Data Scientist
(almost) Master's degree in Data Science with demonstrated ability to work in diverse interdisciplinary domains
As part of a kaggle machine learning competition, I was given tabular training data which consisted of sixty four columns of 1s and 0s representing the presence of symptoms ranging from back pain to nose bleeding. Instructions were asked to predict which of the 13 possible diseases was present for each of the 300 row in the test data. Iterations of my solution involved a deep neural network and a random forest classifier model which were trained on data which I cleaned and feature engineered. My best solution however was obtained using the open source framework developed by Yandex.
Another kaggle competition entry.I learned a lot and got my feet feet wet with some mixed tabular as well as data in string and float format. My focus was more on data analysis, preparation and feature engineering.