Grep Health

Don't waste time looking up your symptoms online! Use Grep Health to speed up the process.

Comment

Inspiration

As tech-reliant people such as us, we've always found it a no-brainer to look up our symptoms online when we felt ill. There were a few issues with this process however:

  • There are a lot of sites that might say different things
  • Much of the information provided was irrelevant or bloat

So we set out to fix that problem.

What it does

Grep Health will scrape the most common health sites to extract information about a given symptom, digest all that data, then present it to the user in a short and simple format. It will also link to the same sites it scanned so you can learn more if you wish.

How we built it

We built the backend using Node + Express, while the frontend was built using pure HTML, CSS, and JS. The server scrapes the code using Puppeteer, and caches its findings using Firebase to avoid scraping the same things.

Challenges we ran into

Most of our issues boiled down to us not having prior Firebase experience. During our transition to Firebase Functions for caching, our Google Cloud Project was not being listed as being able to be linked to the Firebase project. Luckily, one of the mentors who was highly experienced with Firebase was able to help us with that issue.

Another issue we had earlier on was finding the right APIs to use for our project. We were looking for an NLP API capable of identifying entities in text and their saliences, and we were considering using a dedicated solution before realizing GCP (Google Cloud Platform) had the Google Cloud Natural Language, which was exactly what we needed.

Accomplishments that we're proud of

Albeit very time-consuming, search caching is a feature which we're very happy with, especially with how performant it is. It takes little to no time at all to retrieve past searches, which makes for a rather satisfying experience!

The styling of the webpage also looked bleak at the beginning, but ended up becoming much smoother in the final product.

What we learned

All of us who worked on Grep Health learned something. This is the first time most of us had worked with the GCP, and some haven't even worked with databases before this event. Though, more importantly, we all learned something about collaboration. Working together as a team brought us closer to our goals than we could have thought.

What's next for Grep Health

If the project is well-recieved, we will try to push out the features we missed and clean up the dirty spots.

Share this project:

Updates