C# Класс MyMediaLite.Correlation.AdjustedCosine

Class for (shrunk) adjusted cosine similarity

Badrul Sarwar, George Karypis, Joseph Konstan, John Riedl: Item-based collaborative filtering recommendation algorithms. WWW 2001

Наследование: RatingCorrelationMatrix

Открытые свойства

Свойство	Тип	Описание
shrinkage	float

Метод	Описание
AdjustedCosine ( int num_entities ) : System	Constructor. Create a AdjustedCosine matrix
ComputeCorrelation ( IRatings ratings, EntityType entity_type, int i, int j, float shrinkage ) : float	Compute correlations between two entities for given ratings
ComputeCorrelations ( IRatings ratings, EntityType entity_type ) : void	Compute correlations for given ratings
Create ( IRatings ratings, EntityType entity_type, float shrinkage ) : CorrelationMatrix	Create a AdjustedCosine matrix from given data

Constructor. Create a AdjustedCosine matrix

public AdjustedCosine ( int num_entities ) : System
num_entities	int	the number of entities
Результат	System

Compute correlations between two entities for given ratings

public static ComputeCorrelation ( IRatings ratings, EntityType entity_type, int i, int j, float shrinkage ) : float
ratings	IRatings	the rating data
entity_type	EntityType	the entity type, either USER or ITEM
i	int	the ID of first entity
j	int	the ID of second entity
shrinkage	float	the shrinkage parameter
Результат	float

Compute correlations for given ratings

public ComputeCorrelations ( IRatings ratings, EntityType entity_type ) : void
ratings	IRatings	the rating data
entity_type	EntityType	the entity type, either USER or ITEM
Результат	void

Create a AdjustedCosine matrix from given data

public static Create ( IRatings ratings, EntityType entity_type, float shrinkage ) : CorrelationMatrix
ratings	IRatings	the ratings data
entity_type	EntityType	the entity type, either USER or ITEM
shrinkage	float	a shrinkage parameter
Результат	CorrelationMatrix

shrinkage parameter

public float shrinkage
Результат	float