IP Library Granted Patent US 10,073,837
Granted Patent B2
US 10,073,837 · App. 14/815,714 · Granted Sep 11, 2018

Method and system for implementing alerts in semantic analysis technology

Inventors: Mehrshad Setayesh (Lafayette, CO); Marta Lynn Studinger (Centennial, CO); Sonal Bablani (Broomfield, CO); Laxmi Narsimha Rao Oruganti (Telengana, IN); Mukesh Jha (Telengana, IN)
Assignee: ORACLE INTERNATIONAL CORPORATION
G06F17/2785G06F9/542G06F17/241G06F17/30707H04L43/16H04L51/14H04L51/32
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,073,837
App. No.
14/815,714
Granted
Sep 11, 2018
Kind
B2
Abstract

Disclosed is an approach for allowing an entity to perform semantic analysis upon data associated with an entity, and to automatically perform categorization of that data and to generate alerts based on a set of predefined rules. A topic is created based on which to categorize a set of semantically analyzed messages. A set of rules is defined, wherein an alert is generated when the set of rules is satisfied for the semantically analyzed messages.

Claims (62)

1. A method comprising:

generating a topic vector representing semantic significance of one or more terms, wherein generating the topic vector comprises:

performing a search on a sample corpus to obtain a plurality of samples;

accessing a sematic space to determine a plurality of semantic vectors based at least in part on the plurality of samples,

presenting a user interface comprising a plurality of themes based at least in part on the plurality of semantic vectors,

receiving a user selection, via the user interface, of one or more themes from the plurality of themes, and

saving the topic vector based at least in part on the user selection of one or more themes;

associating a set of alert rules with the topic vector, wherein the set of alert rules define one or more thresholds that trigger one or more alerts;

categorizing a set of messages based at least in part on analysis of the set of messages against the topic vector, to obtain a categorized set of messages;

determining whether the categorized set of messages satisfies the one or more thresholds defined in the set of alert rules; and

if the one or more thresholds are satisfied, generating a corresponding alert of the one or more alerts.

2. The method of claim 1 , wherein the corresponding alert comprises at least one of an email message, a text message, a social media message, an audio communication, and a video communication.

3. The method of claim 1 , further comprising performing an automatic processing task in response to the corresponding alert.

4. The method of claim 1 , further comprising accessing a rules database to retrieve the set of alert rules, the rules database comprising rules for a plurality of topics.

5. The method of claim 1 , wherein categorizing the set of messages comprises vectorizing the set of messages using a collection of vector values for a plurality of dimensions for a plurality of terms.

6. The method of claim 1 , wherein at least one alert rule of the set of alert rules corresponds to a sentiment associated with the categorized set of messages.

7. The method of claim 1 , wherein at least one alert rule of the set of alert rules corresponds to a volume of the categorized set of messages.

8. The method of claim 1 , wherein at least one alert rule of the set of alert rules corresponds to a change in a volume of the categorized set of messages.

9. The method of claim 1 , wherein the analysis of the set of messages against the topic vector comprises, for each particular message in the set of messages:

vectorizing the particular message to obtain a vectorized message; and

comparing the vectorized message to the topic vector to determine a similarity, if any, between the topic vector and the vectorized message.

10. A computer program product embodied in a non-transitory computer readable medium, the computer readable medium having stored thereon a sequence of instructions which, when executed by a processor causes the processor to execute a method comprising:

generating a topic vector representing semantic significance of one or more terms, wherein generating the topic vector comprises:

performing a search on a sample corpus to obtain a plurality of samples;

accessing a sematic space to determine a plurality of semantic vectors based at least in part on the plurality of samples,

presenting a user interface comprising a plurality of themes based at least in part on the plurality of semantic vectors,

receiving a user selection, via the user interface, of one or more themes from the plurality of themes, and

saving the topic vector based at least in part on the user selection of one or more themes;

associating a set of alert rules with the topic vector, wherein the set of alert rules define one or more thresholds that trigger one or more alerts;

categorizing a set of messages based at least in part on analysis of the set of messages against the topic vector, to obtain a categorized set of messages;

determining if the categorized set of messages satisfies the one or more thresholds defined in the set of alert rules; and

if the one or more thresholds are satisfied, generating a corresponding alert of the one or more alerts.

11. The computer program product of claim 10 , wherein the corresponding alert comprises at least one of an email message, a text message, a social media message, an audio communication, and a video communication.

12. The computer program product of claim 10 , further comprising performing an automatic processing task in response to the corresponding alert.

13. The computer program product of claim 10 , further comprising accessing a rules database to retrieve the set of alert rules, the rules database comprising rules for a plurality of topics.

14. The computer program product of claim 10 , wherein categorizing the set of messages comprises vectorizing the set of messages using a collection of vector values for a plurality of dimensions for a plurality of terms.

15. The computer program product of claim 10 , wherein at least one alert rule of the set of alert rules corresponds to a sentiment associated with the categorized set of messages.

16. The computer program product of claim 10 , wherein at least one alert rule of the set of alert rules corresponds to a volume of the categorized set of messages.

17. The computer program product of claim 10 , wherein at least one alert rule of the set of alert rules corresponds to a change in a volume of the categorized set of messages.

18. The computer program product of claim 10 , wherein the analysis of the set of messages against the topic vector comprises, for each particular message in the set of messages:

vectorizing the particular message to obtain a vectorized message; and

comparing the vectorized message to the topic vector to determine a similarity, if any, between the topic vector and the vectorized message.

19. A computer system comprising:

a computer processor to execute a set of program code instructions;

a memory to hold the program code instructions, in which the program code instructions comprises program code to perform operations comprising:

generating a topic vector representing semantic significance of one or more terms, wherein generating the topic vector comprises:

performing a search on a sample corpus to obtain a plurality of samples;

accessing a sematic space to determine a plurality of semantic vectors based at least in part on the plurality of samples,

presenting a user interface comprising a plurality of themes based at least in part on the plurality of semantic vectors,

receiving a user selection, via the user interface, of one or more themes from the plurality of themes, and

saving the topic vector based at least in part on the user selection of one or more themes;

associating a set of alert rules with the topic vector, wherein the set of alert rules define one or more thresholds that trigger one or more alerts;

categorizing a set of messages based at least in part on analysis of the set of messages against the topic vector, to obtain a categorized set of messages;

determining whether the categorized set of messages satisfies the one or more thresholds defined in the set of alert rules; and

if the one or more thresholds are satisfied, to generate a corresponding alert of the one or more alerts.

20. The system of claim 19 , wherein the corresponding alert comprises at least one of an email message, a text message, a social media message, an audio communication, and a video communication.

21. The system of claim 19 , further comprising performing an automatic processing task in response to the corresponding alert.

22. The system of claim 19 , further comprising accessing a rules database to retrieve the set of alert rules, the rules database comprising rules for a plurality of topics.

23. The system of claim 19 , wherein categorizing the set of messages comprises vectorizing the set of messages using a collection of vector values for a plurality of dimensions for a plurality of terms.

24. The system of claim 19 , wherein at least one alert rule of the set of alert rules corresponds to a sentiment associated with the categorized set of messages.

25. The system of claim 19 , wherein at least one alert rule of the set of alert rules corresponds to a volume of the categorized set of messages.

26. The system of claim 19 , wherein at least one alert rule of the set of alert rules corresponds to a change in a volume of the categorized set of messages.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 2, 2015
From: SETAYESH, MEHRSHAD; STUDINGER, MARTA LYNN; BABLANI, SONAL; ORUGANTI, LAXMI NARASIMHA RAO; JHA, MUKESH
To: ORACLE INTERNATIONAL CORPORATION
Reel/Frame 036720/0792 →
Continuity (2)
Provisional Application 62031771 · Jul 31, 2014
Related Publication 20160034571A1 · Feb 4, 2016