Package diva.sketch.classification

This package provides basic pattern classification.

See: Description

Interface Summary

Interface	Description
Classifier	A Classifier performs generic classification on feature sets, the semantics of which it knows nothing about.
TrainableClassifier	A Classifier performs generic classification on feature sets, the semantics of which it knows nothing about.

Class Summary

Class	Description
AbstractClassifier	Given a training set containing multiple classes, for each class, an AbstractClassifier would compute the mu and sigma of each feature of that class.
BayesClassifier	A naive bayes classifier.
Classification	Data structure for storing a list of classifer type and confidence value pairs in the order of descending confidence values.
CrossValidation
CrossValidation.CVData	Containing the training set on which the cross-vali
CrossValidation.CVResult
FeatureSet	A data structure for storing features for a classifier; it is basically a typesafe array of doubles with appropriate accessor methods.
GaussianWeightSet	Given a set of training examples (each example is a feature vector), a Gaussian classifier computes the mu and sigma for each type of features.
KNNClassifier	A K Nearest Neighbor classifier compares a given example (feature set) to the training set and make its prediction based on the majority match in the top K candidates.
RubineClassifier	This classifier implements the classic linear discriminator.
TrainingSet	A TrainingSet contains a set of types, and for each type a corresponding set of positive and negative examples.
WeightedEuclideanClassifier	WeightedEuclideanClassifier is a trainable classifier that uses a weighted N-dimensional distance between feature sets to classify its input.
WeightSet	A WeightSet object represents a training type (e.g.

Exception Summary

Exception	Description
ClassifierException	Thrown when there is some internal error in the training or classification process.

Package diva.sketch.classification Description

This package provides basic pattern classification. A classifier is trained with a set of examples in the form of feature vectors. Once the classifier is trained, it can classify input feature vectors. The resulting classification is a list of [type, confidence] pairs according to the similarity of the input vector with the examples.