Search in sources :

Example 11 with ArrayDBIDs

use of de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs in project elki by elki-project.

the class ABOD method run.

/**
 * Run ABOD on the data set.
 *
 * @param relation Relation to process
 * @return Outlier detection result
 */
public OutlierResult run(Database db, Relation<V> relation) {
    ArrayDBIDs ids = DBIDUtil.ensureArray(relation.getDBIDs());
    // Build a kernel matrix, to make O(n^3) slightly less bad.
    SimilarityQuery<V> sq = db.getSimilarityQuery(relation, kernelFunction);
    KernelMatrix kernelMatrix = new KernelMatrix(sq, relation, ids);
    WritableDoubleDataStore abodvalues = DataStoreUtil.makeDoubleStorage(ids, DataStoreFactory.HINT_STATIC);
    DoubleMinMax minmaxabod = new DoubleMinMax();
    MeanVariance s = new MeanVariance();
    DBIDArrayIter pA = ids.iter(), pB = ids.iter(), pC = ids.iter();
    for (; pA.valid(); pA.advance()) {
        final double abof = computeABOF(kernelMatrix, pA, pB, pC, s);
        minmaxabod.put(abof);
        abodvalues.putDouble(pA, abof);
    }
    // Build result representation.
    DoubleRelation scoreResult = new MaterializedDoubleRelation("Angle-Based Outlier Degree", "abod-outlier", abodvalues, relation.getDBIDs());
    OutlierScoreMeta scoreMeta = new InvertedOutlierScoreMeta(minmaxabod.getMin(), minmaxabod.getMax(), 0.0, Double.POSITIVE_INFINITY);
    return new OutlierResult(scoreMeta, scoreResult);
}
Also used : WritableDoubleDataStore(de.lmu.ifi.dbs.elki.database.datastore.WritableDoubleDataStore) OutlierResult(de.lmu.ifi.dbs.elki.result.outlier.OutlierResult) DBIDArrayIter(de.lmu.ifi.dbs.elki.database.ids.DBIDArrayIter) InvertedOutlierScoreMeta(de.lmu.ifi.dbs.elki.result.outlier.InvertedOutlierScoreMeta) DoubleRelation(de.lmu.ifi.dbs.elki.database.relation.DoubleRelation) MaterializedDoubleRelation(de.lmu.ifi.dbs.elki.database.relation.MaterializedDoubleRelation) OutlierScoreMeta(de.lmu.ifi.dbs.elki.result.outlier.OutlierScoreMeta) InvertedOutlierScoreMeta(de.lmu.ifi.dbs.elki.result.outlier.InvertedOutlierScoreMeta) KernelMatrix(de.lmu.ifi.dbs.elki.distance.similarityfunction.kernel.KernelMatrix) MeanVariance(de.lmu.ifi.dbs.elki.math.MeanVariance) DoubleMinMax(de.lmu.ifi.dbs.elki.math.DoubleMinMax) ArrayDBIDs(de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs) MaterializedDoubleRelation(de.lmu.ifi.dbs.elki.database.relation.MaterializedDoubleRelation)

Example 12 with ArrayDBIDs

use of de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs in project elki by elki-project.

the class LBABOD method run.

/**
 * Run LB-ABOD on the data set.
 *
 * @param relation Relation to process
 * @return Outlier detection result
 */
@Override
public OutlierResult run(Database db, Relation<V> relation) {
    ArrayDBIDs ids = DBIDUtil.ensureArray(relation.getDBIDs());
    DBIDArrayIter pB = ids.iter(), pC = ids.iter();
    SimilarityQuery<V> sq = db.getSimilarityQuery(relation, kernelFunction);
    KernelMatrix kernelMatrix = new KernelMatrix(sq, relation, ids);
    // Output storage.
    WritableDoubleDataStore abodvalues = DataStoreUtil.makeDoubleStorage(ids, DataStoreFactory.HINT_STATIC);
    DoubleMinMax minmaxabod = new DoubleMinMax();
    double max = 0.;
    // Storage for squared distances (will be reused!)
    WritableDoubleDataStore sqDists = DataStoreUtil.makeDoubleStorage(ids, DataStoreFactory.HINT_TEMP | DataStoreFactory.HINT_HOT);
    // Nearest neighbor heap (will be reused!)
    KNNHeap nn = DBIDUtil.newHeap(k);
    // Priority queue for candidates
    ModifiableDoubleDBIDList candidates = DBIDUtil.newDistanceDBIDList(relation.size());
    // get Candidate Ranking
    for (DBIDIter pA = relation.iterDBIDs(); pA.valid(); pA.advance()) {
        // Compute nearest neighbors and distances.
        nn.clear();
        double simAA = kernelMatrix.getSimilarity(pA, pA);
        // Sum of 1./(|AB|) and 1./(|AB|^2); for computing R2.
        double sumid = 0., sumisqd = 0.;
        for (pB.seek(0); pB.valid(); pB.advance()) {
            if (DBIDUtil.equal(pB, pA)) {
                continue;
            }
            double simBB = kernelMatrix.getSimilarity(pB, pB);
            double simAB = kernelMatrix.getSimilarity(pA, pB);
            double sqdAB = simAA + simBB - simAB - simAB;
            sqDists.putDouble(pB, sqdAB);
            final double isqdAB = 1. / sqdAB;
            sumid += FastMath.sqrt(isqdAB);
            sumisqd += isqdAB;
            // Update heap
            nn.insert(sqdAB, pB);
        }
        // Compute FastABOD approximation, adjust for lower bound.
        // LB-ABOF is defined via a numerically unstable formula.
        // Variance as E(X^2)-E(X)^2 suffers from catastrophic cancellation!
        // TODO: ensure numerical precision!
        double nnsum = 0., nnsumsq = 0., nnsumisqd = 0.;
        KNNList nl = nn.toKNNList();
        DoubleDBIDListIter iB = nl.iter(), iC = nl.iter();
        for (; iB.valid(); iB.advance()) {
            double sqdAB = iB.doubleValue();
            double simAB = kernelMatrix.getSimilarity(pA, iB);
            if (!(sqdAB > 0.)) {
                continue;
            }
            for (iC.seek(iB.getOffset() + 1); iC.valid(); iC.advance()) {
                double sqdAC = iC.doubleValue();
                double simAC = kernelMatrix.getSimilarity(pA, iC);
                if (!(sqdAC > 0.)) {
                    continue;
                }
                // Exploit bilinearity of scalar product:
                // <B-A, C-A> = <B, C-A> - <A,C-A>
                // = <B,C> - <B,A> - <A,C> + <A,A>
                double simBC = kernelMatrix.getSimilarity(iB, iC);
                double numerator = simBC - simAB - simAC + simAA;
                double sqweight = 1. / (sqdAB * sqdAC);
                double weight = FastMath.sqrt(sqweight);
                double val = numerator * sqweight;
                nnsum += val * weight;
                nnsumsq += val * val * weight;
                nnsumisqd += sqweight;
            }
        }
        // Remaining weight, term R2:
        double r2 = sumisqd * sumisqd - 2. * nnsumisqd;
        double tmp = (2. * nnsum + r2) / (sumid * sumid);
        double lbabof = 2. * nnsumsq / (sumid * sumid) - tmp * tmp;
        // Track maximum?
        if (lbabof > max) {
            max = lbabof;
        }
        abodvalues.putDouble(pA, lbabof);
        candidates.add(lbabof, pA);
    }
    // Put maximum from approximate values.
    minmaxabod.put(max);
    candidates.sort();
    // refine Candidates
    int refinements = 0;
    DoubleMinHeap topscores = new DoubleMinHeap(l);
    MeanVariance s = new MeanVariance();
    for (DoubleDBIDListIter pA = candidates.iter(); pA.valid(); pA.advance()) {
        // Stop refining
        if (topscores.size() >= k && pA.doubleValue() > topscores.peek()) {
            break;
        }
        final double abof = computeABOF(kernelMatrix, pA, pB, pC, s);
        // Store refined score:
        abodvalues.putDouble(pA, abof);
        minmaxabod.put(abof);
        // Update the heap tracking the top scores.
        if (topscores.size() < k) {
            topscores.add(abof);
        } else {
            if (topscores.peek() > abof) {
                topscores.replaceTopElement(abof);
            }
        }
        refinements += 1;
    }
    if (LOG.isStatistics()) {
        LoggingConfiguration.setVerbose(Level.VERYVERBOSE);
        LOG.statistics(new LongStatistic("lb-abod.refinements", refinements));
    }
    // Build result representation.
    DoubleRelation scoreResult = new MaterializedDoubleRelation("Angle-based Outlier Detection", "abod-outlier", abodvalues, ids);
    OutlierScoreMeta scoreMeta = new InvertedOutlierScoreMeta(minmaxabod.getMin(), minmaxabod.getMax(), 0.0, Double.POSITIVE_INFINITY);
    return new OutlierResult(scoreMeta, scoreResult);
}
Also used : DoubleDBIDListIter(de.lmu.ifi.dbs.elki.database.ids.DoubleDBIDListIter) DoubleMinHeap(de.lmu.ifi.dbs.elki.utilities.datastructures.heap.DoubleMinHeap) WritableDoubleDataStore(de.lmu.ifi.dbs.elki.database.datastore.WritableDoubleDataStore) ModifiableDoubleDBIDList(de.lmu.ifi.dbs.elki.database.ids.ModifiableDoubleDBIDList) OutlierResult(de.lmu.ifi.dbs.elki.result.outlier.OutlierResult) DBIDArrayIter(de.lmu.ifi.dbs.elki.database.ids.DBIDArrayIter) InvertedOutlierScoreMeta(de.lmu.ifi.dbs.elki.result.outlier.InvertedOutlierScoreMeta) KNNHeap(de.lmu.ifi.dbs.elki.database.ids.KNNHeap) DoubleRelation(de.lmu.ifi.dbs.elki.database.relation.DoubleRelation) MaterializedDoubleRelation(de.lmu.ifi.dbs.elki.database.relation.MaterializedDoubleRelation) OutlierScoreMeta(de.lmu.ifi.dbs.elki.result.outlier.OutlierScoreMeta) InvertedOutlierScoreMeta(de.lmu.ifi.dbs.elki.result.outlier.InvertedOutlierScoreMeta) DBIDIter(de.lmu.ifi.dbs.elki.database.ids.DBIDIter) KernelMatrix(de.lmu.ifi.dbs.elki.distance.similarityfunction.kernel.KernelMatrix) MeanVariance(de.lmu.ifi.dbs.elki.math.MeanVariance) DoubleMinMax(de.lmu.ifi.dbs.elki.math.DoubleMinMax) KNNList(de.lmu.ifi.dbs.elki.database.ids.KNNList) LongStatistic(de.lmu.ifi.dbs.elki.logging.statistics.LongStatistic) ArrayDBIDs(de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs) MaterializedDoubleRelation(de.lmu.ifi.dbs.elki.database.relation.MaterializedDoubleRelation)

Example 13 with ArrayDBIDs

use of de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs in project elki by elki-project.

the class MaterializedKNNAndRKNNPreprocessorTest method testKNNQueries.

private void testKNNQueries(Relation<DoubleVector> rep, KNNQuery<DoubleVector> lin_knn_query, KNNQuery<DoubleVector> preproc_knn_query, int k) {
    ArrayDBIDs sample = DBIDUtil.ensureArray(rep.getDBIDs());
    List<? extends KNNList> lin_knn_ids = lin_knn_query.getKNNForBulkDBIDs(sample, k);
    List<? extends KNNList> preproc_knn_ids = preproc_knn_query.getKNNForBulkDBIDs(sample, k);
    for (int i = 0; i < rep.size(); i++) {
        KNNList lin_knn = lin_knn_ids.get(i);
        KNNList pre_knn = preproc_knn_ids.get(i);
        DoubleDBIDListIter lin = lin_knn.iter(), pre = pre_knn.iter();
        for (; lin.valid() && pre.valid(); lin.advance(), pre.advance(), i++) {
            assertTrue(DBIDUtil.equal(lin, pre) || lin.doubleValue() == pre.doubleValue());
        }
        assertEquals("kNN sizes do not agree.", lin_knn.size(), pre_knn.size());
        for (int j = 0; j < lin_knn.size(); j++) {
            assertTrue("kNNs of linear scan and preprocessor do not match!", DBIDUtil.equal(lin_knn.get(j), pre_knn.get(j)));
            assertEquals("kNNs of linear scan and preprocessor do not match!", lin_knn.get(j).doubleValue(), pre_knn.get(j).doubleValue(), 0.);
        }
    }
}
Also used : DoubleDBIDListIter(de.lmu.ifi.dbs.elki.database.ids.DoubleDBIDListIter) KNNList(de.lmu.ifi.dbs.elki.database.ids.KNNList) ArrayDBIDs(de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs)

Example 14 with ArrayDBIDs

use of de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs in project elki by elki-project.

the class MaterializedKNNAndRKNNPreprocessorTest method testRKNNQueries.

private void testRKNNQueries(Relation<DoubleVector> rep, RKNNQuery<DoubleVector> lin_rknn_query, RKNNQuery<DoubleVector> preproc_rknn_query, int k) {
    ArrayDBIDs sample = DBIDUtil.ensureArray(rep.getDBIDs());
    List<? extends DoubleDBIDList> lin_rknn_ids = lin_rknn_query.getRKNNForBulkDBIDs(sample, k);
    List<? extends DoubleDBIDList> preproc_rknn_ids = preproc_rknn_query.getRKNNForBulkDBIDs(sample, k);
    for (int i = 0; i < rep.size(); i++) {
        DoubleDBIDList lin_rknn = lin_rknn_ids.get(i);
        DoubleDBIDList pre_rknn = preproc_rknn_ids.get(i);
        DoubleDBIDListIter lin = lin_rknn.iter(), pre = pre_rknn.iter();
        for (; lin.valid() && pre.valid(); lin.advance(), pre.advance(), i++) {
            assertTrue(DBIDUtil.equal(lin, pre) || lin.doubleValue() == pre.doubleValue());
        }
        assertEquals("rkNN sizes do not agree for k=" + k, lin_rknn.size(), pre_rknn.size());
        for (int j = 0; j < lin_rknn.size(); j++) {
            assertTrue("rkNNs of linear scan and preprocessor do not match!", DBIDUtil.equal(lin_rknn.get(j), pre_rknn.get(j)));
            assertEquals("rkNNs of linear scan and preprocessor do not match!", lin_rknn.get(j).doubleValue(), pre_rknn.get(j).doubleValue(), 0.);
        }
    }
}
Also used : DoubleDBIDListIter(de.lmu.ifi.dbs.elki.database.ids.DoubleDBIDListIter) ArrayDBIDs(de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs) DoubleDBIDList(de.lmu.ifi.dbs.elki.database.ids.DoubleDBIDList)

Example 15 with ArrayDBIDs

use of de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs in project elki by elki-project.

the class SilhouetteOutlierDetection method run.

@Override
public OutlierResult run(Database database) {
    Relation<O> relation = database.getRelation(getDistanceFunction().getInputTypeRestriction());
    DistanceQuery<O> dq = database.getDistanceQuery(relation, getDistanceFunction());
    // TODO: improve ELKI api to ensure we're using the same DBIDs!
    Clustering<?> c = clusterer.run(database);
    WritableDoubleDataStore scores = DataStoreUtil.makeDoubleStorage(relation.getDBIDs(), DataStoreFactory.HINT_DB);
    DoubleMinMax mm = new DoubleMinMax();
    List<? extends Cluster<?>> clusters = c.getAllClusters();
    for (Cluster<?> cluster : clusters) {
        if (cluster.size() <= 1 || cluster.isNoise()) {
            switch(noiseOption) {
                case IGNORE_NOISE:
                case TREAT_NOISE_AS_SINGLETONS:
                    // As suggested in Rousseeuw, we use 0 for singletons.
                    for (DBIDIter iter = cluster.getIDs().iter(); iter.valid(); iter.advance()) {
                        scores.put(iter, 0.);
                    }
                    mm.put(0.);
                    continue;
                case MERGE_NOISE:
                    // Treat as cluster below
                    break;
            }
        }
        ArrayDBIDs ids = DBIDUtil.ensureArray(cluster.getIDs());
        // temporary storage.
        double[] as = new double[ids.size()];
        DBIDArrayIter it1 = ids.iter(), it2 = ids.iter();
        for (it1.seek(0); it1.valid(); it1.advance()) {
            // a: In-cluster distances
            // Already computed distances
            double a = as[it1.getOffset()];
            for (it2.seek(it1.getOffset() + 1); it2.valid(); it2.advance()) {
                final double dist = dq.distance(it1, it2);
                a += dist;
                as[it2.getOffset()] += dist;
            }
            a /= (ids.size() - 1);
            // b: other clusters:
            double min = Double.POSITIVE_INFINITY;
            for (Cluster<?> ocluster : clusters) {
                if (ocluster == /* yes, reference identity */
                cluster) {
                    continue;
                }
                if (ocluster.isNoise()) {
                    switch(noiseOption) {
                        case IGNORE_NOISE:
                            continue;
                        case MERGE_NOISE:
                            // No special treatment
                            break;
                        case TREAT_NOISE_AS_SINGLETONS:
                            // Treat noise cluster as singletons:
                            for (DBIDIter it3 = ocluster.getIDs().iter(); it3.valid(); it3.advance()) {
                                double dist = dq.distance(it1, it3);
                                if (dist < min) {
                                    min = dist;
                                }
                            }
                            continue;
                    }
                }
                final DBIDs oids = ocluster.getIDs();
                double b = 0.;
                for (DBIDIter it3 = oids.iter(); it3.valid(); it3.advance()) {
                    b += dq.distance(it1, it3);
                }
                b /= oids.size();
                if (b < min) {
                    min = b;
                }
            }
            final double score = (min - a) / Math.max(min, a);
            scores.put(it1, score);
            mm.put(score);
        }
    }
    // Build result representation.
    DoubleRelation scoreResult = new MaterializedDoubleRelation("Silhouette Coefficients", "silhouette-outlier", scores, relation.getDBIDs());
    OutlierScoreMeta scoreMeta = new InvertedOutlierScoreMeta(mm.getMin(), mm.getMax(), -1., 1., .5);
    return new OutlierResult(scoreMeta, scoreResult);
}
Also used : WritableDoubleDataStore(de.lmu.ifi.dbs.elki.database.datastore.WritableDoubleDataStore) ArrayDBIDs(de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs) DBIDs(de.lmu.ifi.dbs.elki.database.ids.DBIDs) OutlierResult(de.lmu.ifi.dbs.elki.result.outlier.OutlierResult) DBIDArrayIter(de.lmu.ifi.dbs.elki.database.ids.DBIDArrayIter) InvertedOutlierScoreMeta(de.lmu.ifi.dbs.elki.result.outlier.InvertedOutlierScoreMeta) DoubleRelation(de.lmu.ifi.dbs.elki.database.relation.DoubleRelation) MaterializedDoubleRelation(de.lmu.ifi.dbs.elki.database.relation.MaterializedDoubleRelation) OutlierScoreMeta(de.lmu.ifi.dbs.elki.result.outlier.OutlierScoreMeta) InvertedOutlierScoreMeta(de.lmu.ifi.dbs.elki.result.outlier.InvertedOutlierScoreMeta) DBIDIter(de.lmu.ifi.dbs.elki.database.ids.DBIDIter) DoubleMinMax(de.lmu.ifi.dbs.elki.math.DoubleMinMax) ArrayDBIDs(de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs) MaterializedDoubleRelation(de.lmu.ifi.dbs.elki.database.relation.MaterializedDoubleRelation)

Aggregations

ArrayDBIDs (de.lmu.ifi.dbs.elki.database.ids.ArrayDBIDs)45 DBIDIter (de.lmu.ifi.dbs.elki.database.ids.DBIDIter)23 DBIDs (de.lmu.ifi.dbs.elki.database.ids.DBIDs)16 DBIDArrayIter (de.lmu.ifi.dbs.elki.database.ids.DBIDArrayIter)14 FiniteProgress (de.lmu.ifi.dbs.elki.logging.progress.FiniteProgress)13 KNNList (de.lmu.ifi.dbs.elki.database.ids.KNNList)12 WritableDoubleDataStore (de.lmu.ifi.dbs.elki.database.datastore.WritableDoubleDataStore)10 ArrayModifiableDBIDs (de.lmu.ifi.dbs.elki.database.ids.ArrayModifiableDBIDs)9 DoubleDBIDListIter (de.lmu.ifi.dbs.elki.database.ids.DoubleDBIDListIter)9 ModifiableDBIDs (de.lmu.ifi.dbs.elki.database.ids.ModifiableDBIDs)8 DoubleRelation (de.lmu.ifi.dbs.elki.database.relation.DoubleRelation)8 MaterializedDoubleRelation (de.lmu.ifi.dbs.elki.database.relation.MaterializedDoubleRelation)7 DoubleMinMax (de.lmu.ifi.dbs.elki.math.DoubleMinMax)7 OutlierResult (de.lmu.ifi.dbs.elki.result.outlier.OutlierResult)7 OutlierScoreMeta (de.lmu.ifi.dbs.elki.result.outlier.OutlierScoreMeta)7 ArrayList (java.util.ArrayList)7 Clustering (de.lmu.ifi.dbs.elki.data.Clustering)6 KNNHeap (de.lmu.ifi.dbs.elki.database.ids.KNNHeap)6 ModifiableDoubleDBIDList (de.lmu.ifi.dbs.elki.database.ids.ModifiableDoubleDBIDList)6 LongStatistic (de.lmu.ifi.dbs.elki.logging.statistics.LongStatistic)5