How I won UNICEF Data Visualization Hackathon
I am not joking this time. It’s been so long since I last wrote on medium. A lot of things changed since then. I deleted FB. I deleted LinkedIn. Most importantly, my focus shifted from competitive programming. Yeah, I do like it. I’d rather solve a puzzle than playing some kind of computer games because I suck at games. I just don’t want to give much time to be the best. I’d rather learn maching learning.
When reading back my posts, I don’t even understand what the hell I was writing about. I promise I would write in a very simple language, with baby steps. Like some great dude said,
“If you can’t explain in a simple language, you don’t get it.”
I’ve been studying two algorithms specialization on Coursera. You know what? I applied financial aids to all courses. I truly can’t afford otherwise. You may do it like me if you can’t afford. I will write about the lessons I learnt. I feel the need to share. It’s win-win. For various reasons and excuses, I couldn’t. I added it in 2018 resolutions. So, stay tuned for algorithms stories and of course, competitive puzzles. I am trying to study machine learning courses after it. Creating super cool A.I is my dream since I started loving Computer Science. Machine learning is one way to do it. I am learning it because I love. I don’t expect a career with it. I don’t care I have to beg for food. I will follow what I love. I will see how far I will go. (MOANA).
Let’s get back to Data Visualization Hackathon. It was about ten days before I learnt about the hackathon. At that time, I didn’t know how to draw graphs. I never did with programming. Yeah, of course, I took a few courses on marketing and marketing researches about two years ago. I know how to make sense of data although not an expert at data. I draw a few graphs with Powerpoint, Excel!! Well, they were dots that connected in this case. Then, I realized what Steve Jobs meant by connecting the dots.
All I need to do is to use some libraries to plot the graphs. And then, I asked on Quora
What is the best data visualization library to integrate with React frontend and Firebase backend…
Answer (1 of 2): Unless you are dealing with weird clients, I would recommend going with the 800 pound gorilla in the…
At first, I overestimated my abilities to code. I couldn’t use React and Firebase in the hackathon. I used HTML, CSS, jQuery, D3 for front-end app. As suggested by the answer on Quora, I chose D3 undoubtedly. I looked for excellent resources to learn. I had only five days to finish graphs if I was to be ready before the event. So, I purchased a course on Udemy.
Learn and Understand D3.js for Data Visualization
Dive deep under the hood of D3.js. Learn the core concepts and build stunning data visualizations for your audience.
It was very good and worth the money. Then, I studied it all time except sleep and eat. I started on Sunday and finished drawing graphs I think I would need for hackathon on Thursday. Hackathon was from Friday to Sunday. I also studied how to choose graphs to visualize data by a free course on Udacity.
On Friday, we couldn’t do much except forming a team. I got a bad news. The choropleth map of Myanmar I drew already had only 14 states. Myanmar added a new state some time ago. Then, I have to coded and solved it till 2:00 a.m SAT when I got home. Saturday was real coding marathon. In my team, there were a data analyst, data collector (from API) and UX/UI designer and me (Coding). They were very helpful. The UX/UI designer is my younger brother. I met the other two there. So, we worked together from 9 to 5. It was great experience. However, we didn’t finished the graphs. So, I had to code them from 9 p.m to 3 a.m when I got back home.
Sunday was a moment of truth. The graph the judge liked best is scatter plot. I drew it to find the correlation between number of people with no access to clean water and number of people with no clean sanitation. I knew about scatter plot when I took a lecture on Machine Learning with Tensorflow. It uses many scatter plots. I didn’t understand why the teacher used it a lot.