LureLogic: Project Description

What Inspired My Idea

This project was inspired by my experience volunteering to teach kids how to fish. I saw firsthand that the biggest barrier for a new angler isn't the gear; it's the "wall of choice": that moment of staring into a tackle box with no idea where to start. This problem of "angler uncertainty" creates a barrier to entry that I wanted to solve.

How It Connects to My Hobby

As an avid angler myself, I know that lure selection is a complex skill learned over years. It's a "gut feeling" based on weather, water conditions, and time of day. My goal was to build a "digital co-pilot" that could quantify that gut feeling. I wanted to make a predictive model that moves from guesswork to data-driven guidance, helping both new and experienced anglers feel more confident on the water.

What Technologies I Used

I engineered a complete, end-to-end IoT-to-AI system:

  • IoT Sensor Hub (Hardware): The "senses" of the project. I built a custom device using an Arduino R4 WiFi to read from a DS18B20 waterproof temperature sensor, a DHT11 (air temp/humidity), and a BMP280 (barometric pressure). This device hosts its own web server to provide live data.

  • AI Backend (The "Brain"): A Python Flask server hosts the machine learning model. The model itself is a Scikit-learn KNeighborsClassifier trained on my personal 1,000-catch dataset, which uses the live sensor data to find patterns and make predictions.

  • Frontend (The Interface): A clean, mobile-first web app built with HTML, Tailwind CSS, and JavaScript. It fetches data from the Arduino, sends it to the Python server, and displays the top 3 lure recommendations in a simple, ranked list.

What I Learned While Building It

While I was forced to learn a lot about everything from predictive modeling to soldering, my main takeaway was learning to spend real effort creating a realistic user with a real problem and creating the best solution for them. I'm passionate about what I've made and how it can help people.

Built With

Share this project:

Updates