EVENT-PROCESSING OPERATORS
First Claim
Patent Images
1. A method comprising:
- receiving a stream of events;
for each event in the received stream of events, determining whether the event is associated with one of a plurality of event clusters;
for each event determined to not be associated with any of the plurality of event clusters, placing an alert event into an output stream; and
periodically redefining the plurality of event clusters based on a subset of the events in the received stream of events.
1 Assignment
0 Petitions
Accused Products
Abstract
Novel event-processing operators are provided. These novel operators can be advantageously utilized in implementing event processors and event-processing systems. The novel event-processing operators include operators for event filtering by clustering, operators for event partitioning by classification, operators for event abstraction by hypothesis testing, and operators for event filtering by point estimation.
116 Citations
21 Claims
-
1. A method comprising:
-
receiving a stream of events; for each event in the received stream of events, determining whether the event is associated with one of a plurality of event clusters; for each event determined to not be associated with any of the plurality of event clusters, placing an alert event into an output stream; and periodically redefining the plurality of event clusters based on a subset of the events in the received stream of events. - View Dependent Claims (2, 3, 4, 5, 6)
-
-
7. A method comprising:
-
receiving a first stream of events; for each event in the received stream of events, determining whether the event has a probability equal to or greater than a predetermined threshold of being associated with one of a plurality of classifications in accordance with a generalized linear model comprising a predetermined distribution function, a linear predictor, and a predetermined link function; for each event determined to have a probability equal to or greater than the predetermined threshold of being associated with one of the plurality of classifications, placing the event in an output stream along with a mark indicating the associated classification; for each event determined to not have a probability equal to or greater than the predetermined threshold of being associated with one of the plurality of classifications, placing the event in an output stream along with a mark indicating that the event is not associated with one of the plurality of classifications; receiving a second stream of events, each event in the second stream being marked as being associated with one of the plurality of classifications; and based on events in the second stream, estimating unknown parameters in the linear predictor. - View Dependent Claims (8, 9, 10)
-
-
11. A method comprising:
-
receiving a stream of events; identifying a subset of events in the received stream; determining whether the subset of events fails a hypothesis about a statistical distribution of events in the received stream; placing an alert event into an output stream when the subset of events fails the hypothesis. - View Dependent Claims (12, 13, 14, 15, 16)
-
-
17. A method comprising:
-
receiving a stream of events; for each event in the received stream of events, evaluating a predicate comprising predetermined parameters; for each event for which the predicate evaluates to true, placing an alert event into an output stream; and estimating values for the predetermined parameters based on a subset of the events in the received stream. - View Dependent Claims (18, 19, 20, 21)
-
Specification