Examples of org.apache.mahout.cf.taste.impl.model.file.FileDataModel$UserIteratorOverData

Package org.apache.mahout.cf.taste.impl.model.file

Examples of org.apache.mahout.cf.taste.impl.model.file.FileDataModel$UserIteratorOverData

org.apache.mahout.cf.taste.impl.model.file.FileDataModel

A {@link DataModel} backed by a comma-delimited file. This class typically expects a file where eachline contains a user ID, followed by item ID, followed by preferences value, separated by commas. You may also use tabs.

The preference value is assumed to be parseable as a double. The user and item IDs are ready literally as Strings and treated as such in the API. Note that this means that whitespace matters in the data file; they will be treated as part of the ID values.

This class will reload data from the data file when {@link #refresh(Collection)} is called, unlessthe file has been reloaded very recently already.

This class will also look for update "delta" files in the same directory, with file names that start the same way (up to the first period). These files should have the same format, and provide updated data that supersedes what is in the main data file. This is a mechanism that allows an application to push updates to {@link FileDataModel} without re-copying the entire data file.
The line may contain a blank preference value (e.g. "123,ABC,"). This is interpreted to mean "delete preference", and is only useful in the context of an update delta file (see above).

Finally, for application that have no notion of a preference value (that is, the user simply expresses a preference for an item, but no degree of preference), the caller can simply omit the third token in each line altogether -- for example, "123,ABC".

This class is not intended for use with very large amounts of data (over, say, tens of millions of rows). For that, a JDBC-backed {@link DataModel} and a database are more appropriate.
It is possible and likely useful to subclass this class and customize its behavior to accommodate application-specific needs and input formats. See {@link #processLine(String,Map,Map)}, {@link #buildItem(String)}, {@link #buildUser(String,List)}and {@link #buildPreference(User,Item,double)}.

    if( args.length > 0){
      INPUT_FILE = args[0];
    }
  
    UserSimilarity similarity;
    DataModel model = new FileDataModel(new File(INPUT_FILE));
    
    
    if( USE_LOG_LIKELIHOOD ){
      similarity =  new LogLikelihoodSimilarity(model);
    }
    else {
      similarity =  new TanimotoCoefficientSimilarity(model);
    }
    UserNeighborhood neighborhood = new NearestNUserNeighborhood(2, similarity, model);
    UserBasedRecommender recommender = new GenericUserBasedRecommender(model, neighborhood, similarity);
    
    int counter = 0;
    LongPrimitiveIterator users = model.getUserIDs();
    while (users.hasNext()) {
      long userID = users.nextLong();
      List<RecommendedItem> recommendations = recommender.recommend(userID, NUM_OF_RECOMMENDATIONS_RETURNED);
      for (RecommendedItem recommendation : recommendations) {

View Full Code Here


  private UserBaseRecommenderEvaluation() {}
  
  public static void main(String[] args) throws Exception {


  DataModel model = new FileDataModel(new File("ua.base.boolean-large.csv"));


  RecommenderBuilder builder = new RecommenderBuilder() {
    @Override
    public Recommender buildRecommender(DataModel model) throws TasteException {
      UserSimilarity similarity = new LogLikelihoodSimilarity(model);

View Full Code Here


  private ItemBaseRecommenderEvaluation() {}
  
  public static void main(String[] args) throws Exception {


  DataModel model = new FileDataModel(new File("ua.base.boolean.csv"));


  RecommenderBuilder builder = new RecommenderBuilder() {
    @Override
    public Recommender buildRecommender(DataModel model) throws TasteException {
      return new GenericBooleanPrefItemBasedRecommender(model, new LogLikelihoodSimilarity(model));

View Full Code Here

      INPUT_FILE = args[0];
      OUTPUT_FILE = args[1];
    }
  
    ItemSimilarity similarity;
    DataModel model = new FileDataModel(new File(INPUT_FILE));
    BufferedWriter out = new BufferedWriter(new FileWriter(OUTPUT_FILE));


  
  
    if( USE_LOG_LIKELIHOOD ){
      similarity =  new LogLikelihoodSimilarity(model);
    }
    else {
      similarity =  new TanimotoCoefficientSimilarity(model);
    }


    Recommender recommender = new GenericBooleanPrefItemBasedRecommender(model, similarity);
    
    int counter = 0;
    LongPrimitiveIterator users = model.getUserIDs();
    while (users.hasNext()) {
      long userID = users.nextLong();
      List<RecommendedItem> recommendations = recommender.recommend(userID, NUM_OF_RECOMMENDATIONS_RETURNED);
      for (RecommendedItem recommendation : recommendations) {

View Full Code Here

0 1 2

TOP

Related Classes of org.apache.mahout.cf.taste.impl.model.file.FileDataModel$UserIteratorOverData

com.manning.hip.ch9.MovieUserEvaluator

com.manning.hip.ch9.MovieUserRecommender

mia.recommender.ch02.EvaluatorIntro

mia.recommender.ch02.IREvaluatorIntro

mia.recommender.ch02.ItemBaseRecommender

mia.recommender.ch02.ItemBaseRecommenderEvaluation

mia.recommender.ch02.UserBaseRecommender

mia.recommender.ch02.UserBaseRecommenderEvaluation

mia.recommender.ch03.IREvaluatorBooleanPrefIntro1

mia.recommender.ch03.IREvaluatorBooleanPrefIntro2

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.