Skip to main content
Colin Jaffe/2 min read

Exploring the K-Nearest Neighbors Algorithm

Common ML Algorithms

Linear/Logistic Regression

Interpretable baselines — start here.

Random Forest

Robust, handles mixed data, minimal tuning.

Gradient Boosting

XGBoost/LightGBM dominate tabular ML competitions.

Neural Networks

Best for images, audio, text, and high-dimensional data.

Master Machine Learning at Noble Desktop

Noble Desktop's Python Machine Learning Bootcamp covers scikit-learn, Keras, neural networks, and applied ML.

This lesson is a preview from our Data Science & AI Certificate Online (includes software) and Python Certification Online (includes software & exam). Enroll in a course for detailed lessons, live instructor support, and project-based training.

Here in part three, we're going to be working on the k-nearest neighbors machine learning algorithm. The k-nearest neighbors algorithm is a supervised machine learning algorithm for classifying data points based on, hey, what is the value of the closest existing points?

Here in part three, we're going to be working on the k-nearest neighbors machine learning algorithm. The k-nearest neighbors algorithm is a supervised machine learning algorithm for classifying data points based on, hey, what is the value of the closest existing points? So it learns in a different way than the original algorithms, where we did regressions.

KNN is not a regression. KNN looks instead at its memory of the data points it has seen. Right? Again, we give it, hey, this data point is this value.

This data point is this value. And we can plot all kinds of things with that. For example, height and weight.

We could plot X against Y and get a lot of data points of height and weight, maybe of animals. And be able to classify, saying, "Okay, this height and weight area over here seems to be dogs, " and this one over here seems to be cats.

We're going to look at some data next in the second part of part three. And we're going to look at some flower data and classify flowers in a classic machine learning dataset. But for now, we're going to explore just the concept of k-nearest neighbors and try to visualize it and try to get a sense of what it's doing.

All right. So very first thing, let's make sure we've got all of our ducks in a row regarding getting everything imported. We've combined a couple of things here.

But these are our basic data science imports, including our Jupyter Notebook ability to display images. Here we have our k-nearest neighbors classifier, our model. And we've also set up Google Drive.

So run that. And then as is typical, the base URL for any files we need to look at. Execute that.

And we can also execute this and start talking about k and N. All right. We'll talk about this image and what it means in just a moment.