Intent management tool for identifying concepts associated with a plurality of users' queries

• US 8,781,813 B2
• Filed: 08/14/2006
• Issued: 07/15/2014
• Est. Priority Date: 08/14/2006
• Status: Active Grant

• Current Assignee: Oracle OTC Subsidiary LLC (Oracle Corporation)
• Original Assignee: Oracle OTC Subsidiary LLC (Oracle Corporation)
• Inventors: Cooper, Edwin Riley, Dukes, Michael Peter, Bierner, Gann Alexander, Beghelli, Filippo Ferdinando Paulo
• Primary Examiner(s): Kazeminezhad, Farzad
• Application Number:
  US11/464,446
• Publication Number:
  US 20090089044A1
• Time in Patent Office: 2,892 Days
• Field of Search: 704/9, 705/10, 707/6, 707/100, 707/3, 395/12
• US Class Current: 704/9
• CPC Class Codes: G06F 16/283 Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
• CPC Class Codes: G06F 16/3338 Query expansion
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 40/263 Language identification
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

Linguistic analysis is used to identify queries from a plurality of users over a period of time that use different natural language formations to request similar information. Common intent categories are identified for the queries requesting similar information. Intent responses can then be provided that are associated with the identified intent categories. An intent management tool can be used for identifying new intent categories, identifying obsolete intent categories, or refining existing intent categories. The said intent categories are used in identifying ontologies associated with the intent categories to help in selecting concepts as ontology parameters.

Claims

1. An article of manufacture including a non-transitory computer-readable medium having instructions stored thereon that, in response to execution by a computing device, cause the computing device to perform operations comprising:

receiving a group of text questions from a plurality of users over a period of time;

identifying groups of the logged questions that use a variety of different linguistic expressions to express similar information requests;

assigning intent categories applicable to each group of questions requesting similar information;

identifying ontologies associated with the intent categories;

presenting concepts in the identified ontologies for selection by a user;

assigning any of the selected concepts as ontology parameters for the associated intent categories;

displaying intent responses for the ontology parameters assigned to the intent categories;

configuring the intent responses for at least some of the intent categories, wherein each intent response provides applicable information to all of the questions assigned to the associated intent category;

identifying ones of the intent categories associated with at least a first threshold number of the questions, and web-posting all of the intent responses associated with said identified intent categories together on an enterprise home web page if not currently web-posted, wherein said web-posting is in response to said identification, and wherein said web-posting of all of the intent responses is visible to a user of the enterprise home web page;

identifying ones of the intent categories that are not associated with at least a second threshold number of the questions, and removing intent responses for any of said identified intent categories if currently web-posted, wherein said removal is in response to said identification; and

for those ones of the intent categories with no associated intent response, identifying ones of the intent categories associated with at least a third threshold number of the questions, and in response to said identification, initiating generation of an intent response for each of said identified intent categories.

2. The article of manufacture of claim 1, wherein the operations further comprise assigning parameters to the intent categories that control how the associated intent responses are displayed.

3. The article of manufacture of claim 2, wherein the operations further comprise associating user parameters with the intent categories that cause a search engine to display different information associated with a user submitting the question or associated with a query history associated with the user submitting the question.

4. The article of manufacture of claim 1, wherein the operations further comprise:

conducting a spelling analysis to replace unidentifiable words in the questions;

conducting a punctuation analysis to correct punctuation in the questions;

conducting a sentence analysis to identify sentence structures and sentence elements in the questions;

conducting a stem analysis to add and identify other forms of words to the questions;

conducting a concept analysis using one or more ontologies to identify concepts related to the questions; and

conducting a linguistic analysis using the identified concepts, sentence structures, sentence elements, and other added and identified forms of words, wherein the intent category assignment is based on said linguistic analysis.

5. The article of manufacture of claim 1, wherein the operations further comprise configuring at least some of the intent categories into an intent hierarchy that causes the intent responses associated with the intent categories to be displayed according to the configured intent hierarchy.

6. The article of manufacture of claim 1, wherein the operations further comprise assigning response parameters to the intent responses that cause a search engine to display both the intent responses associated with the intent categories and also display any additional information associated with the response parameters assigned to the intent responses.

7. A method, comprising:

logging queries received from a plurality of users by an enterprise information system;

using a clustering engine to identify clusters of the logged queries;

generating names for the clusters of logged queries;

using the generated names to create intent categories pertinent to the queries in the same clusters;

using a linguistic matching language to match the queries in the same clusters with the intent categories;

identifying ontologies associated with the intent categories;

presenting concepts in the identified ontologies for selection by a user;

assigning any of the selected concepts as ontology parameters for the associated intent categories;

displaying intent responses for the ontology parameters assigned to the intent categories;

configuring the intent responses for at least some of the intent categories;

after matching queries to the intent categories, identifying ones of the intent categories matched with at least a first threshold number of the queries, and configuring all of the intent responses associated with said identified intent categories to be displayed together on an enterprise home web page if not currently configured to be displayed, wherein said display of all of the intent responses is visible to a user of the enterprise home web page;

after matching queries to the intent categories, identifying ones of the intent categories that are not matched with at least a second threshold number of the queries, and removing a display configuration for intent responses for any of said identified intent categories if currently configured to be displayed; and

after matching queries to the intent categories, for those ones of the intent categories with no associated intent response, identifying ones of the intent categories matched with at least a third threshold number of the questions, and initiating generation of an intent response for each of said identified intent categories.

8. The method according to claim 7, including using ontology elements associated with the enterprise information system as features for the clustering engine and then using at least some of the cluster names generated by the clustering engine to create the intent categories.

9. The method according to claim 8, including:

identifying queries that do not match any of the created intent categories;

submitting the identified queries to the clustering engine;

generating new cluster names for the clusters of non-matching queries; and

using the new cluster names to generate additional intent categories.

10. The method according to claim 9, including:

automatically generating an intent hierarchy by identifying a new intent category having an associated set of queries that comprise a sub-set of queries for an existing intent category and identifying the new intent category as a child of the existing intent category; and

identifying a new intent category having a subgroup of associated queries that comprise all of the queries matching the existing intent category and identifying the new intent category as a parent of the existing intent category.

11. The method according to claim 7, including processing queries that remain unmatched after using the linguistic matching language, said processing including:

identifying relationships between the unmatched queries using the clustering engine;

grouping the unmatched queries according to an output of the clustering engine; and

filtering or ranking the query groupings according to the amount of queries contained therein on a per-group basis, wherein said filtering or ranking identifies a first one of the query groupings having more queries contained therein than a second different one of the filtered query groupings.

12. The method according to claim 11, wherein said processing of the queries that remain unmatched after using the linguistic matching language further includes:

outputting a request to correlate the first query grouping with one of the existing intent categories or a new intent category; and

receiving back an input assigning the first query grouping with a particular one of the existing intent categories or the new intent category;

wherein a duration of time passes after assigning the first query grouping where the queries of the second query grouping are unassigned to the intent categories including the new intent category, wherein said duration of time is greater than zero.

13. The method according to claim 11, wherein said processing of the queries that remain unmatched after using the linguistic matching language further includes:

generating a new intent category for the first query grouping and correlating the newly generated intent category with the queries of the first query grouping;

wherein the queries of the second query grouping remain uncorrelated with the intent categories including the new intent category for a duration of time after said correlation, wherein said duration of time is greater than zero.

14. The method according to claim 7, including:

identifying relationships between unmatched queries using the clustering engine; and

modifying one of the existing intent categories based on a relationship identified by the clustering engine.

15. The method of claim 7, including:

inputting a set of features into the clustering engine;

grouping the logged queries into different clusters according to the input feature set; and

outputting the clusters from the cluster engine, wherein the matching using the linguistic matching language is based on the output of the cluster engine.

16. A non-transitory computer readable medium encoded with instructions that, if executed, result in:

receiving a first group of text questions logged from a plurality of users over a first period of time;

identifying groups of the logged questions that use a variety of different linguistic expressions to express similar information requests;

assigning intent categories applicable to each group of questions requesting similar information;

identifying ontologies associated with the intent categories;

presenting concepts in the identified ontologies for selection by a user;

assigning any of the selected concepts as ontology parameters for the associated intent categories;

displaying intent responses for the ontology parameters assigned to the intent categories;

configuring the intent responses for at least some of the intent categories, wherein each intent response provides applicable information to all of the questions assigned to the associated intent category;

identifying ones of the intent categories associated with at least a first threshold number of the questions, and configuring all of the intent responses associated with said identified intent categories to be displayed together on an enterprise home web page if not currently configured to be displayed, wherein said display of all of the intent responses is visible to a user of the enterprise home web page;

identifying ones of the intent categories that are not associated with at least a second threshold number of the questions, and deleting a display configuration for intent responses for any of said identified intent categories if currently configured to be displayed; and

for those ones of the intent categories with no associated intent response, identifying ones of the intent categories associated with at least a third threshold number of the questions, and initiating generation of an intent response for each of said identified intent categories.

17. The non-transitory computer readable medium of claim 16, wherein the instructions, if executed, result in:

intelligently adapting the existing intent categories over a second subsequent period of time based on changes over time in searching terminology, said intelligent adapting of the existing intent categories including;

receiving a second group of text questions logged for the one or more servers during the second subsequent period of time;

identifying text questions from the second group that do not match with the existing intent categories; and

modifying one of the existing intent categories based on information from the identified text questions of the second group.

18. The non-transitory computer readable medium of claim 17, wherein the instructions, if executed, result in:

processing the identified text questions of the second group using a cluster engine, wherein said one of the existing intent categories is modified based on an output of the cluster engine.

19. A non-transitory computer readable medium encoded with instructions that, if executed, result in:

using a clustering engine to identify clusters of logged queries received from a plurality of users, including;

inputting a set of features into the clustering engine;

grouping the logged queries into different clusters according to the input feature set; and

outputting the clusters;

generating names for the output clusters of the logged queries;

using the generated names to create intent categories pertinent to the queries in the same clusters;

using a linguistic matching language to match the queries in the same clusters with the intent categories, wherein the linguistic matching is performed after conducting the clustering using the clustering engine, wherein the linguistic matching is performed on the output of the clustering engine, wherein the matched queries are associated with each other through a combination of the linguistic matching and clustering operations;

identifying ontologies associated with the intent categories;

presenting concepts in the identified ontologies for selection by a user;

assigning any of the selected concepts as ontology parameters for the associated intent categories;

displaying intent responses for the ontology parameters assigned to the intent categories;

configuring the intent responses for at least some of the intent categories;

after matching queries to the intent categories, identifying ones of the intent categories matched with at least a first threshold number of the queries, and configuring all of the intent responses associated with said identified intent categories to be displayed together on an enterprise home web page if not currently configured to be displayed, wherein said display of all of the intent responses is visible to a user of the enterprise home web page;

after matching queries to the intent categories, identifying ones of the intent categories that are not matched with at least a second threshold number of the queries, wherein intent responses for any of said identified intent categories are not configured to be displayed with the other intent responses; and

after matching queries to the intent categories, for those ones of the intent categories with no associated intent response, identifying ones of the intent categories matched with at least a third threshold number of the questions, and initiating generation of an intent response for each of said identified intent categories.