book

Practical Machine Learning: A New Look at Anomaly Detection

August 2014

Intermediate to advanced

66 pages

1h 25m

English

Read now

Unlock full access

Finding “Normal”If you enjoy math, read this description of a probabilistic model of “normal”…Human Insight HelpsFinding AnomaliesOnce again, if you like math, this description of anomalies is for you…Take-Home Lesson: Key Steps in Anomaly DetectionA Simple Approach: Threshold Models
The Philosophy Behind Setting the ThresholdUsing t-Digest for Accurate Calculation of Extreme QuantilesIssues with Simple Thresholds
Windows and ClustersMatches with the Windowed Reconstruction: Normal FunctionMismatches with the Windowed Reconstruction: Anomalous FunctionA Powerful But Simple TechniqueLooking Toward Modeling More Problematic Inputs
Counts Don’t Work WellArrival Times Are the KeyAnd Now with the Math…Event Rate in a Worked Example: Website Traffic PredictionExtreme Seasonality Effects
The Phishing AttackThe No-Phishing-Allowed Anomaly DetectorHow the Model WorksPutting It All Together
GitHubApache Mahout Open Source ProjectAdditional Publications

Colophon

The text font is Adobe Minion Pro; the heading font is Adobe Myriad Condensed; and the code font is Dalton Maag’s Ubuntu Mono.

Vikram Chandra Jha, Atul Tripathi

Jojo Moolayil

Samuel Holt

Santanu Pattanayak

ISBN: 9781491914151Errata