Intact Data Science Challenge by DOT JDSN

Inspiration

As a beginner in data science, we were excited to learn new skills and gain practical experience in a real-world scenario by participating in the Intact Data Science Challenge.

What it does

Our project aimed to classify a medical transcription as a medical specialty using machine learning algorithms.

How we built it

We uploaded the provided notebook to Google Colab and wrote all of our code there. We spent several days reading through the code and understanding the problem at hand. We also did some exploratory data analysis to gain insights into the dataset and visualize the distribution of medical specialties.

Challenges we ran into

The biggest challenge we faced was starting from almost 0 experience in data science. None of the software used in the challenge was familiar to us, but we were determined to learn and understand the tools provided. We spent time researching and learning the basics of Python, and eventually felt confident enough to start writing our code. Another challenge we faced was understanding the code provided by the challenge, but we overcame it through research and perseverance.

Accomplishments that we're proud of

We are proud of what we were able to accomplish with our limited knowledge of this subject matter. Our code had an accuracy of 49%, which was higher than the initial run accuracy rate of 4%.

What we learned

Through this project, we learned new skills in data science, including Python, Pandas, and Scikit-Learn. We also gained practical experience in a real-world scenario by participating in the Intact Data Science Challenge.