Producing a measure regarding cases associated with an issue after one or more events have occurred
First Claim
Patent Images
1. A method comprising:
- producing, based on output from a categorizer executed on one or more processors, a first measure regarding cases associated with an issue;
receiving information regarding additional cases associated with the issue after one or more events have occurred with respect to the issue;
after the one or more events have occurred, producing, based on further output from the categorizer executed on the one or more processors, a second measure regarding the additional cases associated with the issue; and
training the categorizer, wherein training the categorizer comprises;
receiving a query relating to the issue;
identifying cases in response to the query;
receiving user confirmation or disconfirmation that the identified cases belong to the issue; and
developing training cases for training the categorizer based on receiving the user confirmation or disconfirmation.
1 Assignment
0 Petitions
Accused Products
Abstract
A categorizer produces a first measure regarding cases associated with an issue. Information regarding additional cases associated with the issue is received after one or more events have occurred with respect to the issue. Based on further output from the categorizer, a second measure is produced regarding the additional cases associated with the issue.
66 Citations
36 Claims
-
1. A method comprising:
-
producing, based on output from a categorizer executed on one or more processors, a first measure regarding cases associated with an issue; receiving information regarding additional cases associated with the issue after one or more events have occurred with respect to the issue; after the one or more events have occurred, producing, based on further output from the categorizer executed on the one or more processors, a second measure regarding the additional cases associated with the issue; and training the categorizer, wherein training the categorizer comprises; receiving a query relating to the issue; identifying cases in response to the query; receiving user confirmation or disconfirmation that the identified cases belong to the issue; and developing training cases for training the categorizer based on receiving the user confirmation or disconfirmation. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 32, 33)
-
-
20. A system, comprising:
-
one or more processors; a categorizer executable on the one or more processors for categorizing cases in a data set into at least a first category, wherein the categorizer is a trained categorizer trained using a technique comprising; receiving a query relating to the first category; identifying cases in response to the query; receiving user confirmation or disconfirmation that the identified cases belong to the first category; and developing training cases for training the categorizer based on receiving the user confirmation or disconfirmation; a quantifier executable on the one or more processors in cooperation with the trained categorizer to compute a first quantification measure for a first group of cases, and a second quantification measure for a second group of cases, wherein the first group of cases relates to cases prior to occurrence of an event corresponding to an issue associated with the first category, and the second group of cases relates to cases after occurrence of the event; and a module executable on the one or more processors to determine an impact of the event on the issue based on comparing the first and second quantification measures. - View Dependent Claims (21, 34)
-
-
22. A computer-readable storage medium storing instructions that when executed cause a system having a processor to:
-
produce, based on output from a categorizer, a first measure regarding cases associated with an issue; receive information regarding additional cases associated with the issue after one or more events have occurred with respect to the issue; produce, after the one or more events have occurred and based on further output from the categorizer, a second measure regarding the additional cases associated with the issue; and train the categorizer, wherein training the categorizer comprises; receiving a query relating to the issue; identifying cases in response to the query; receiving user confirmation or disconfirmation that the identified cases belong to the issue; and developing training cases for training the categorizer based on receiving the user confirmation or disconfirmation. - View Dependent Claims (23, 24, 25, 26, 27, 28, 35)
-
-
29. A method comprising:
-
computing, based on output from a categorizer executed on one or more processors, a first measure regarding a first set of cases associated with a first issue and a second measure regarding a second set of cases associated with the first issue, wherein the first set of cases is associated with a first time period prior to occurrence of an event, and wherein the second set of cases is associated with a second time period after occurrence of the event; comparing, by the one or more processors, the first and second measures to determine an impact of the event on the first issue; identifying plural issues including the first issue; selecting the first issue from among the plural issues based on an expected return on investment; and effecting action with respect the first issue, wherein the event comprises the action. - View Dependent Claims (30)
-
-
31. A system, comprising:
-
one or more processors; a categorizer executable on the one or more processors; a search engine to identify at least one case in response to a query relating to an issue; an interface to receive an indication that the identified at least one case belongs to the issue; and a training module executable on the one or more processors to train the categorizer based on the received indication; the trained categorizer executable on the one or more processors to classify cases in a data set into the issue; a quantifier executable on the one or more processors in cooperation with the trained categorizer to compute a first quantification measure for a first time window, and a second quantification measure for a second time window, wherein the first time window is prior to occurrence of an event corresponding to the issue, and the second time window is after occurrence of the event; and a module executable on the one or more processors to determine an impact of the event on the issue based on comparing the first and second quantification measures. - View Dependent Claims (36)
-
Specification