All Classes

Class	Description
AbstractStableMarriage<T>
AbstractStableMarriage.AbstractMatcher
AbstractStableMarriage.Matcher
AggregatingSimilarityMeasure<T>	Aggregates similarities with a given aggregator.
AggregatingSimilarityMeasure.AggregatingSimilarityMeasureBuilder<T>
AnnotatedValue<T>	A value with lineage information; in particular, the Source and the timestamp.
Assigner<T>	Implements an algorithm that solves an assignment problem.
BipartiteMatcher<T>	Implements an algorithm that finds a matching in a bipartite graph.
CachingSimilarity<T,I>	A light-weight caching layer over SimilarityMeasure to allow implementations to repeatedly calculate expensive similarities without implementing a cache on their own.
Candidate<T>	Represents a candidate pair that was generated with an OfflineCandidateSelection.
CandidateSelection<T>	Selects candidates from a static or dynamic dataset accessible through Iterables.
Classification	Contains the possible classification classes.
ClassificationException	An exception thrown by a Classifier whenever an exception during classification occurred.
ClassificationResult	The classification of a Candidate with additional information.
ClassificationResult.ClassificationResultBuilder
ClassifiedCandidate<T>	Contains a ClassificationResult with the corresponding Candidate.
ClassifiedCandidate.ClassifiedCandidateBuilder<T>
Classifier<T>	Classifies a OnlineCandidate as duplicate, non-duplicate, or other Classifications.
Cluster<C extends java.lang.Comparable<C>,T>	A cluster is a coherent collection of duplicate records.
Cluster.ClusterBuilder<C extends java.lang.Comparable<C>,T>
ClusterIdGenerators	A collection of typical cluster id generators.
Clustering<C extends java.lang.Comparable<C>,T>	A clustering algorithm takes a list of ClassifiedCandidates and creates a coherent Cluster, such that all pairs of records inside the cluster are duplicate and no record outside the cluster is a duplicate with any record inside the cluster.
Clusters	Utility methods for Cluster
ClusterSplitHandler<C extends java.lang.Comparable<C>,T>	A callback that is invoked when an already existing cluster is split up during the Clustering of (online) deduplication.
CollectionSimilarityMeasure<C extends java.util.Collection<? extends E>,E>	A SimilarityMeasure that is defined over Collections.
CommonConflictResolutions	Provides factory methods for common conflict resolutions.
CommonSimilarityMeasures	A utility class that offers factory methods for common similarity measures.
CommonTransformations	A utility class that offers factory methods for common similarity transformations.
CompositeValue<T1 extends java.lang.Comparable<T1>,T2 extends java.lang.Comparable<T2>>	Helper for composed (sorting) keys, which will perform position-wise comparison of the key elements.
ConflictResolution<I,O>	Solves a conflict during resolution - two or more different values that need to be merged into a single value for the fused representation.
ConflictResolutionFusion<R>	A fusion approach based on conflict resolution.
ConflictResolutionFusion.ConflictResolutionFusionBuilder<R>
ConsistentClustering<C extends java.lang.Comparable<C>,T,I extends java.lang.Comparable<? super I>>	Wraps another clustering and keeps clusters together, when the wrapped clustering would split it. Example: consider a stable marriage-based clustering where A1-B have been previously matched and subsequently clustered.
ConsistentClustering.ConsistentClusteringBuilder<C extends java.lang.Comparable<C>,T,I extends java.lang.Comparable<? super I>>
CosineSimilarityMeasure<C extends java.util.Collection<? extends E>,E>	Calculates the cosine similarity measure over two bags of elements.
CutoffSimiliarityMeasure<T>	Cuts the similarity returned by this similarity, such that all values <threshold result in a similarity of 0, and all values [threshold, 1] are left untouched.
Deduplication<T>	A full deduplication process, which ensures that no duplicate record is emitted.
DuplicateDetection<C extends java.lang.Comparable<C>,T>	A duplicate detection algorithm processes a dataset of records and returns the distinct Clusters.
ExceptionContext	The exception context allows the safe execution of code that may throw an exception.
FunctionalClass<T>	A wrapper around Class that can be used to extract callable lambdas to methods and fields.
FunctionalConstructor<T>	A lambda wrapper around the no-arg constructor of a class.
FunctionalMethod<T>	A lambda wrapper around the a method of a class.
FunctionalProperty<T,F>	A lambda wrapper around the a property of a class.
FusedValue<T>	A fused value with contextual information.
FusingOnlineDuplicateDetection<C extends java.lang.Comparable<C>,T>	A full online deduplication process, which Retrieves duplicate clusters through OnlineDeduplication. Fuses the duplicate clusters into reconciled records.
FusingOnlineDuplicateDetection.FusingOnlineDuplicateDetectionBuilder<C extends java.lang.Comparable<C>,T>
Fusion<T>	Fuses a cluster of duplicates to one representation.
FusionContext	A fusion context captures exceptions in an ExceptionContext during resolution and provides additional context values.
FusionContext.FusionContextBuilder
FusionException	An exception thrown by a ConflictResolution whenever an exception during fusion occurs.
IncompleteFusionHandler<T>	A callback that allows incomplete fusions to be treated either with direct repair algorithms or through side-channel means (a.k.a ignore for now and repair asynchronously with the help of domain experts).
Levenshtein<T extends java.lang.CharSequence>	Provides the Levenshtein similarity calculation, which calculates the number of insertions, deletions, and replacements needed to transform one string into another.
MatchingSimilarity<C extends java.util.Collection<? extends E>,E>	A similarity measure that finds the best matches between two bags of entities and calculates an overall similarity by summing the similarity of these matches and normalize it over the (max) number of elements per collection.
Merge<R>	A nested conflict resolution for complex types.
Merge.AdditionalFieldMergeBuilder<F,R>
Merge.FieldMergeBuilder<F,R>
Merge.IllTypedFieldMergeBuilder<I,F,R>
Merge.MergeBuilder<R>
OfflineCandidate<T>	Represents a candidate pair that was generated with an OfflineCandidateSelection.
OfflineCandidateSelection<T>	Selects candidates from a static dataset accessible through Iterables.
OfflineDeduplication<T>	A full offline deduplication process, which ensures that no duplicate record is in the results.
OfflineDuplicateDetection<C extends java.lang.Comparable<C>,T>	The offline duplicate detection returns all duplicate records within a dataset.
OnlineCandidate<T>	Represents a candidate pair that was generated with an OnlineCandidateSelection.
OnlineCandidateSelection<T>	Selects candidates from a streaming dataset by processing the incoming elements one-at-a-time.
OnlineDeduplication<T>	A full online deduplication process, which ensures that no duplicate record is emitted.
OnlineDuplicateDetection<C extends java.lang.Comparable<C>,T>	An online duplicate detection algorithm processes a stream of records and returns all changed Clusters for each record.
OnlinePairBasedDuplicateDetection<C extends java.lang.Comparable<C>,T>	A pair-based duplicate detection algorithm, which Performs OnlineCandidateSelection Applies a Classifier to the found Candidates Transforms the found duplicate pairs with a Clustering into Clusterss
OnlinePairBasedDuplicateDetection.OnlinePairBasedDuplicateDetectionBuilder<C extends java.lang.Comparable<C>,T>
OnlineSortedNeighborhoodMethod<T>	A sorted neighborhood method (SNM) for online deduplication.
OnlineSortedNeighborhoodMethod.OnlineSortedNeighborhoodMethodBuilder<T>
OnlineSortedNeighborhoodMethod.Pass<T,K extends java.lang.Comparable<K>>	Represents a pass over the dataset with a specific sorting key and window size.
OracleClassifier<T>	A classifier that knows the results perfectly as it receives the gold standard during creation.
OracleClustering<C extends java.lang.Comparable<C>,T,I>	A clustering that knows the results perfectly as it receives the gold standard during creation.
PossibleDuplicateHandler<T>	A callback that is invoked when a Classification.POSSIBLE_DUPLICATE has been founds during DuplicateDetection.
RefineCluster<C extends java.lang.Comparable<C>,T>	Splits large clusters into smaller clusters when the inter-cluster similarities are sub-optimal.
RefineCluster.RefineClusterBuilder<C extends java.lang.Comparable<C>,T>
RefinedTransitiveClosure<C extends java.lang.Comparable<C>,T,I extends java.lang.Comparable<? super I>>	Executes TransitiveClosure and successively RefineCluster.
RefinedTransitiveClosure.RefinedTransitiveClosureBuilder<C extends java.lang.Comparable<C>,T,I extends java.lang.Comparable<? super I>>
ResolutionTag<T>	A resolution tag is a way to refer to a previously, resolved value in the current FusionContext.
RuleBasedClassifier<T>	Successively applies a list of rules to the record and returns the respective ClassificationResult with the following cases: If any rule classifies the pair unambiguously as Classification.DUPLICATE or Classification.NON_DUPLICATE, the classification is immediately returned. If no rule can be applied, the classification is RuleBasedClassifier.defaultClassificationResult. There are three types of rules: Negatives rule are used to exclude false positives.
RuleBasedClassifier.Rule<T>	A rule has a name for lineage/debugging and the similarity measure.
RuleBasedClassifier.RuleBasedClassifierBuilder<T>	A builder for RuleBasedClassifier with convenience methods to create positive and negative rules.
SetSimilarityMeasure<C extends java.util.Collection<? extends E>,E>	A SimilarityMeasure that is defined over Collections that are treated as sets.
SimilarityContext	A similarity context captures exceptions in an ExceptionContext and provides additional configurations that represent cross-cutting concerns, such as null handling.
SimilarityContext.SimilarityContextBuilder
SimilarityMeasure<T>	A SimilarityMeasure compares two values and calculates a similarity in [-1; 1], where 0 means no similarity and 1 equal values in a given context.
SortingKey<T,K extends java.lang.Comparable<K>>	A sorting key allows a dataset to be indexed by a specific (calculated) value of a record, such that duplicates have a higher probability of being in close proximity and thus a CandidateSelection may prune the search space drastically.
Source	The source of a value to be fused.
StreamUtil	Adds some functions that are missing in Java Streams.
StronglyStableMarriage<T>	Implements a strongly stable matching based on the stable marriage with indifference.
TerminalConflictResolution<I,O>	A conflict resolution function that is guaranteed to produce a single value.
TransformingSimilarityMeasure<R,T>	Applies a SimilarityMeasure to transformed input values.
TransitiveClosure<C extends java.lang.Comparable<C>,T,I extends java.lang.Comparable<? super I>>	An amortized linear transitive closure implementation over the number of pairs.
TransitiveClosure.TransitiveClosureBuilder<C extends java.lang.Comparable<C>,T,I extends java.lang.Comparable<? super I>>
ValueTransformation<T,R>	Performs a transformation on a value, especially on input values for SimilarityMeasure.
WeaklyStableMarriage<T>	Implements a weakly stable matching based on the stable marriage with indifference (i.e., ties).
WeightedAggregatingSimilarityMeasure<R>
WeightedAggregatingSimilarityMeasure.WeightedAggregatingSimilarityMeasureBuilder<R>
WeightedAggregatingSimilarityMeasure.WeightedSimilarity<T>
WeightedAggregatingSimilarityMeasure.WeightedValue
WeightedEdge<T>	A weighted (directed) edge between two records.