TD Hospital Exploration

What it does

Our program takes in a dataframe, cleans the given data, finds the Principal Component Analysis (PCA) which informed us of which factors of a patient to include in our machine learning model, uses logistic regression, random forest and neural network machine learning models to predict survivability, and combines our differing approaches through a multi modal machine learning model to formulate a more accurate prediction of a patient's survivability.

How we built it

We utilized python, numpy, tensorflow, scikit-learn, and matplotlib to clean and organize data, view different approaches to finding a prediction of a patient's survivability, and combine approaches through a multi modal model of machine learning.

Challenges we ran into

Initially we did not account for the need to keep a machine learning model for each run of our program, rather than re-training for each iteration of testing. We resolved this issue by reorganizing our files, adding new functions to help a clean transition from our results to the required submission results, and refining our submission output.

Accomplishments that we're proud of

In general, we are extremely proud of being able to create an entire machine learning model, as we had no prior experience with machine learning, and for it to be able to accurately predict a patient's survivability in a professional context.

What we learned

We learned the basics of machine learning as well as multi modal models in order to accurately predict results based off of a given data set.

What's next for TD Hospital Exploration

We plan to revise our calculations and refine our submission results for an even more accurate result.

Built With

• anaconda
• matplotlib
• numpy
• python
• scikit-learn
• tensorflow