Machine Learning in Education

This project explores how machine learning concepts can be made accessible through structured, hands-on learning systems that connect abstract ideas with physical interaction.

Overview

Approach

System Design

Key Insight

Machine learning is often taught as an abstract, code-heavy discipline, which can make it difficult for beginners to build intuition.

This project investigates how structured instructional design, combined with physical systems, can make machine learning more intuitive and accessible.

Instead of focusing only on models and code, the goal was to create a learning experience where users interact with systems and understand behavior through guided exploration.

Instructional Design
Developed structured, step-by-step learning guides using Dozuki to break down complex technical processes.
Hardware Integration
Used Arduino-based systems to create physical interactions that reflect computational processes.
3D Modeling
Designed supporting components in SolidWorks to enable hands-on engagement.
Concept Translation
Transformed abstract machine learning ideas into intuitive, observable behaviors.

Learning Pipeline

The system was designed as a feedback-driven learning loop:

Concept → Instruction → Interaction → Feedback → Understanding

Each stage reinforces the next, allowing learners to move from abstract ideas to concrete understanding.

Machine learning becomes significantly easier to understand when it is grounded in physical interaction and structured guidance.

Learners are able to build intuition not by memorizing models, but by observing how systems behave and responding to feedback.

This suggests that the future of technical education lies not just in better tools-but in better systems for translating complexity into experience.

← Back