I’ve been working my way through the TensorFlow in Practice Specialization on Coursera. I’m learning how to use neural networks to solve problems like image recognition. I decided to take a break from the course and try applying what I’ve learned so far to one of the Kaggle competitions. The MNIST is a database of more than 50,000 handwritten numbers. The goal, usually, is to train a model that can be used for digit recognition.
In a previous blog ("Modeling the UCI Heart Disease dataset") I trained a model to predict the presence of heart disease. So I have a model, now what? Machine learning models like this can be put to work generating predictions on new inputs, and they’re great for simulations as well. Let’s say we wanted to know the likelihood of heart disease for a 60 year-old male with a cholesterol value of 244, and a resting blood pressure value of 88.
Does the growth in COVID-19 cases have anything to do with Big 5 Personality traits? To answer this question, I compute country-level aggregates on the Big 5 test, and a country-level aggregate that represents for “growth” over time in coronavirus cases, using data current as of March 20, 2020.
Using logistic regression, I trained a machine learning model to predict heart disease, using 14 attributes and 303 observations (e.g., age, sex, chest pain, resting ECG).
A simple prediction of coronavirus spread in the US, using case trends in China and Italy.
There are times when you must go outside, even during the coronavirus pandemic. What if you had a tool that could tell you if a place, like the grocery store, was off-peak and safe to visit, or off-peak and not safe to visit?
Some observations about high-cardinality features and sparsity.
We’re starting to use SMS at work for communicating with customers, and there was a need for a tool that would allow us to send SMS messages and check the history of calls and messages to the phone number we are sending from. With that in mind, I coded a simple frontend to the Twilio API. The frontend is a page with 4 navigation tabs that allow sending a single SMS, sending SMS in bulk, seeing the status of sent SMS messages, and fetching the inbound call/SMS log.
Continuing to practice my python skills. I decided to try modeling the Telco Customer Churn dataset from Kaggle. Churn is when customers end their relationship with a company (e.g., by cancelling their subscription to a service). Companies want to retain customers, so understanding and preventing churn is naturally an important goal.