Algorithms

What kind of data mining algorithms were implemented.

Classification methods

Binary Classification

PegasosSVM

• The update method

    Input: x, y
    y' = w * x
    nu = 1 / (lambda * t)
    w(t+1) = (1 - 1/t) * w(t)
    if (y' * y < 1)
        w(t+1) += nu * y * x
    t ++
  

• The label prediction

    Input: x
    return (w * x < 0) ? -1 : 1
  

Perceptron

NormaSVM

Logistic Regression

Winnow?

Multi - Label Classification

Multi - Logistic Regression

FilterBoost

• ConstantLearner
• SigmoidStumpLearner
• AdalineLearner
• ProductLearner
• SequentialLearner

Adaptive Classification

Age Variance

Drift Detection

Regression

Linear Regression

NormaSVM

Clustering methods

K-Means

GMM