I love AI, but maybe not the AI that you’re thinking about.
Before the AI hype, I took a machine learning course at the University of Cincinnati. It was by far one of my favorite courses. The reason? It taught me so much about what ML/AI can do and why it’s so important. But first, let me explain the basics of machine learning.
Machine learning is when you give software data, and it makes connections and predictions that our soft human heads just can’t. This is cool because it truly helps us understand the rules of our world—rules we don’t even know exist yet.
Algorithms I want to talk about:
- Linear Regression: Predicting continuous values, which means it’s good for data that is linear.
- Example: Predicting a runner’s finish time based on training data.
- Example: Predicting a runner’s finish time based on training data.
- Logistic Regression: Predicts if an element fits in a classification or not. Basically, it guesses True or False—simple, but massive.
- Example: Predicting if a patient has a condition or not. This has already shaken up the medical world, even before the OpenAI craze. It allows for earlier diagnoses and better treatment plans.
- Example: Predicting if a patient has a condition or not. This has already shaken up the medical world, even before the OpenAI craze. It allows for earlier diagnoses and better treatment plans.
- Decision Tree: Predicts the most correct decision based on the data given. These are classification problems, but the algorithm figures out which data points carry more or less weight—making it more accurate than us.
- Example: Detecting fraud based on things like store, transaction amount, and frequency. This means exploited people can get help faster than ever.
- Example: Detecting fraud based on things like store, transaction amount, and frequency. This means exploited people can get help faster than ever.
- K-Nearest Neighbors: Clustering is like finding out who your data’s closest friends are. The idea is simple—if most of your nearest neighbors are one thing, then you probably are too.
- Example: Recommending content. If you’ve watched five documentaries about space, and so have 200 other users, it’s likely you’ll enjoy what they watched next too. That’s how Netflix and YouTube do their thing.
I think it’s easy to get swept up in the buzzwords—”AGI”, “LLMs”, “autonomous agents”—but machine learning at its core is about using data to improve real-world decisions. It’s not always flashy, but it’s deeply practical.
- Example: Recommending content. If you’ve watched five documentaries about space, and so have 200 other users, it’s likely you’ll enjoy what they watched next too. That’s how Netflix and YouTube do their thing.
Understanding how these algorithms work demystifies a lot of the “magic” around AI and makes you realize how much of it is built on logical, approachable ideas. And honestly, that’s what made me fall in love with it in the first place.
If you’ve been curious about AI but feel overwhelmed by the hype—start here. This is the foundation.
Would love to hear what others are learning about AI right now!