I ended up creating three different final product proposals. Surprisingly (or perhaps unsurprisingly), all of them had to do with machine learning. My first product proposal was the most ambitious and amorphous of the three. I didn't even know how to Google it using the appropriate terminology, so I have no idea if it has been attempted before. It probably has. The second and third proposals had to do with distributed denial of service (DDoS) mitigation and automated filtering of phishing emails. If you're interested, you can view my proposals in the "Projects" section of this website.
Why is it that I always come back to machine learning? It's not as though ML is a *particularly* revolutionary; It's just applied data science. Is it because it sounds cool? Maybe. Is it because I already have some experience with it, making ML projects more accessible to me? Probably. I just can't help but feel that I'm hopping on some sort of bandwagon here. ML is great, but it certainly isn't the optimal solution for every problem. I'm sure there's an infinite number of possibilities out there - I just don't have the knowledge or experience to take on any of them with the resources and time I have at my disposal. Those are just my personal reservations, in any case. I'm still excited to be working with neural networks again. As stressful as last year's project was, I'll be sure to have a lot more fun with this one.