Automated scheme for identifying user intent in real-time

• US 8,082,264 B2
• Filed: 12/18/2007
• Issued: 12/20/2011
• Est. Priority Date: 04/07/2004
• Status: Active Grant

• Current Assignee: Oracle OTC Subsidiary LLC (Oracle Corporation)
• Original Assignee: Inquira Incorporated (Oracle Corporation)
• Inventors: Bierner, Gann Alexander
• Primary Examiner(s): Cottingham, John
• Assistant Examiner(s): Richardson, James E
• Application Number:
  US11/959,307
• Publication Number:
  US 20080104037A1
• Time in Patent Office: 1,463 Days
• Field of Search: 707705-708, 707/771, 707766-768, 715/816
• US Class Current: 707/767
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence

Abstract

An intent guessing system receives partial user queries as they are entered by the user. The partial user queries are compared with different intents derived from previously logged queries. Guesses are made as to which of the intents are associated with the partial user query. The intent guesses are then provided as responses to the user query. Features are identified for the earlier logged queries and associated with the derived intents. The derived intents and associated features are then used to identify intents for the partial user queries.

Claims

1. A method of processing a current user query, comprising:

training a model, said training comprising;

continuously logging previous user queries;

extracting a question from the continuously logged previous user queries;

analyzing the extracted question to obtain at least one feature;

assigning an intent to the extracted question; and

adding an entry to the model and mapping the obtained feature for the extracted question to the assigned intent for the extracted question in the added entry of the model;

receiving an initial portion of the current user query, wherein the initial portion includes a particular string of text;

guessing an intent associated with the initial portion of the current user query while a remainder of the current user query is being received, wherein said guessing using only the initial portion further comprises;

analyzing the initial portion to obtain at least one feature;

comparing, using a processing device, the obtained at least one feature of the initial portion to entries in the model;

selecting, using the processing device, at least one entry from the model based on the comparison; and

determining the intent for the initial portion of the current user query according to an intent included in the selected entry;

wherein the intent guess obtained for the initial portion according to said analyzing, said comparing, said selecting, and said determining, does not include any portion of the particular text string; and

responding to the initial portion of the current user query with the intent guess.

2. The method according to claim 1, wherein the at least one feature includes at least one selected from the group comprising query tokens, query token stems, and concepts.

3. The method according to claim 1, further comprising:

identifying a parameterized intent guess that includes at least one concept;

identifying words in the current user query that are associated with the concept;

modifying the parameterized intent guess by replacing the concept with the identified words from the current user query; and

responding to the current user query with the modified parameterized intent guess.

4. The method according to claim 1, further comprising:

identifying a response associated with the intent guess; and

either displaying the response along with the associated intent guess or displaying the response instead of the associated intent guess.

5. The method according to claim 4, wherein the response includes links to web pages or additional information associated with the intent guess.

6. The method according to claim 1, further comprising:

identifying a plurality of intent guesses associated with the current user query;

identifying an order for the intent guesses of the plurality according to how well the intent guesses match the current user query; and

displaying the intent guesses in the identified order.

7. An article of manufacture including a 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 an initial portion of a current user query, wherein the initial portion includes a particular string of text;

training a model, said training comprising;

continuously logging previous user queries;

extracting a question from the continuously logged previous user queries;

analyzing the extracted question to obtain at least one feature;

assigning an intent to the extracted question; and

adding an entry in the model and mapping the obtained feature for the extracted question to the assigned intent for the extracted question in the added entry of the model;

guessing an intent associated with the initial portion of the current user query while a remainder of the current user query is being received, wherein said guessing using only the initial portion further comprises;

inputting the initial portion into a context generator to obtain at least one feature corresponding to the initial portion;

analyzing the obtained feature of the initial portion according to the model, wherein the analysis includes comparing the obtained at least one feature of the initial portion to entries in the model;

selecting, using the processing device, at least one entry from the model based on the comparison; and

determining the intent for the initial portion of the current user query according to an intent included in the selected entry;

wherein the intent guess obtained for the initial portion according to said analysis, said selecting, and said determining does not include any portion of the particular text string; and

responding to the initial portion of the current user query with the intent guess.

8. The article of manufacture according to claim 7, wherein the operations further comprise:

identifying a first subset of words from the initial portion while the current user query is being entered;

making an intent guess based on the first subset of words; and

replying to the first subset of words with the intent guess while the current user query is still being entered.

9. The article of manufacture according to claim 7, wherein the operations further comprise:

identifying at least one question class that includes at least one concept;

identifying words in the current user query that are associated with the concept;

modifying the identified question class by replacing the concept with the identified words in the current user query; and

responding to the current user query with the modified question class.

10. The article of manufacture according to claim 9, wherein the operations further comprise:

identifying confidence factors for the question classes; and

displaying the question classes according to the identified confidence factors.

11. The article of manufacture according to claim 9, wherein the question classes include associated probability values that indicate a probability that the question classes correctly respond to the current user query and the question classes are sent or not sent in response to the current user query according to the associated probability values.

12. The article of manufacture according to claim 7, wherein the operations further comprise:

identifying additional information associated with the intent guess; and

either displaying the identified additional information with the intent guess or displaying the additional information instead of the intent guess.

13. An apparatus, comprising:

a processor; and

a memory to store instructions that, when executed by the processor, are operable to;

receive an initial portion of a current user query, wherein the initial portion includes a particular string of text;

train a model, said training comprising;

continuously logging previous user queries;

extracting a question from the continuously logged previous user queries;

analyzing the extracted question to obtain at least one feature;

assigning an intent to the extracted question; and

adding an entry to the model and mapping the obtained feature for the extracted question to the assigned intent for the extracted question in the added entry of the model;

guess an intent associated with the initial portion of the current user query while a remainder of the current user query is being received, wherein the guess using only the initial portion includes the following operations;

extracting at least one feature from the initial portion;

comparing, using the processing device, the extracted at least one feature to entries from the model;

selecting, using the processing device, at least one entry from the model based on the comparison; and

determining the intent for the initial portion of the current user query according to an intent included in the selected entry;

wherein the intent guess obtained for the initial portion according to said extracting, said comparing, said selecting, and said determining does not include any portion of the particular text string; and

respond to the initial portion of the current user query with the intent guess.

14. The apparatus of claim 13, wherein the processing device is configured to:

identify a parameterized intent guess that includes at least one concept;

identify words in the current user query that are associated with the concept;

modify the parameterized intent guess by replacing the concept with the identified words from the current user query; and

respond to the current user query with the modified parameterized intent guess.

15. The apparatus of claim 13, wherein the at least one feature includes at least one selected from the group comprising query tokens, query token stems, and concepts.