Method to allow for question and answer system to dynamically return different responses based on roles

• US 10,382,440 B2
• Filed: 09/22/2016
• Issued: 08/13/2019
• Est. Priority Date: 09/22/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Byron, Donna K., Feirouz, Elie, Jamrog, Daniel M., Witherspoon, Kristin A.
• Primary Examiner(s): Turchen, James R
• Application Number:
  US15/273,136
• Publication Number:
  US 20180081934A1
• Time in Patent Office: 1,055 Days
• CPC Class Codes: G06F 21/6245 Protecting personal data, e.g. for financial or medical purposes
• CPC Class Codes: G06F 2221/2141 Access rights, e.g. capability lists, access control lists, access tables, access matrices
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: H04L 63/10 for controlling access to devices or network resources
• CPC Class Codes: H04L 63/102 Entity profiles
• CPC Class Codes: H04L 63/105 Multiple levels of security

Abstract

Embodiments are directed to a question and answer (QA) pipeline system that adjusts answers to input questions based on a user criteria, thus implementing a content-based determination of access permissions. The QA system allows for information to be retrieved based on permission granted to a user. Documents are ingested and assigned an access level based on a defined information access policy. The QA system is implemented with the defined information access policy, the ingested documents, and the inferred access levels. For the QA system implementation, a user enters a question; primary search and answer extraction stages are performed; candidate answer extraction is performed using only content the user is allowed to access; the candidate answers are scored, ranked, and merged; ranked answers based on user permissions are filtered; and answers are provided to the user.

Claims

1. A computer implemented method for adjusting answers to questions based on a user criteria by an information handling system capable of answering questions, the system comprising a processor and a memory comprising instructions executed by the processor, the method comprising:

manually tagging documents for deep natural language processing (NLP);

extracting features from the manually-tagged documents;

providing each of a plurality of documents with an access level based on extracted features;

receiving a question from a user having a user profile comprising a responsibility role of the user and a permission level of the user;

deriving a content access constraint criteria of the user from the responsibility role and the permission level;

generating a plurality of candidate answers for the question;

removing one or more candidate answers from the plurality of candidate answers, wherein the one or more candidate answers have an access level higher than the content access constraint criteria of the user, wherein the access level is provided by performing deep NLP and feature extraction on each text segment of each document, wherein the extracted features include particular entity types, relation types, predicates, and indicating words;

applying natural language processing, deep analytic analysis, and evidence scoring to the plurality of documents having the access level no higher than the content access constraint criteria to select an answer to the question from the remaining candidate answers; and

providing to the user a notification comprising the answer to the question.

2. The method of claim 1, wherein the permission level is based on a subscription.

3. The method of claim 1, wherein the content access constraint criteria controls access to embedded content in documents based on a passage classifier model analysis including accumulated discovery.

4. The method of claim 3, wherein a first responsibility role for a first employee without management responsibility receives a more restrictive content access constraint criteria than a second employee with management responsibility.

5. The method of claim 1, wherein the access level is provided by applying a classification-based policy.

6. The method of claim 1, wherein, in addition to the notification comprising the answer to the question, the user is provided with an indication of the access level of the answer being provided.

7. A computer program product for adjusting answers to questions based on a user criteria by an information handling system capable of answering questions, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:

manually tag documents for deep natural language processing (NLP);

extract features from the manually-tagged documents;

provide each of a plurality of documents with an access level based on extracted features;

receive a question from a user having a user profile comprising a responsibility role of the user and a permission level of the user;

derive a content access constraint criteria of the user from the responsibility role and the permission level;

generate a plurality of candidate answers for the question;

remove one or more candidate answers from the plurality of candidate answers, wherein the one or more candidate answers have an access level higher than the content access constraint criteria of the user, wherein the access level is provided by performing deep NLP and feature extraction on each text segment of each document, wherein the extracted features include particular entity types, relation types, predicates, and indicating words;

apply natural language processing, deep analytic analysis, and evidence scoring to the plurality of documents having the access level no higher than the content access constraint criteria to select an answer to the question from the remaining candidate answers; and

provide to the user a notification comprising the answer to the question.

8. The computer program product of claim 7, wherein the permission level is based on a subscription.

9. The computer program product of claim 7, wherein the content access constraint criteria controls access to embedded content in documents based on a passage classifier model analysis including accumulated discovery.

10. The computer program product of claim 9, wherein a first responsibility role for a first employee without management responsibility receives a more restrictive content access constraint criteria than a second employee with management responsibility.

11. The computer program product of claim 7, wherein the access level is provided by applying a classification-based policy.

12. A system for adjusting answers to questions based on a user criteria by an information handling system capable of answering questions, the system comprising:

a memory comprising executable instructions; and

a processor configured to execute the executable instructions to;

manually tag documents for deep natural language processing (NLP);

extract features from the manually-tagged documents;

provide each of a plurality of documents with an access level based on extracted features;

receive a question from a user having a user profile comprising a responsibility role of the user and a permission level of the user;

derive a content access constraint criteria of the user from the responsibility role and the permission level;

generate a plurality of candidate answers for the question;

remove one or more candidate answers from the plurality of candidate answers,

wherein the one or more candidate answers have an access level higher than the content access constraint criteria of the user, wherein the access level is provided by performing deep NLP and feature extraction on each text segment of each document, wherein the extracted features include particular entity types, relation types, predicates, and indicating words;

apply natural language processing, deep analytic analysis, and evidence scoring to the plurality of documents having the access level no higher than the content access constraint criteria to select an answer to the question from the remaining candidate answers; and

provide to the user a notification comprising the answer to the question.

13. The system of claim 12, wherein the permission level is based on a subscription.

14. The system of claim 12, wherein the content access constraint criteria controls access to embedded content in documents based on a passage classifier model analysis including accumulated discovery.

15. The system of claim 14, wherein a first responsibility role for a first employee without management responsibility receives a more restrictive content access constraint criteria than a second employee with management responsibility.

16. The system of claim 12, wherein the access level is provided by applying a classification-based policy.

17. The system of claim 12, wherein, in addition to the notification comprising the answer to the question, the user is provided with an indication of the access level of the answer being provided.