Oreilly – Outlier Detection in Python, Video Edition 2024-12

Outlier Detection in Python, Video Edition. This course teaches you how to identify unusual, interesting, or suspicious data in your data sets using Python. In addition to discovering patterns, data scientists must also recognize exceptions, as these anomalies often contain valuable insights, hidden problems, or new opportunities. In this course, you will learn practical methods for identifying points that deviate from the overall data pattern, even when these points are hidden among normal information. Topics include using Python’s standard libraries, choosing appropriate methods, combining different techniques to improve accuracy, and effectively interpreting results. Working with numeric, batch, text, and time series data types is also covered. Anomaly detection is used in areas such as fraud detection, security analysis, and data quality control. The course is packed with real-world examples from a variety of industries, including finance, social media, and networking, and teaches essential tools like scikit-learn and PyOD. Prerequisites for this course include a basic understanding of statistics and the Python environment. By the end, you will be able to apply key algorithms and practical techniques to identify and analyze anomalies in a variety of data.

What you will learn:

• Using Python libraries to identify outliers
• Choosing the most appropriate outlier detection methods
• Combining multiple outlier detection methods to improve results
• Effective interpretation of outlier detection results
• Working with numerical data to detect outliers
• Working with batch data to detect outliers
• Working with time series data to detect outliers
• Working with text data to detect outliers

Who is this course suitable for?

• Python programmers who are familiar with tools like pandas and NumPy.
• People who have basic knowledge in statistics.
• Anyone interested in identifying anomalies and outliers in data.
• Data scientists looking to learn outlier detection techniques.
• Data analysts who want to improve their skills in identifying outliers.

Course details for Outlier Detection in Python, Video Edition

• Publisher: Oreilly
• Instructor: Brett Kennedy
• Training level: Beginner to advanced
• Training duration: 19 hours and 35 minutes

Course headings

• Part 1.
  Chapter 1. Introducing outlier detection
  Chapter 1. Outlier detection’s place in machine learning
  Chapter 1. Outlier detection in tabular data
  Chapter 1. Definitions of outliers
  Chapter 1. Trends in outlier detection
  Chapter 1. How does this book teach outlier detection?
  Chapter 1. Summary
  Chapter 2. Simple outlier detection
  Chapter 2. One-dimensional categorical outliers: Rare values
  ​​Chapter 2. Multidimensional outliers
  Chapter 2. Rare combinations of categorical values
  ​​Chapter 2. Rare combinations of numeric values
  ​​Chapter 2. Noise vs. inliers and outliers
  Chapter 2. Local and global outliers
  Chapter 2. Combining the scores of univariate tests
  Chapter 2. Summary
  Chapter 3. Machine learning-based outlier detection
  Chapter 3. Types of algorithms
  Chapter 3. Types of detectors
  Chapter 3. Summary
  Chapter 4. The outlier detection process
  Chapter 4. Determining the types of outliers we are interested in
  Chapter 4. Choosing the type of model to be used
  Chapter 4. Collecting the data
  Chapter 4. Examining the data
  Chapter 4. Cleaning the data
  Chapter 4. Feature selection
  Chapter 4. Feature engineering
  Chapter 4. Encoding categorical values ​​Chapter 4.
  Scaling numeric values
  ​​Chapter 4. Fitting a set of models and generating predictions
  Chapter 4. Evaluating the models
  Chapter 4. Setting up ongoing outlier detection systems
  Chapter 4. Refitting the models as necessary
  Chapter 4. Summary
• Part 2.
  Chapter 5. Outlier detection using scikit-learn
  Chapter 5. Isolation Forest
  Chapter 5. LocalOutlierFactor (LOF)
  Chapter 5. One-class SVM (OCSVM)
  Chapter 5. Elliptic Envelope
  Chapter 5. Gaussian mixture models
  Chapter 5. BallTree and KDTree
  Chapter 5. Summary
  Chapter 6. The PyOD library
  Chapter 6. Histogram-based Outlier Score (HBOS)
  Chapter 6. Empirical Cumulative Distribution Function (ECOD)
  Chapter 6. Copula-based outlier detection (COPOD)
  Chapter 6. Angle-based outlier detection (ABOD)
  Chapter 6. Clustering-based local outlier factor (CBLOF)
  Chapter 6. Local correlation integral (LOCI)
  Chapter 6. Connectivity-based outlier factor (COF)
  Chapter 6. Principal component analysis (PCA)
  Chapter 6. Subspace outlier detection
  Chapter 6. FeatureBagging
  Chapter 6. Cook’s Distance
  Chapter 6. Using SUOD for faster model training
  Chapter 6. The PYOD thresholds module
  Chapter 6. Summary
  Chapter 7. Additional libraries and algorithms for outlier detection
  Chapter 7. The alibi-detect library
  Chapter 7. The PyCaret library
  Chapter 7. Local outlier probability (LoOP)
  Chapter 7. Local distance-based outlier factor (LDOF)
  Chapter 7. Extended Isolation Forest (EIF)
  Chapter 7. Outlier Detection Using In-degree Number (ODIN)
  Chapter 7. Clustering
  Chapter 7. Entropy
  Chapter 7. Association Rules
  Chapter 7. Convex Hull
  Chapter 7. Distance metric learning (DML)
  Chapter 7. NearestSample
  Chapter 7. Summary
• Part 3.
  Chapter 8. Evaluating detectors and parameters
  Chapter 8. Contour plots
  Chapter 8. Visualizing subspaces in real-world data
  Chapter 8. Correlation between detectors with full real-world datasets
  Chapter 8. Modifying real-world data
  Chapter 8. Testing with classification datasets
  Chapter 8. Timing experiments
  Chapter 8. Summary
  Chapter 9. Working with specific data types
  Chapter 9. Special data types
  Chapter 9. Text features
  Chapter 9. Encoding categorical data
  Chapter 9. Scaling numeric values
  ​​Chapter 9. Binning numeric data
  Chapter 9. Distance metrics
  Chapter 9. Summary
  Chapter 10. Handling very large and very small datasets
  Chapter 10. Data with many rows
  Chapter 10. Working with very small datasets
  Chapter 10. Summary
  Chapter 11. Synthetic data for outlier detection
  Chapter 11. Generating new synthetic data
  Chapter 11. Doping
  Chapter 11. Simulations
  Chapter 11. Training classifiers to distinguish real from fake data
  Chapter 11. Summary
  Chapter 12. Collective outliers
  Chapter 12. Preparing the data
  Chapter 12. Testing for duplicates
  Chapter 12. Testing for gaps Chapter 12.
  Testing for missing combinations
  Chapter 12. Creating new tables to capture collective outliers
  Chapter 12. Identifying trends
  Chapter 12. Unusual distributions
  Chapter 12. Rolling windows features
  Chapter 12. Tests for unusual numbers of point anomalies
  Chapter 12. Summary
  Chapter 13. Explainable outlier detection
  Chapter 13. Post hoc explanations
  Chapter 13. Interpretable outlier detectors
  Chapter 13. Summary
  Chapter 14. Ensembles of outlier detectors
  Chapter 14. Accuracy metrics with ensembles
  Chapter 14. Methods to create ensembles
  Chapter 14. Selecting detectors for an ensemble
  Chapter 14. Scaling scores
  Chapter 14. Combining scores
  Chapter 14. Summary
  Chapter 15. Working with outlier detection predictions
  Chapter 15. Examining the flagged outliers
  Chapter 15. Automating the process of sorting outlier detection results
  Chapter 15. Semisupervised learning
  Chapter 15. Regression testing
  Chapter 15. Summary
• Part 4.
  Chapter 16. Deep learning-based outlier detection
  Chapter 16. PyOD
  Chapter 16. Image data
  Chapter 16. alibi-detect
  Chapter 16. Self-supervised learning for outlier detection with tabular data
  Chapter 16. Summary
  Chapter 17. Time-series data
  Chapter 17. Types of time-series outliers
  Chapter 17. Tools for time-series data
  Chapter 17. Summary

Course images

Sample course video

Download link

Download Part 1 – 1 GB

Download Part 2 – 1 GB

Download Part 3 – 1 GB

Download Part 4 – 244 MB

File(s) password: www.downloadly.ir

File size

3.2 GB