Methods and systems for acquiring user related information using natural language processing techniques

• US 9,396,179 B2
• Filed: 08/30/2012
• Issued: 07/19/2016
• Est. Priority Date: 08/30/2012
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Stavrianou, Anna, Brun, Caroline
• Primary Examiner(s): Shah, Paras D
• Assistant Examiner(s): THOMAS-HOMESCU, ANNE L
• Application Number:
  US13/599,043
• Publication Number:
  US 20140067369A1
• Time in Patent Office: 1,419 Days
• Field of Search: 704/275, 704/257, 704/201, 704/9, 704/254, 704/276, 704/235, 709/203, 707/723, 707/769, 707/758, 707/741, 707/708, 707/639, 707/749, 707/765, 707/748, 707/661, 707/728
• US Class Current: 1/1
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06Q 30/016 After-sales
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
• CPC Class Codes: H04M 3/51 Centralised call answering arrangements requiring operator intervention , e.g. call or contact centers for telemarketing
• CPC Class Codes: H04M 3/5158 in combination with automated outdialling systems devices for signalling identity of wanted subscriber H04M1/26
• CPC Class Codes: H04M 3/5191 interacting with the Internet

Abstract

Systems and methods for acquiring information associated with a user by using NLP techniques are disclosed. One or more phrases are classified in one or more categories at least partly on the basis of a period for which a product has been used by the user, the user'"'"'s experience with the product, preferences of the user, or needs of the user by applying one or more natural language processing (NLP) techniques. The one or more phrases are extractable from an electronic publication at least partly on the basis of on a predefined set of verbs, a predefined set of domain-specific terms, and terms indicative of temporal information. One or more terms from the classified phrases are extracted, in which the one or more terms are indicative of the information about the user.

Claims

1. A method of providing product recommendations to a user, the method comprising:

receiving at a remotely-located customer support center, via a user'"'"'s voice-call a query about a product;

searching, at an analytics server, a plurality of sources for one or more electronic documents attributed to the user, wherein the searching is initiated in response to receiving a request from the customer support center to create a user profile;

extracting, by the analytics server, one or more phrases from the searched electronic documents using a natural language parser;

classifying, by the analytics server, one or more phrases extracted from one or more electronic documents in a user'"'"'s experience category by applying a first set of rules comprising phrases indicative of a user'"'"'s experience with a product;

further classifying, by the analytics server, the one or more phrases into at least one category by applying a first set of rules, wherein the categories comprise at least one of;

a user'"'"'s preference category that comprises phrases that indicate preferences of the user, and

a user'"'"'s need category that comprises phrases that indicate needs of the user;

extracting, by the analytics server, a plurality of terms from the one or more classified phrases by applying a second set of rules to the classified phrases, wherein at least one term extracted from a phrase categorized as a user'"'"'s experience comprises a temporal term that relates the product and the user based on a rule from the second set of rules that comprises extracting a temporal term from a phrase categorized as a user'"'"'s experience when a verb in the categorized phrase comprises a past tense verb associated with possession and a portion of the categorized phrase refers to a domain of the product;

creating the user profile, at the analytics server, based on the one or more temporal terms extracted from the classified phrases by applying the second set of rules;

communicating, by the analytics server, the created user profile to the customer service center;

determining, at the analytics server, product information based on the created user profile to determine needs and preferences of the user in response to the received query; and

returning to the user, via an interactive voice response, product recommendations based on the product information and received query.

2. The method of claim 1 further comprising providing, by the analytics server, one or more services based on the user profile, wherein the one or more services correspond to a product support, a product recommendation, and a troubleshooting.

3. The method of claim 1, wherein a plurality of sources for one or more electronic documents attributed to the user include online sources corresponding to at least one of a product review website, a blog, a forum, or an e-mail.

4. The method of claim 1, wherein the classification is performed using one or more NLP techniques comprising one or more of word tokenization, word lemmatization, part-of-speech tagging, Named Entity detection, or syntactic parsing.

5. The method of claim 1, wherein at least a subset of the one or more electronic documents are creatable from a voice call by applying one or more speech to text (STT) conversion techniques, wherein the user initiates the voice call to obtain information about a product.

13. The method of claim 1, wherein determining, by the analytics server, product information is based on a length of experience for the user with the product, wherein the length of experience is calculated based on the extracted temporal term and a current date.

6. A method of providing one or more services to a user, the method comprising:

receiving at a remotely-located customer support center, via a user'"'"'s voice-call a query about a product;

searching, at an analytics server, a plurality of sources for one or more electronic documents attributed to a user, wherein the searching is initiated in response to receiving a request from the customer support center to create a user profile;

extracting, by the analytics server, one or more phrases from the searched electronic documents using a natural language parser;

classifying, by the analytics server, one or more phrases extracted from one or more electronic documents in a user'"'"'s experience category by applying a first set of rules comprising phrases indicative of a user'"'"'s experience with a product;

further classifying, by the analytics server, the one or more phrases into at least one category by applying a first set of rules, wherein the categories comprise at least one of;

a user'"'"'s preference category that comprises phrases that indicate preferences of the user, and

a user'"'"'s need category that comprises phrases that indicate needs of the user; and

extracting, by the analytics server, a plurality of terms from the one or more classified phrases by applying a second set of rules to the classified phrases, wherein at least one term extracted from a phrase categorized as a user'"'"'s experience comprises a temporal term that relates a product and the user based on a rule from the second set of rules that comprises extracting a temporal term from a phrase categorized as a user'"'"'s experience when a verb in the categorized phrase comprises a past tense verb associated with possession and a portion of the categorized phrase refers to a domain of the product;

creating the user profile, at the analytics server, based on the one or more temporal terms extracted from the classified phrases by applying the second set of rules;

communicating, by the analytics server, the created user profile to the customer service center;

determining, at the analytics server, product information based on the created user profile to determine needs and preferences of the user in response to the received query; and

returning to a user, via an interactive voice response, product recommendations based on the product information and received query.

7. The method of claim 6, a plurality of sources for one or more electronic documents attributed to the user include online sources corresponding to at least one of a product review website, a blog, a forum, or an e-mail.

8. The method of claim 6, wherein the classifying is performed using one or more NLP techniques comprising one or more of word tokenization, word lemmatization, part-of-speech tagging, Named Entity detection, or syntactic parsing.

14. The method of claim 6, wherein determining, by the analytics server, product information is based on a length of experience for the user with the product, wherein the length of experience is calculated based on the extracted temporal term and a current date.

9. A system for creating a user profile, the system comprising:

a remotely-located customer support center configured to;

receive, via a user'"'"'s voice-call a query about a product; and

an analytics server configured to;

search a plurality of sources for one or more electronic documents attributed to a user, wherein the searching is initiated in response to receiving a request from the customer support center to create a user profile;

extract one or more phrases from the searched electronic documents using a natural language parser;

classify one or more phrases extracted from one or more electronic documents in a user'"'"'s experience category by applying a first set of rules comprising phrases indicative of a user'"'"'s experience with a product;

further classify the one or more phrases into at least one category by applying a first set of rules, wherein the categories comprise at least one of;

a user'"'"'s preference category that comprises phrases that indicate preferences of the user, and

a user'"'"'s need category that comprises phrases that indicate needs of the user;

extract a plurality of terms from the one or more classified phrases by applying a second set of rules to the classified phrases, wherein at least one term extracted from a phrase categorized as a user'"'"'s experience comprises a temporal term that relates a product and the user based on a rule from the second set of rules that comprises extracting a temporal term from a phrase categorized as a user'"'"'s experience when a verb in the categorized phrase comprises a past tense verb associated with possession and a portion of the categorized phrase refers to a domain of the product;

create the user profile based on the one or more temporal terms extracted from the classified phrases by applying the second set of rules;

communicate the created user profile to the customer service center;

determine product information based on the created user profile to determine needs and preferences of the user in response to the received query; and

return to a user, via an interactive voice response, product recommendations based on the product information and received query.

10. The system of claim 9, wherein a plurality of sources for one or more electronic documents attributed to the user include online sources corresponding to at least one of a product review website, a blog, a forum, or an e-mail.

11. The system of claim 9, wherein the analytics server utilizes one or more NLP techniques comprising one or more of word tokenization, word lemmatization, part-of-speech tagging, Named Entity detection, or syntactic parsing, to classify the one or more phrases.

12. The system of claim 9, wherein at least a subset of the one or more electronic documents are creatable from a voice call by applying one or more speech to text (STT) conversion techniques, wherein the user initiates the voice call to obtain information about a product.

15. The system of claim 9, wherein determining, by the analytics server, product information is based on a length of experience for the user with the product, wherein the length of experience is calculated based on the extracted temporal term and a current date.