Using business process model to create machine translation dictionaries

Patent metadata

Field	Value
Publication number	US-9720910-B2
Application number	US-201514938493-A
Country	US
Kind code	B2
Filing date	Nov 11, 2015
Priority date	Nov 11, 2015
Publication date	Aug 1, 2017
Grant date	Aug 1, 2017

Abstract

An approach is provided to receive a term that is included in a Business Process Model (BPM) data store with the term being from one natural language. The approach identifies that first descriptive text of the term is not available in the same natural language. A translated version of the term is retrieved from a different natural language stored in the BPM data store with descriptive text of the term being present in the different language. The descriptive text is translated to the given natural language, resulting in translated descriptive text that is, in turn, provided as a meaning of the term in the given language.

First claim

The invention claimed is:

1. A method implemented by an information handling system that includes a memory and a processor, the method comprising:
receiving a term that is included in a Business Process Model (BPM) data store, wherein the term is from a first natural language, and wherein the term is included in a natural language question from a requestor to a question answering (QA) system, and wherein the BPM data store is included in a corpus utilized by the QA system;
identifying that a first descriptive text of the term in the first natural language is lacking;
retrieving a translated version of the term in a second natural language stored in the BPM data store, wherein a second descriptive text of the translated version of the term is present in the second natural language;
translating the second descriptive text from the second natural language to the first natural language, resulting in a translated descriptive text;
providing, to the QA system, the translated descriptive text as a meaning of the term in the first natural language; and
answering, by the QA system, the natural language question from the requestor, wherein the answering comprises:
searching the corpus for a plurality of candidate answers to the natural language question;
scoring the plurality of candidate answers, wherein the scoring indicates an extent to which each of the plurality of candidate answers is inferred by the natural language question, wherein the scoring further comprises weighting the plurality of scored candidate answers using a machine generated statistical model, wherein the statistical model indicates a level of confidence of the QA system, the weighting resulting in weighted scores corresponding to each of the candidate answers; and
providing one or more of the candidate answers based on the weighted scores.

2. The method of claim 1 further comprising:
updating the BPM with the translated descriptive text, wherein the translated descriptive text is stored in the BPM as the first descriptive text and wherein the first descriptive text is associated with the term.

3. The method of claim 1 further comprising:
receiving, at the QA system, a question expression from the requestor;
identifying that the term is a business related term that is included in the received question expression;
receiving, at the QA system, the provided translated descriptive text as a context of the question expression; and
wherein the answering utilizes the context of the question expression.

4. The method of claim 1 wherein the BPM data store includes a plurality of natural languages that include the first and second natural languages, and wherein the method further comprises:
searching through one or more of the plurality of natural languages to identify one of the natural languages with a stored term corresponding to the received term and a stored descriptive text that describes the stored term, wherein the identified natural language is used as the second natural language.

5. The method of claim 1 further comprising:
identifying that the term is lacking from the BPM data store;
translating the term to the second natural language forming a translated version of the term; and
searching the BPM data store for the translated version of the term in the second natural language.

6. The method of claim 1 further comprising:
harvesting a set of BPM-based pairs of labels and descriptions in the first and second languages; and
updating the BPM data store with the translated descriptive text.

7. An information handling system comprising:
one or more processors;
one or more data stores accessible by at least one of the processors;
a memory coupled to at least one of the processors; and
a set of computer program instructions stored in the memory and executed by at least one of the processors in order to perform actions of:
receiving a term that is included in a Business Process Model (BPM) data store, wherein the term is from a first natural language, and wherein the term is included in a natural language question from a requestor to a question answering (QA) system, and wherein the BPM data store is included in a corpus utilized by the QA system;
identifying that a first descriptive text of the term in the first natural language is lacking;
retrieving a translated version of the term in a second natural language stored in the BPM data store, wherein a second descriptive text of the translated version of the term is present in the second natural language;
translating the second descriptive text from the second natural language to the first natural language, resulting in a translated descriptive text; and
providing, to the QA system, the translated descriptive text as a meaning of the term in the first natural language; and
answering, by the QA system, the natural language question from the requestor, wherein the answering comprises:
searching the corpus for a plurality of candidate answers to the natural language question;
scoring the plurality of candidate answers, wherein the scoring indicates an extent to which each of the plurality of candidate answers is inferred by the natural language question, wherein the scoring further comprises weighting the plurality of scored candidate answers using a machine generated statistical model, wherein the statistical model indicates a level of confidence of the QA system, the weighting resulting in weighted scores corresponding to each of the candidate answers; and
providing one or more of the candidate answers based on the weighted scores.

8. The information handling system of claim 7 wherein the actions further comprise:
updating the BPM with the translated descriptive text, wherein the translated descriptive text is stored in the BPM as the first descriptive text and wherein the first descriptive text is associated with the term.

9. The information handling system of claim 7 wherein the actions further comprise:
receiving, at the QA system, a question expression from the requestor;
identifying that the term is a business related term that is included in the received question expression;
receiving, at the QA system, the provided translated descriptive text as a context of the question expression; and
wherein the answering utilizes the context of the question expression.

10. The information handling system of claim 7 wherein the BPM data store includes a plurality of natural languages that include the first and second natural languages, and wherein the actions further comprise:
searching through one or more of the plurality of natural languages to identify one of the natural languages with a stored term corresponding to the received term and a stored descriptive text that describes the stored term, wherein the identified natural language is used as the second natural language.

11. The information handling system of claim 7 wherein the actions further comprise:
identifying that the term is lacking from the BPM data store;
translating the term to the second natural language forming a translated version of the term; and
searching the BPM data store for the translated version of the term in the second natural language.

12. The information handling system of claim 7 wherein the actions further comprise:
harvesting a set of BPM-based pairs of labels and descriptions in the first and second languages; and
updating the BPM data store with the translated descriptive text.

13. A computer program product stored in a computer readable storage medium, comprising computer program code that, when executed by an information handling system, causes the information handling system to perform actions c

Assignees

• IBM

Inventors

• Byron Donna K
• Krishnamurthy Lakshminarayanan
• Sinha Ravi S
• Trim Craig M

Classifications

• G06F40/242

  Dictionaries · CPC title

• G06F40/58 Primary

  Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation · CPC title

• G06F16/3329 Primary

  Natural language query formulation · CPC title

• G06F17/2735

  Physics · mapped topic

• G06F17/289 Primary

  Physics · mapped topic