US 2020/0265218 A1 · SEMI-SUPERVISED HYBRID CLUSTERING/CLASSIFICATION SYSTEM

Abstract

System and method for classifying data objects occurring in an unstructured dataset, comprising: extracting feature vectors from the unstructured dataset, each feature vector representing an occurrence of a data object in the unstructured dataset; classifying the feature vectors into feature vector sets that each correspond to a respective object class from a plurality of object classes; for each feature vector set: performing multiple iterations of a clustering operation, each iteration including clustering feature vectors from the feature vector set into clusters of similar feature vectors and identifying outlier feature vectors, wherein for at least one iteration after a first iteration of the clustering operation, outlier feature vectors identified in a previous iteration are excluded from the clustering operation; and outputting a key cluster for the feature vector set from a final iteration of the multiple iterations, the key cluster including a greater number of similar feature vectors than any of the other clusters of the final iteration; and assembling a dataset that includes at least the feature vectors from the key clusters of the feature vector sets.

Claims20

Dependent claims

A computer implemented method for classifying data objects occurring in an unstructured dataset, comprising:

extracting feature vectors from the unstructured data, each feature vector representing an occurrence of a data object in the unstructured data;
classifying the feature vectors into feature vector sets that each correspond to a respective object class from a plurality of object classes;
for each feature vector set:
performing multiple iterations of a clustering operation, each iteration including clustering feature vectors from the feature vector set into clusters of similar feature vectors and identifying outlier feature vectors, wherein for at least one iteration after a first iteration of the clustering operation, outlier feature vectors identified in a previous iteration are excluded from the clustering operation;
and
outputting a key cluster for the feature vector set from a final iteration of the multiple iterations, the key cluster including a greater number of similar feature vectors than any of the other clusters of the final iteration;
and
assembling a dataset that includes at least the feature vectors from the key clusters of the feature vector sets.

5 more elements

A system for classifying data objects occurring in an unstructured dataset, the system including:

a classification system configured to:
extract feature vectors from the unstructured data, each feature vector representing an occurrence of a data object in the unstructured data;
and
classify the feature vectors into feature vector sets that each correspond to a respective object class from a plurality of object classes;
a clustering system configured to, for each feature vector set:
perform multiple iterations of a clustering operation on the feature vector set, each iteration including clustering feature vectors from the feature vector set into clusters of similar feature vectors and identifying outlier feature vectors, wherein for at least one iteration after a first iteration of the clustering operation, outlier feature vectors identified in a previous iteration are excluded from the clustering operation;
and
output a key cluster for the feature vector set from a final iteration of the multiple iterations, the key cluster including a greater number of similar feature vectors than any of the other clusters of the final iteration;
and
a dataset assembly module configured to assemble a dataset that includes at least the feature vectors from the key clusters of the feature vector sets.

7 more elements

Dependencies are inferred from the claim text.

Semi-supervised hybrid clustering/classification system

Life of this patent

Semi-supervised hybrid clustering/classification system

Life of this patent

Abstract

Claims20