Stay organized with collections Save and categorize content based on your preferences.
To cluster your data, you'll follow these steps:
Prepare data.
Create similarity metric.
Run clustering algorithm.
Interpret results and adjust your clustering.
This page briefly introduces the steps. We'll go into depth in subsequent sections.
Prepare data
As with any ML problem, you must normalize, scale, and transform feature data before training or fine-tuning a model on that data. In addition, before clustering, check that the prepared data lets you accurately calculate similarity between examples.
Create similarity metric
Before a clustering algorithm can group data, it needs to know how similar pairs of examples are. You can quantify the similarity between examples by creating a similarity metric, which requires a careful understanding of your data.
Run clustering algorithm
A clustering algorithm uses the similarity metric to cluster data. This course uses k-means.
Interpret results and adjust
Because clustering doesn't produce or include a ground "truth" against which you can verify the output, it's important to check the result against your expectations at both the cluster level and the example level. If the result looks odd or low-quality, experiment with the previous three steps. Continue iterating until the quality of the output meets your needs.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-02-25 UTC."],[[["\u003cp\u003eData needs to be prepared through normalization, scaling, and transformation before using it for clustering.\u003c/p\u003e\n"],["\u003cp\u003eA similarity metric is crucial for clustering algorithms as it quantifies how similar data points are to each other.\u003c/p\u003e\n"],["\u003cp\u003eThe k-means algorithm is employed in this course to group data based on the defined similarity metric.\u003c/p\u003e\n"],["\u003cp\u003eEvaluating and adjusting clustering outcomes is an iterative process involving reviewing cluster quality and individual data point assignments.\u003c/p\u003e\n"]]],[],null,["\u003cbr /\u003e\n\nTo cluster your data, you'll follow these steps:\n\n1. Prepare data.\n2. Create similarity metric.\n3. Run clustering algorithm.\n4. Interpret results and adjust your clustering.\n\nThis page briefly introduces the steps. We'll go into depth in subsequent\nsections.\n\nPrepare data\n\nAs with any ML problem, you must normalize, scale, and transform feature data\nbefore training or fine-tuning a model on that data. In addition, before\nclustering, check that the prepared data lets you accurately calculate\nsimilarity between examples.\n| **Review:** For a review of data transformation, see [Working with numerical data](/machine-learning/crash-course/numerical-data) from Machine Learning Crash Course.\n\nCreate similarity metric\n\nBefore a clustering algorithm can group data, it needs to know how similar\npairs of examples are. You can quantify the similarity between examples by\ncreating a similarity metric, which requires a careful understanding of your\ndata.\n\nRun clustering algorithm\n\nA clustering algorithm uses the similarity metric to cluster data.\nThis course uses k-means.\n\nInterpret results and adjust\n\nBecause clustering doesn't produce or include a ground \"truth\" against which you\ncan verify the output, it's important to check the result against your\nexpectations at both the cluster level and the example level. If the result\nlooks odd or low-quality, experiment with the previous three steps. Continue\niterating until the quality of the output meets your needs."]]