knn.itemToItem.similarities.ItemsSimilarities Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of cf4j-recsys Show documentation

A Java's Collaborative Filtering library to carry out experiments in research of Collaborative Filtering based Recommender Systems. The library has been designed from researchers to researchers.

The newest version!

package cf4j.knn.itemToItem.similarities;

import cf4j.Item;
import cf4j.Kernel;
import cf4j.TestItem;
import cf4j.TestItemsPartible;

/**
 * This class process the similarity measure between two items. If you want to define your own similarity
 * metric implementation, you must extend this class and implements the abstract method similarity (...).
 * 
 * When the execution of the similarity metric is complete, all the test items will have a double array
 * on his map saved with the key "similarities". The similarities can be retrieved using getSimilarities() 
 * method of TestItem class. The positions of this array overlaps with the array returned by the method 
 * getItems() of the Kernel class. For example, testItem.getSimilarities()[i] will contains the similarity 
 * between testItem and Kernel.getInstance().getItems()[i] item.
 * 
 * @author Fernando Ortega
 */
abstract public class ItemsSimilarities implements TestItemsPartible {

	/**
	 * Method to calculate the similarity measure between a pair of items.
	 * If not able to calculate the similarity measure between two items Double.NEGATIVE_INIFINITY 
	 * is returned.
	 * The similarity measure must be greater the more similar are the items.
	 * @param activeItem Active item
	 * @param targetItem Item with which the similarity is computed
	 * @return Similarity measure between the two items
	 */
	abstract public double similarity (TestItem activeItem, Item targetItem);

	@Override
	public void beforeRun () { }

	@Override
	public void run (int testItemIndex) {
		
		TestItem activeItem = Kernel.getInstance().getTestItems()[testItemIndex];
		
		int numItems = Kernel.gi().getNumberOfItems();		
		double [] similarities = new double [numItems];
		
		for (int i = 0; i < numItems; i++) {
			Item targetItem = Kernel.gi().getItems()[i];
			if (activeItem.getItemCode() == targetItem.getItemCode()) {
				similarities[i] = Double.NEGATIVE_INFINITY;
			} else { 
				similarities[i] = this.similarity(activeItem, targetItem);
			}
		}
		
		activeItem.setSimilarities(similarities);
	}
	
	@Override
	public void afterRun () { }
}