1.1.1 Tools: Open-Source Revolution
2.2.1 Classifier Evaluation in Data Streams
2.2.2 Majority Class Classifier
3 Hands-on Introduction to MOA
3.2 The Graphical User Interface for Classification
4.1 Setting: Approximation Algorithms
4.2 Concentration Inequalities
4.5 Counting Distinct Elements
4.5.2 Cohen’s Logarithmic Counter
4.5.3 The Flajolet-Martin Counter and HyperLogLog
4.5.4 An Application: Computing Distance Functions in Graphs
4.5.5 Discussion: Log vs. Linear
4.6.1 The S PACE S AVING Sketch
4.7 Exponential Histograms for Sliding Windows
4.8 Distributed Sketching: Mergeability
4.9 Some Technical Discussions and Additional Material
4.9.2 Creating ( 𝜖 , δ )-Approximation Algorithms
4.9.3 Other Sketching Techniques
5.1 Notion of Change in Streams
5.2.1 Sliding Windows and Linear Estimators
5.2.2 Exponentially Weighted Moving Average
5.2.3 Unidimensional Kalman Filter
5.3.1 Evaluating Change Detection
5.3.2 The CUSUM and Page-Hinkley Tests
5.4 Combination with Other Sketches and Multidimensional Data
6.1.3 Performance Evaluation Measures
6.1.4 Statistical Significance
6.1.5 A Cost Measure for the Mining Process
6.3.1 Estimating Split Criteria
6.4 Handling Numeric Attributes
6.4.3 Greenwald and Khanna’s Quantile Summaries
6.7 Multi-label Classification
6.7.1 Multi-label Hoeffding Trees
6.8.2 Fixed Uncertainty Strategy
6.8.3 Variable Uncertainty Strategy
6.8.4 Uncertainty Strategy with Randomization
7.1 Accuracy-Weighted Ensembles
7.4.1 Online Bagging Algorithm
7.4.2 Bagging with a Change Detector
7.6 Ensembles of Hoeffding Trees
7.6.3 Perceptron Stacking of Restricted Hoeffding Trees
7.6.4 Adaptive-Size Hoeffding Trees
9.3 BIRCH, BICO, and C LU S TREAM
9.4 Density-Based Methods: DBSCAN and Den-Stream
10.1 An Introduction to Pattern Mining
10.1.1 Patterns: Definitions and Examples
10.1.2 Batch Algorithms for Frequent Pattern Mining
10.1.3 Closed and Maximal Patterns
10.2 Frequent Pattern Mining in Streams: Approaches
10.2.1 Coresets of Closed Patterns
10.3 Frequent Itemset Mining on Streams
10.3.1 Reduction to Heavy Hitters
10.4 Frequent Subgraph Mining on Streams
11 Introduction to MOA and Its Ecosystem
11.3 Recent Developments in MOA
12 The Graphical User Interface
12.1 Getting Started with the GUI
12.2 Classification and Regression
12.2.2 Data Feeds and Data Generators
12.2.5 Meta Classifiers (Ensembles)
12.2.8 Active Learning Classifiers
12.3.1 Data Feeds and Data Generators
12.3.2 Stream Clustering Algorithms
12.3.3 Visualization and Analysis
13.1 Learning Task for Classification and Regression
13.2 Evaluation Tasks for Classification and Regression
13.3 Learning and Evaluation Tasks for Classification and Regression
13.4 Comparing Two Classifiers
14.3 Prequential Evaluation Example
15 Developing New Methods in MOA
15.2 Creating a New Classifier