View source for US Patent Application 17737065. ANOMALY DETECTION AND ANOMALOUS PATTERNS IDENTIFICATION simplified abstract

=ANOMALY DETECTION AND ANOMALOUS PATTERNS IDENTIFICATION=


==Organization Name==

[[INTERNATIONAL BUSINESS MACHINES CORPORATION]]


==Inventor(s)==

[[Xi Yang of Apex NC (US)]]

[[Larisa Shwartz of Greenwich CT (US)]]

[[Ruchi Mahindru of Elmsford NY (US)]]

[[Ian Manning of Church Hill (IE)]]

[[Ruchir Puri of Baldwin Place NY (US)]]

[[MUDHAKAR Srivatsa of White Plains NY (US)]]

==ANOMALY DETECTION AND ANOMALOUS PATTERNS IDENTIFICATION - A simplified explanation of the abstract==


This abstract first appeared for US patent application 17737065 titled 'ANOMALY DETECTION AND ANOMALOUS PATTERNS IDENTIFICATION

==Simplified Explanation==

The patent application describes a method for detecting anomalies and identifying anomalous patterns using a GMM-LASSO algorithm.
* GMM-LASSO is a selection operator-type algorithm that uses generalized method of moments (GMM) estimation.
* The method involves a feedback loop where an anomalous window is detected and then used to identify anomalous patterns.
* The approach can be applied to classify sequential data.
* It generates vectors based on the sequential data.
* These vectors are then clustered into clusters.
* The method determines the membership of the vectors in the clusters.
* The clusters are updated and optimized with respect to a predefined threshold.


==Original Abstract Submitted==

An approach for end-to-end anomaly detection and anomalous patterns identification is disclosed. The approach leverages the use of a GMM-LASSO (a selection operator-type, Lasso-type, generalized method of moments (GMM) estimator) algorithm and proposes a feedback loop where the window (i.e., anomalous window) is detected and then it is used to detect the anomalous patterns. For example, the approach can classify one or more sequential data; generates one or more vectors based on the one or more sequential data; clusters the one or more vectors into one or more clusters; determines a membership of the one or more vectors associated with the one or more clusters; updates the one or more clusters; and optimizes the one or more clusters with respect to a predefined threshold.