a small set of classes, an observation belongs to. Mathematically, the aim is to find y , a label based on knowing a feature vector x . For instance, consider predicting gender from seeing a person's face, something we do fairly well as humans. To have a machine do this well, we would typically feed the machine a bunch of images of people which have been labelled "male" or "female" (the training set), and have it learn the gender of the person in the image from the labels and the features used to determine gender. Then, given a new photo, the trained algorithm returns us the gender of the person in the photo. There are different ways of making classifications. One idea is shown schematically in the image below, where we find a line that divides "things" of two different types in a 2-dimensional feature space. The classification show in the figure below is an example of a maximum-margin classifier where construct a decision boundary that is far as possible away from both classes of points. The fact that a line can be drawn to separate the two classes makes the problem linearly separable. Support Vector Machines (SVM) are an example of a maximum-margin classifier.
Examining Racial Discrimination in the US Job Market
In this project, I will develop a deep learning model to achieve a near state-of-the-art performance on the MNIST handwritten dataset. I'm going to use Keras with TensorFlow.
In October 2012, the US government's Center for Medicare and Medicaid Services (CMS) began reducing Medicare payments for Inpatient Prospective Payment System hospitals with excess readmissions. Excess readmissions are measured by a ratio, by dividing a hospital’s number of “predicted” 30-day readmissions for heart attack, heart failure, and pneumonia by the number that would be “expected,” based on an average hospital with similar patients. A ratio greater than 1 indicates excess readmissions.
Convolutional Neural Network is going to learn differences between a cat and a dog and make predictions.
In this example we'll classify an image with the bundled CaffeNet model (which is based on the network architecture of Krizhevsky et al. for ImageNet).
In this exercise, I get work with JSON strings and files using packages such as numpy, pandas, and json.
Gaming. Do we perform the best during the day or night? After work or before work? During work? This quick study uses Riot's API to find out.
Something relatively simple but fundamental to any business is customer segmentation. At the core of this, we are able to identify different types of customers so we can get more customers.
Linear regression is used to model and predict continuous outcomes while logistic regression is used to model binary outcomes. We'll see some examples of linear regression as well as Train-test splits.
In the mini-project, I'll go through the basics of text analysis using a subset of movie reviews from the rotten tomatoes database. I'll also use a fundamental technique in Bayesian inference, called Naive Bayes. This mini-project is based on Lab 10 of Harvard's CS109 class.
MapReduce using Spark
We build a recommendation engine with a data set which contains 1 million ratings collected from 6000 users on 4000 movies from MovieLens.
The goal of this study is to improve user experience by analyzing and predicting data. Improved user experience can be achieved through more matches. To get more matches, a better representation of data would be helpful. This could include self-reports. However, an even better method could be achieved through predictions using this data for machine learning.
Predict sales prices and practice feature engineering, RFs, and gradient boosting. Top 4% as of Aug 2017.
Predicting Traffic Signs using Neural Networks (Deep Learning with TensorFlow)
Repository for Programming Assignment 2 for R Programming on Coursera
Recurrent Neural Network to Predict Tesla Stock Prices (TensorFlow and Keras)
In this project, I will create a neural network model with Keras. I will also use scikit-learn to evaluate models using cross-validation. Finally, I will tune the network topology of models with Keras. My backend will be TensorFlow.
A recreation of hurtpiggypig's Simply Love SM3.95 theme made to now run in StepMania 5
Advanced rhythm game for Windows, Linux and OS X. Designed for both home and arcade use.
Time Series Analysis
What is the True Normal Human Body Temperature?
A ride-sharing data science challenge with exploratory data analysis, experiment/metrics design, and predictive modeling.
XML exercise with mondial database