SEMANTIC TRANSLATION MODEL TRAINING

• US 20150006143A1
• Filed: 06/27/2013
• Published: 01/01/2015
• Est. Priority Date: 06/27/2013
• Status: Active Grant

• Current Assignee: Arlington Technologies, LLC (Dominion Harbor Enterprises, LLC)
• Original Assignee: Avaya Incorporated
• Inventors: Skiba, David, Erhart, George, Becker, Lee, Matula, Valentine C.
• Granted Patent: US 10,599,765 B2
• Time in Patent Office: Days
• US Class Current: 704/2
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/263 Language identification
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation

Abstract

A semantic translation model system is described along with various methods and mechanisms for administering the same. The semantic translation model system proposed herein creates an intermediate representation and a knowledge base in multiple languages, reducing the amount of time and expensive resources typically required for translation and automatic response to written communications. The system also removes the problem of a translation being influenced by a person'"'"'s writing style and human misinterpretation and provides ongoing translation to keep the system current.

Claims

1. A method for creating semantic translation model training, comprising:

receiving, by a processor, an electronic communication, the electronic communication being in a first language;

performing a semantic analysis on content of the electronic communication;

determining from the semantic analysis of the electronic communication, that the content of the electronic communication is a question in the first language;

performing a syntactic analysis of the content based on the first language; and

translating at least some of the content to a second language by applying linguistic rules of the second language to the content.

2. The method of claim 1, wherein the syntactic analysis comprises identifying, wherein the question comprises one or more of a question asked, a comment made, and a statement posted in the first language.

3. The method of claim 2, wherein the semantic analysis comprises determining structure associated with how a question is asked in the first language.

4. The method of claim 3, wherein the translating comprises translating the identified at least one of keywords, phrases, questions, and answers in the first language to at least one of keywords, phrases, questions, and answers in the second language and then structuring the translated at least one of keywords, phrases, questions, and answers according to a structure associated with how a question is asked in the second language.

5. The method of claim 4, wherein an answer to the question is identified in the first language, wherein the answer includes one or more of a response, a post, and a comment to the question is translated into the second language to create a question and answer pair in the second language.

6. The method of claim 5, wherein the question and answer pair are used in a contact center to provide at least one of an agent and customer with answers in the second language.

7. The method of claim 5, wherein the question and answer pair are provided in an HTML document and served to a customer via a web server.

8. The method of claim 1, further comprising:

building a language model in the second language by automatically repeating the translating step for all question and answer pairs in the first language; and

providing the language model in the second language to at least one of a contact center agent and contact center customer that are speaking the second language but not the first language.

9. A non-transitory computer-readable medium comprising processor-executable instructions, the instruction comprising:

instructions configured to receive, by a processor, an electronic communication, the electronic communication being in a first language;

instructions configured to perform a semantic analysis on content of the electronic communication;

instructions configured to determine from the semantic analysis of the electronic configuration, that the content of the electronic communication is in the first language;

instructions configured to perform a syntactic analysis of the content based on the first language; and

instructions configured to translate at least some of the content to a second language by applying linguistic rules of the second language to the content.

10. The computer-readable medium of claim 9, wherein the semantic analysis comprises determining structure associated with how a question is asked in the first language.

11. The computer-readable medium of claim 9, wherein the syntactic analysis comprises identifying at least one of keywords, phrases, questions, and answers in the first language.

12. The computer-readable medium of claim 10, wherein an answer to the question is identified in the first language and wherein the answer to the question is translated into the second language to create a question and answer pair in the second language.

13. The computer-readable medium of claim 12, wherein the translating further comprises translating the identified at least one of keywords, phrases, questions, and answers in the first language to at least one of keywords, phrases, questions, and answers in the second language and then structuring the translated at least one of keywords, phrases, questions, and answers according to a structure associated with how a question is asked in the second language.

14. The computer-readable medium of claim 13, wherein the question and answer pair are used in a contact center to provide at least one of an agent and customer with answers in the second language.

15. The computer-readable medium of claim 14, wherein the question and answer pair are provided in an HTML document and served to a customer via a web server.

16. A communication system, comprising:

a translation module configured to determine that the system has received an electronic communication, the electronic communication being in a first language and performs the following operations;

receive an electronic communication, the electronic communication being in a first language;

perform a semantic analysis on content of the electronic communication;

determine from the semantic analysis of the electronic communication, that the content of the electronic communication is in the first language in a first language domain;

perform a syntactic analysis of the content based on the first language; and

translate at least some of the content to a second language in a second language domain by applying linguistic rules of the second language to the content.

17. The system of claim 16, wherein the electronic communication comprises at least one of a transcribed voice message, an email, an instant message, and a short message service text.

18. The system of claim 16, wherein the semantic analysis results in the structural determination associated with how a question is asked in the first language.

19. The system of claim 16, wherein the syntactic analysis results in the identification of at least one of keywords, phrases, questions, and answers in the first language.

20. The system of claim 16, wherein the translation comprises a combination of the semantic analysis results, the syntactic analysis results, and application of the linguistic rules.