Willian Kazuo Solution

[williankazuo]

Pull Request Template

Description

Create an algorithm that will process a set of historical patient data and compute a score for each patient that (1 as the lowest, 10 as the highest) that represents the chance of a patient accepting the offer off the waitlist.

Checklist:

• I have performed a self-review of my own code
• I have tested my code to prove my fix is effective or that my feature works
• I have commented my code in areas where it's hard to make the code speak for itself
• My changes breaks no tests

Implementation

• Load json data once, parse the patients data, and get min and max for each property to use for normalization later.
• Calculate distance between patients location and facility using Haversine Formula, get max and min distances.
• Normalize data. The data has different scales of values, so we have to normalize and transform the scale to 0~1.
• Limit distance to 1000 kilometers, not scoring when patient is more than 1000 km away. This would still be a high value in real life, but we're just going to limit to 1000 for this demo.
• Scoring, apply the weight to normalized values.
• Created endpoint /api/v1/top-patients that takes latitude and longitude as query params, and returns an ordered list of 10 patients who will most likely accept the appointment offer.

Considerations

• Younger patients are more likely to accept.
• Penalize patients with many canceled offers.
• Boost patients closer to facility.
• Boost patients with lower response time.

How to run

• Clone
• Run npm install
• Run npm run dev
• If using port default, it would be 3000. Make a get request to localhost:3000/api/v1/top-patients?latitude=?&longitude=?

Tests

npm run test
npm run test:coverage