net.librec.recommender.cf.ranking

Class SLIMRecommender

java.lang.Object

net.librec.recommender.AbstractRecommender

net.librec.recommender.cf.ranking.SLIMRecommender

All Implemented Interfaces:

Recommender

@ModelData(value={"isRanking","slim","coefficientMatrix","trainMatrix","similarityMatrix","knn"})
public class SLIMRecommender
extends AbstractRecommender

Xia Ning and George Karypis, SLIM: Sparse Linear Methods for Top-N Recommender Systems, ICDM 2011.

Related Work:

• Levy and Jack, Efficient Top-N Recommendation by Linear Regression, ISRS 2013. This paper reports experimental results on the MovieLens (100K, 10M) and Epinions datasets in terms of precision, MRR and HR@N (i.e., Recall@N).
• Friedman et al., Regularization Paths for Generalized Linear Models via Coordinate Descent, Journal of Statistical Software, 2010.

Author:

guoguibing and Keqiang Wang

Field Summary

Fields

Modifier and Type	Field and Description
protected static int	knn number of nearest neighbors
protected int	numIterations the number of iterations

Fields inherited from class net.librec.recommender.AbstractRecommender

conf, context, decay, earlyStop, globalMean, isBoldDriver, isRanking, itemMappingData, lastLoss, LOG, loss, maxRate, minRate, numItems, numRates, numUsers, ratingScale, recommendedList, testMatrix, topN, trainMatrix, userMappingData, validMatrix, verbose

Constructor Summary

Constructors

Constructor and Description
SLIMRecommender()

Method Summary

Methods

Modifier and Type	Method and Description
void	createItemNNs() Create item KNN list.
protected boolean	isConverged(int iter) Post each iteration, we do things: print debug information check if converged if not, adjust learning rate
protected double	predict(int userIdx, int itemIdx) predict a specific ranking score for user userIdx on item itemIdx.
protected double	predict(int userIdx, int itemIdx, int excludedItemIdx) predict a specific ranking score for user userIdx on item itemIdx.
protected void	setup() initialization
protected void	trainModel() train model

Methods inherited from class net.librec.recommender.AbstractRecommender

cleanup, evaluate, evaluateMap, getContext, getDataModel, getRecommendedList, loadModel, predict, recommend, recommend, recommendRank, recommendRating, saveModel, setContext

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

numIterations

protected int numIterations

the number of iterations

knn

protected static int knn

number of nearest neighbors

Constructor Detail

SLIMRecommender

public SLIMRecommender()

Method Detail

setup

protected void setup()
              throws LibrecException

initialization

Overrides:

setup in class AbstractRecommender

Throws:

LibrecException - if error occurs

trainModel

protected void trainModel()
                   throws LibrecException

train model

Specified by:

trainModel in class AbstractRecommender

Throws:

LibrecException - if error occurs

predict

protected double predict(int userIdx,
             int itemIdx,
             int excludedItemIdx)

predict a specific ranking score for user userIdx on item itemIdx.

Parameters:

userIdx - user index

itemIdx - item index

excludedItemIdx - excluded item index

Returns:

a prediction without the contribution of excluded item

isConverged

protected boolean isConverged(int iter)

Description copied from class: AbstractRecommender

Post each iteration, we do things:

1. print debug information
2. check if converged
3. if not, adjust learning rate

Overrides:

isConverged in class AbstractRecommender

Parameters:

iter - current iteration

Returns:

boolean: true if it is converged; false otherwise

predict

protected double predict(int userIdx,
             int itemIdx)
                  throws LibrecException

predict a specific ranking score for user userIdx on item itemIdx.

Specified by:

predict in class AbstractRecommender

Parameters:

userIdx - user index

itemIdx - item index

Returns:

predictive ranking score for user userIdx on item itemIdx

Throws:

LibrecException - if error occurs

createItemNNs

public void createItemNNs()

Create item KNN list.