We decided to try to visualize all of the cool sample data Vitech provided us with in a nice format that is easy for marketing managers to make use of.
What it does
V3 Dashboard provides a visualization of Vitech's Insurance data in a way that allows for marketing departments to more specifically tailor their advertising to certain demographics. The dashboard itself is a web-based application that queries the Vitech API and displays data related to Vitech marketing efforts. Each marketing activity is broken down by success rate among gender, location, age and marital status.
In addition to this data about the success of each marketing effort, we also utilized the machine learning capabilities scikit learn to come up with the type of person that each campaign works best on. We used Scikit learn to compile a list of known people who would make the best targets for each marketing campaign but have not yet received any advertising materials and allow marketing managers to download this list as a CSV.
How we built it
The frontend is built with React and Chart.js. The machine learning capabilities are provided by Scikit learn.
Challenges we ran into
Vitech's API provided us with a lot of data (~1.5 million people, ~ 1.5 million active insurance policies) and processing all of this was slow and cumbersome at times. Additionally, on the frontend, a lot of time went into optimizing into how much memory is being used by the browser due to the large amounts of data provided.
What we learned
We learned a lot about React, Chart.js, Scikit learn and fidget spinners.