Implicit, specialized search of business objects using unstructured text

• US 7,634,467 B2
• Filed: 10/31/2006
• Issued: 12/15/2009
• Est. Priority Date: 10/31/2006
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Vogt, Bob, Platt, John C., Venolia, Gina, Anderson, Glen H., Ryan, Patricia A.
• Primary Examiner(s): Lu; Kuen S
• Assistant Examiner(s): OWYANG, MICHELLE N
• Application Number:
  US11/555,228
• Publication Number:
  US 20080104052A1
• Time in Patent Office: 1,141 Days
• Field of Search: 707 1- 7, 709/206
• US Class Current: 1/1
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 40/258 Heading extraction; Automatic titling; Numbering
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A method and system for providing business context corresponding to unstructured documents is provided. A business context correlation system analyzes the contents of a user'"'"'s unstructured document, identifies the business context of the unstructured document from the analyzed contents, and presents the business context along with the unstructured document to the user. The business context includes the relevant contextual objects and documents associated with one or more business applications. Providing a view of the business context along with the unstructured document allows the user more fully understand and act on the displayed document.

Claims

1. A computer-implemented method in a computer system with a processor for presenting a user business context corresponding to an unstructured document, the method comprising:

providing classifiers for classifying terms of unstructured documents into actions with parameters representing business objects derived from the unstructured documents, each classifier having been trained with training data of terms extracted from the unstructured documents and associated actions associated with the unstructured documents of the training data, a business object including a customer having a customer name;

providing a business application database storing business information relating to different types of business objects;

providing a tailored dictionary derived from a master dictionary of names of business objects and other business-related terms, the tailored dictionary containing terms of the master dictionary determined, based on characteristics of the user, to be more relevant to the user than other terms of the master dictionary, a characteristic of the user including job title of the user;

receiving an unstructured document;

extracting by the processor relevant terms from the received unstructured document that are in the provided tailored dictionary;

applying by the processor the provided classifiers to the extracted relevant terms to classify the relevant terms extracted from the received unstructured document into a plurality of actions related to a business object associated with a business application, at least some of the plurality of the actions having a parameter identifying the business object, each classifier providing a score associated with each of the plurality of actions;

identifying which of the plurality of actions have scores that satisfy a predetermined threshold score;

extracting from the unstructured document a parameter identifying the business object;

retrieving from the business application database business information of the business object to which the received unstructured document relates;

presenting to a user the unstructured document, an indication of each identified action along with its score for possible performance, and the retrieved business information of the business object to which the received unstructured document relates, wherein an action is performed using the extracted parameter; and

receiving from the user an indication to modify an identified action and to perform the modified identified action

so that the user can view the received unstructured document, the retrieved business information, and the score of each identified action to assist the user in deciding whether to modify and effect performance of an identified action.

2. The method of claim 1, wherein the unstructured document is an email message.

3. The method of claim 1, wherein the unstructured document is an IM message.

4. The method of claim 1, wherein the unstructured document is a transcribed telephone conversation.

5. The method of claim 1, wherein the at least one relevant term of the extracted relevant terms is extracted from the unstructured document using natural language processing.

6. The method of claim 1, further comprising:

determining whether the at least one action is appropriate to present based on one or more characteristics of the user; and

presenting the at least one action to the user upon determining that the at least one action is appropriate to present based on one or more characteristics of the user.

7. The method of claim 1, further comprising:

classifying the relevant terms extracted from the unstructured document into at least one document category;

identifying a work-flow process associated with the document category;

identifying a task to perform in the work-flow process; and

presenting the work-flow task to the user for possible execution.

8. The method of claim 7, wherein the work-flow task is related to a business object associated with a business application.

9. A computer-readable media encoded with computer executable instructions for controlling a computer system to present a user business context corresponding to an unstructured document, the computer-readable media including a storage device, by a method comprising:

accessing a tailored dictionary derived from a master dictionary of names of business objects and other business-related terms, the tailored dictionary containing terms of the master dictionary determined, based on characteristics of the user, to be more relevant to the user than other terms of the master dictionary, a characteristic of the user including job title of the user;

providing classifiers for classifying the unstructured document into document categories, each classifier having been trained with terms of the tailored dictionary extracted from unstructured documents and document categories associated with the unstructured documents;

providing an indication of work-flow processes associated with each category;

providing a business application database storing business information relating to different types of business objects;

receiving an unstructured document;

extracting one or more relevant terms of the tailored dictionary from the received unstructured document;

applying the provided classifiers to the extracted relevant terms to classify the relevant terms extracted from the unstructured document into a plurality of document categories, each document category having a score;

identifying from the provided indication of work-flow processes a work-flow process associated with the document category;

identifying a task to perform in the work-flow process;

extracting a parameter from the received unstructured document that identifies a business object;

retrieving from the business application database business information associated with the identified business object; and

presenting to the user the unstructured document, the retrieved business information, and for each document category with a score above a predetermined threshold score, the work-flow task for that document category for possible execution, wherein the work-flow task is associated with the extracted parameter.

10. The computer-readable media of claim 9, wherein the document category is a sales opportunity.

11. The computer-readable media of claim 9, wherein the document category is a request for information.

12. The computer-readable media of claim 9, wherein the document category is a customer issue.

13. The computer-readable media of claim 9, further comprising:

determining whether the work-flow task is appropriate to present based on one or more characteristics of the user; and

presenting the work-flow task to the user upon determining that the at least one action is appropriate to present based on one or more characteristics of the user.

14. The computer-readable media of claim 9, further comprising:

classifying the relevant terms extracted from the unstructured document into at least one action related to a business object associated with a business application; and

presenting the at least one action to the user for possible execution.

15. The computer-readable media of claim 14, further comprising:

determining whether the at least one action is appropriate to present based on one or more characteristics of the user; and

presenting the at least one action to the user upon determining that the at least one action is appropriate to present based on one or more characteristics of the user.

16. A business context correlation system with a processor and a memory, comprising:

a tailored dictionary derived from a master dictionary of names of business objects and other business-related terms, the tailored dictionary containing terms of the master dictionary determined, based on characteristics of a user, to be more relevant to the user than other terms of the master dictionary, a characteristic of the user including job title of the user;

classifiers for classifying terms of unstructured documents into actions with parameters representing business objects derived from the unstructured documents, each classifier providing a score and having been trained with terms of the tailored dictionary extracted from the unstructured documents and associated actions associated with the unstructured documents;

a business application database associated with a business application storing business information relating to business objects;

a component that extracts relevant terms from an unstructured document associated with a business object;

a component that classifies using the classifiers the relevant terms extracted from the unstructured document into a plurality of actions related to the business object associated with the unstructured document;

a component that extracts a parameter from the unstructured document identifying the business object;

a component that retrieves from the business application database business information for the business object identified by the parameter; and

a component that presents to a user the unstructured document, the business information, and for each of the plurality of actions with a score above a predetermined threshold score, the action along with its score for possible execution in reference to the business object identified by the extracted parameter

wherein the classifiers and components are implemented as computer-executable instructions stored in the memory for execution by the processor.

17. The system of claim 16, further comprising:

a component that classifies the relevant terms extracted from the unstructured document into at least one document category;

a component that identifies a work-flow process associated with the document category;

a component that identifies a task to perform in the work-flow process; and

a component that presents the work-flow task for possible execution.