Patent · US Active

Systems, methods, and media for outputting data based upon anomaly detection

US9003523B2 · kind B2 · utility

3Cited by

7References

12Claims

0Family size

Assignee

The Trustees of Columbia University in the City of New York · US

Inventors

Salvatore J. Stolfo · Ridgewood, US
Ke Wang · New York, US
Janak Parekh · Manhasset, US

Key dates

Filing date	May 9, 2013
Grant date	Apr 7, 2015
Priority date	—
Expiry date	May 9, 2033

Classification

Technology area (CPC G)Physics
CPC primaryG06F2221/034
WIPO fieldDigital communication
WIPO sectorElectrical engineering

Abstract

Systems, methods, and media for outputting data based on anomaly detection are provided. In some embodiments, a method for outputting data based on anomaly detection is provided, the method comprising: receiving, using a hardware processor, an input dataset; identifying grams in the input dataset that substantially include distinct byte values; creating an input subset by removing the identified grams from the input dataset; determining whether the input dataset is likely to be anomalous based on the identified grams, and determining whether the input dataset is likely to be anomalous by applying the input subset to a binary anomaly detection model to check for an n-gram in the input subset; and outputting the input dataset based on the likelihood that the input dataset is anomalous.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.