Driven by Data

Inspiration

We wanted to have a chance to handle large amounts of data as well as participate in a our first datathon.

What it does

We made 3 models that predict whether or not a user takes their first uber drive with a 92%-93% accuracy.

How we built it

We used a multitude of models to successfully take in data and use that data to predict if the uber user would go on their first ride. We also spent a good amount of time cleaning and making sure that the data is only the data that was relevant to us.

Challenges we ran into

Learning how to manage a large set of data, and cleaning and balancing the dataset so that the models are trained accurately and reliability.

Accomplishments that we're proud of

Creating a model with 3 working models with 90+ accuracy in a first hackathon. Cleaning and training our model using big data.

What we learned

We learned sci-kit-learn, pandas, SQL, interpreting data, and establishing a correlation between real-life data vs predictive data.

What's next for Driven by Data

We are looking to participate in future hackathons to keep expanding our knowledge around coding. We also see ourselves coming back here next year as it was a fun and knowledgable endeavor.

Built With

• matplot
• numpy
• pandas
• plotly
• python
• scikit-learn
• seaborn
• xgboost