Introduction

This section of the document aims to provide a brief explanation of some of the technical terms used throughout the Wiki. The explanations given here will be aimed at the embedded ML world.

Terms

Term	Area used	Explanation
Backpropagation	Machine Learning	Algorithm designed to look for errors, working from output nodes back to input nodes
Branch	Decision Tree	An entire tree's subsection
Classification	Algorithms (All)	A supervised learning task where the aim is to predict a class label for a new data point
Decision Node	Decision Trees	The place where a sub-node splits into another sub-node
Latency	TinyML	The time an embedded system takes to receive data, process it, and generate a response
Leaf Node	Decision Tree	A node that does not split
Over-fitting	Algorithms (All)	Occurs when a model starts to fit the noise in the data, rather than the underlying pattern
Regression	Algorithms (All)	A supervised learning technique that uses historical data as input to predict a new or continuous value
Root Node	Algorithms (Decision Tree)	Represents entire dataset or population; Is split into two or more sets
Splitting	Decision Trees	Process of splitting a node into two or more sets
Under-fitting	Algorithms (All)	Occurs when the model is not sufficient to capture the underlying pattern in the data