Inspiration
We want to make it easier for city developers and environmentalists to access aggregate data from multiple datasets at one time, in such a way that it requires little to no knowledge of statistics.
What it does
Blobs provides aggregate analytical data from a variety of datasets, in an easy to digest way that allows anyone to use it for development purposes to environmental awareness.
How I built it
The variety of skills present across our team proved to be helpful. We decided to build a web-app using React. We originally planned on using Flask and Python for our back-end, as we were planning on incorporating massive datasets to plot on our map. However, we eventually switched over to using Firebase's realtime database while still using Python to do so, as we discovered that we could both deploy our app as well as store data with Firebase. All of us got to work on the front-end of things (React, HTML/CSS) as well as the back-end with Python, manipulating the fetched data into a format that our front-end could easily use.
Challenges I ran into
The bulk of the project was getting to work with the different datasets from different APIs. It proved to especially difficult to manipulate the API data into a format that our map API could read as a heatmap, and so much of our efforts in the first half involved getting a heatmap to even show up.
Accomplishments that I'm proud of
This was most of our first times working with external APIs, as well as our first times making a project from the group up, from scratch, to deployment. We were also able to try out development strategies such as Agile during the making of our app, which definitely made us appreciate how much work needs to go into developing an application for public use.
What I learned
We learned a lot!!! For starters... we learned what NOT to do. Having little hackathon experience on our team (two first-timers and two beginners still), we felt really rushed right away to try and do everything as quickly as possible. Looking back, this turned out to be fatal, as we did not plan out exactly what we envisioned we could realistically accomplish as well as what would actually work. We spent a lot of (wasted) time trying to connect Flask and React, trying to add different heat maps, and doing other things that we ended up ditching after realizing they were too difficult or not needed. And so, we learned that it is really important to take some time to plan things, even if time is limited!!! The other things we learned was how to pick up code quickly. All of us were relatively decent at Python, but that was it, and three of us picked up React on the way, which as great. As mentioned before, we also learned a bit of setting up Flash, and though we didn't end up using it, it was still good to have the experience of trying something new.
What's next for Blobs
One of our biggest hopes is to incorporate machine learning into our project. The original reason we thought about combining Flask/Python with React/JavaScript was to use Python for machine learning on the datasets we fetched from the external APIs. Looking at our map, it is clearly noticeable that many places do not have data. Using machine learning, we want to provide best estimates of data (GPD, expected population growth, air quality, etc.) using different machine learning techniques. A cool feature to implement would be the ability to place a marker anywhere in the world and have our project give an estimate of the beforementioned features and explain how those estimates were created.
Log in or sign up for Devpost to join the conversation.