[MLS-C01] [Algorithms] Clustering Algorithms

Defined

• Unsupervised learning algorithm
• Attempts to find discrete groupings within data, where members of a group are as similar as possible to one another and as different as possible from members of other groups
• Define the attributes that you want the algorithm to use to determine similarity

Use Cases

1. Delivery source location
2. Identifying crime centers
3. Customer segmentation
4. Fraud detection based on clusters of fraud patterns
5. Cyber-profiling criminals
6. Clustering of IT alerts
7. Call center recording analysis

SageMaker Algorithms

K-Means

• Expects tabular data, where rows represent the observations that you want to cluster, and the columns represent attributes of the observations
• n attributes in each row represent a point in n-dimensional space
• Euclidean distance between these points represents the similarity of the corresponding observations
• Groups observations with similar attribute values (the points corresponding to these observations are closer together)
• Example use case: using census data find clusters of populations in counties across the US to focus political activity

Labs

1. Clustering Lab.ipynb
2. Absenteeism_at_work.csv

本文由 Oscaner 创作, 采用 知识共享署名4.0 国际许可协议进行许可
本站文章除注明转载/出处外, 均为本站原创或翻译, 转载前请务必署名

• Previous
  [MLS-C01] [Algorithms] Regression Algorithms
• Next
  [MLS-C01] [Algorithms] Classification Algorithms