/**
* Build a rating matrix from the rating data. Each user's ratings are first normalized by
* subtracting a baseline score (usually a mean).
*
* @param userMapping The index mapping of user IDs to column numbers.
* @param itemMapping The index mapping of item IDs to row numbers.
* @return A matrix storing the <i>normalized</i> user ratings.
*/
private RealMatrix createRatingMatrix(IdIndexMapping userMapping, IdIndexMapping itemMapping) {
final int nusers = userMapping.size();
final int nitems = itemMapping.size();
// Create a matrix with users on rows and items on columns
logger.info("creating {} by {} rating matrix", nusers, nitems);
RealMatrix matrix = MatrixUtils.createRealMatrix(nusers, nitems);
// populate it with data
Cursor<UserHistory<Event>> users = userEventDAO.streamEventsByUser();
try {
for (UserHistory<Event> user : users) {
// Get the row number for this user
int u = userMapping.getIndex(user.getUserId());
MutableSparseVector ratings = Ratings.userRatingVector(user.filter(Rating.class));
MutableSparseVector baselines = MutableSparseVector.create(ratings.keySet());
baselineScorer.score(user.getUserId(), baselines);
// TODO Populate this user's row with their ratings, minus the baseline scores
for (VectorEntry entry : ratings.fast(State.SET)) {
long itemid = entry.getKey();
int i = itemMapping.getIndex(itemid);
double rating = entry.getValue();
double baseline = baselines.get(itemid);
matrix.setEntry(u, i, rating - baseline);
}
}
} finally {
users.close();
}
return matrix;
}
/**
* Build the SVD model.
*
* @return A singular value decomposition recommender model.
*/
@Override
public SVDModel get() {
// Create index mappings of user and item IDs.
// You can use these to find row and columns in the matrix based on user/item IDs.
IdIndexMapping userMapping = IdIndexMapping.create(userDAO.getUserIds());
logger.debug("indexed {} users", userMapping.size());
IdIndexMapping itemMapping = IdIndexMapping.create(itemDAO.getItemIds());
logger.debug("indexed {} items", itemMapping.size());
// We have to do 2 things:
// First, prepare a matrix containing the rating data.
RealMatrix matrix = createRatingMatrix(userMapping, itemMapping);
// Second, compute its factorization
// All the work is done in the constructor
SingularValueDecomposition svd = new SingularValueDecomposition(matrix);
// Third, truncate the decomposed matrix
// TODO Truncate the matrices and construct the SVD model
RealMatrix userMatrix = svd.getU();
RealMatrix weights = svd.getS();
RealMatrix itemMatrix = svd.getV();
userMatrix = userMatrix.getSubMatrix(0, userMatrix.getRowDimension() - 1, 0, featureCount - 1);
weights = weights.getSubMatrix(0, featureCount - 1, 0, featureCount - 1);
itemMatrix = itemMatrix.getSubMatrix(0, itemMatrix.getRowDimension() - 1, 0, featureCount - 1);
return new SVDModel(userMapping, itemMapping, userMatrix, itemMatrix, weights);
}