Browse Prior Art Database

Method And System For Alert Prediction

IP.com Disclosure Number: IPCOM000196839D
Publication Date: 2010-Jun-18
Document File: 5 page(s) / 87K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system for predicting alerts based on a set of configurations and data associated with previously generated alerts is disclosed.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 43% of the total text.

Page 1 of 5

Method And System For Alert Prediction

Disclosed is a method and system for predicting alerts. The method predicts alerts based on a set of configurations and data associated with previously generated alerts,

which increases the likelihood that the alert prediction is a valid precursor to an actual

alert being generated. For example, the method can predict new alerts for specific rows in a database table by examining the patterns of previously generated alerts for the specific rows. The row values, individual column of row values, and whole column of row values that generated the previous alerts are examined and rated to predict new alerts.

The method uses a database to store system data, including the data that defines alerts and types of alert.

APV) is

added to each database table to rate the importance of each row value. In addition, a first new table is created to store the column of row values and APVs associated with the column of row values. The information that leads to the actual generated alert is given an APV based on the significance it had on generating the alert. These values may vary depending on the system used. The information may have a hierarchy. In the hierarchy, individual column of row values, the column values and combined row values each have an APV.

Additionally, each APV may reflect incremental values or

A numbered value equal to a percentage of alerts where a specific attribute data

column value was part of an entity that contributed to the triggering of an alert. The

APV may be manually updated by the user.

A system generated sequential APV identification number.

An APV assigned to the entity attribute value

An APV assigned to the entity attribute type of the attribute data.

A type value that determines the form of the numbered APV. The valid types of

APV are:

a.

b.

A generation type value that indicates whether a current APV is generated by the

system or the user.

A backup value associated with the last APV, if the generation type value was

A new column containing an Alert Prediction Values

(

percentages.

Accordingly, an APV may consist of following values:

1.

2.

3.

(the attribute data).

4.

5.

Percentage = (

specific attribute data column value contributed to the triggering of the alert).

An Incremental Value.

An incremental value) / (Total number of alerts where a

6.

7.

1

Page 2 of 5

changed. These values are automatically updated when the standard values are updated. This facilitates to keep a valid history in case the user chooses to revert to the previous generation type values.

Further, the method also creates a second new table to store Alert Prediction Rules (

APRs).

An APR is a configurable set of parameters. The set of parameters comprises

alerts, events, row values, and column of row values and their APVs based on previously generated alerts that allows for alert prediction.

An APV is a configured set

of APV values that must be met in order to predict an alert. When an alert is generated, all...