meta::classify::nearest_centroid Class Reference

Implements the nearest centroid classification algorithm. More...

#include <nearest_centroid.h>

Inheritance diagram for meta::classify::nearest_centroid:

Classes
class	nearest_centroid_exception
	Basic exception for nearest_centroid interactions. More...

Public Member Functions
	nearest_centroid (std::shared_ptr< index::inverted_index > idx, std::shared_ptr< index::forward_index > f_idx)
void	train (const std::vector< doc_id > &docs) override
	Creates a classification model based on training documents. More...
class_label	classify (doc_id d_id) override
	Classifies a document into a specific group, as determined by training data. More...
void	reset () override
	Resets any learning information associated with this classifier.
Public Member Functions inherited from meta::classify::classifier
	classifier (std::shared_ptr< index::forward_index > idx)
virtual confusion_matrix	test (const std::vector< doc_id > &docs)
	Classifies a collection document into specific groups, as determined by training data; this function will make repeated calls to classify(). More...
virtual confusion_matrix	cross_validate (const std::vector< doc_id > &input_docs, size_t k, bool even_split=false, int seed=1)
	Performs k-fold cross-validation on a set of documents. More...

Static Public Attributes
static const std::string	id = "nearest-centroid"
	Identifier for this classifier.

Private Member Functions
double	cosine_sim (const std::vector< std::pair< term_id, double >> &doc, const std::unordered_map< term_id, double > &centroid)

Private Attributes
std::shared_ptr< index::inverted_index >	inv_idx_
	Inverted index used for ranking.
std::unordered_map< class_label, std::unordered_map< term_id, double > >	centroids_
	The document centroids for this learner.

Additional Inherited Members
Protected Attributes inherited from meta::classify::classifier
std::shared_ptr< index::forward_index >	idx_
	the index that the classifer is run on

Detailed Description

Implements the nearest centroid classification algorithm.

nearest_centroid creates a prototype document for each distinct class as an average of all documents in that class. This is called the centroid. A query (testing document) is then compared against each centroid. The class label of the centroid they query is closest to is returned.

See also

Centroid-Based Document Classification: Analysis and Experimental Results, Eui-Hong Han and George Karypis, 2000

Constructor & Destructor Documentation

meta::classify::nearest_centroid::nearest_centroid	(	std::shared_ptr< index::inverted_index >	idx,
		std::shared_ptr< index::forward_index >	f_idx
	)

Parameters

idx	The index to run the classifier on

Member Function Documentation

void meta::classify::nearest_centroid::train ( const std::vector< doc_id > &  docs )

overridevirtual

Creates a classification model based on training documents.

Parameters

docs	The training documents

Implements meta::classify::classifier.

class_label meta::classify::nearest_centroid::classify ( doc_id  d_id )

overridevirtual

Classifies a document into a specific group, as determined by training data.

Parameters

d_id	The document to classify

Returns

the class it belongs to

Implements meta::classify::classifier.

double meta::classify::nearest_centroid::cosine_sim ( const std::vector< std::pair< term_id, double >> &  doc, const std::unordered_map< term_id, double > &  centroid  )

private

Parameters

d_id
centroid

Returns

the cosine similarity between the query and a centroid

---

The documentation for this class was generated from the following files:

• /home/chase/projects/meta/include/classify/classifier/nearest_centroid.h
• /home/chase/projects/meta/src/classify/classifier/nearest_centroid.cpp