# US 2009/0222256 A1 · Apparatus and method for machine translation

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2009/0222256 A1
- Document ID: US20090222256A1
- Status: Published application
- Kind: A1
- Application number: 12/234,009
- Filed: 2008-09-19
- Published: 2009-09-03
- Earliest claimed: 2008-02-28
- Assignees: Kabushiki Kaisha Toshiba
- Inventors: Satoshi Kamatani; Tetsuro Chino; Kentaro Furihata
- Web page: https://patentgrep.com/patent/US20090222256A1

## CPC

- G06F 40/45: Example-based machine translation; Alignment
- G06F 40/42: Data-driven translation
- G06F 40/47: Machine-assisted translation, e.g. using translation memory
- G06F 40/55: Rule-based translation
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

In a machine-translation apparatus, an example storage unit stores therein target language examples in a target language and source language examples in a source language, while keeping the target language examples and the source language examples in correspondence with one another. An input receiving unit receives an input sentence in the source language. A searching unit conducts a search in the example storage unit for one of the target language examples corresponding to one of the source language examples that either matches or is similar to the input sentence. A translating unit generates a reverse-translated sentence by translating the one of the target language examples found in the search into the source language. A detecting unit detects a difference portion between the reverse-translated sentence and the input sentence. An output unit outputs the difference portion.

## Claims

### Claim 1

A machine-translation apparatus comprising;
an example storage unit that stores target language examples in a target language and source language examples in a source language in correspondence with one another, each of the target language examples having a meaning equivalent to a meaning of a corresponding one of the source language examples;
an input receiving unit that receives an input sentence in the source language;
a searching unit that conducts a search in the example storage unit for one of the target language examples corresponding to one of the source language examples that either matches or is similar to the input sentence;
a translating unit that generates a target language sentence by translating the input sentence into the target language, and generates a reverse-translated sentence by translating the one of the target language examples found in the search into the source language;
a detecting unit that detects difference portions between the reverse-translated sentence and the input sentence; and
an output unit that outputs the difference portions.

### Claim 2 (depends on 1)

The apparatus according to claim 1, further comprising a generating unit that generates an output sentence by adding the difference portions to the input sentence, wherein
the output unit outputs the output sentence to which the difference portions are added.

### Claim 3 (depends on 1)

The apparatus according to claim 1, further comprising a selection receiving unit that receives a selection of one of the output difference portions, wherein
the translating unit further generates the target language sentence by translating the input sentence to which the selected one of the difference portion are added into the target language, and
the output unit further outputs the target language sentence.

### Claim 4 (depends on 3)

The apparatus according to claim 3, wherein
the example storage unit further stores priority levels in correspondence with the target language examples, the priority levels indicating degrees with which the target language examples are prioritized, respectively,
the searching unit conducts a search by giving a higher priority to some of the target language examples having higher priority levels than other target language examples having lower priority levels, and
the apparatus further comprises an updating unit that raises, by a predetermined value, the priority level of the target language example from which the reverse-translated sentence containing the selected difference portion are generated.

### Claim 5 (depends on 1)

The apparatus according to claim 1, wherein
the input receiving unit further receives an input of one or more words to be added to the input sentence,
the translating unit further generates the target language sentence by translating the input sentence to which the received one or more words are added into the target language, and
the output unit further outputs the target language sentence.

### Claim 6 (depends on 1)

The apparatus according to claim 1, wherein
the example storage unit stores the target language examples and the source language examples in correspondence with one another, one or more words contained in a corresponding one of the target language examples being capable of omitting from each of the source language examples, and
the detecting unit detects, as one of the difference portions, the one or more words omitted from the one of the source language examples that either matches or is similar to the input sentence.

### Claim 7

A machine-translation apparatus comprising:
an example storage unit that stores target language examples in a target language and source language examples in a source language in correspondence with one another, each of the target language examples having a meaning equivalent to a meaning of a corresponding one of the source language examples;
an input receiving unit that receives an input sentence in the source language;
a searching unit that conducts a search in the example storage unit for one of the target language examples corresponding to one of the source language examples that either matches or is similar to the input sentence;
a first translating unit that generates a first reverse-translated sentence by translating the one of the target language examples found in the search into the source language;
a second translating unit that generates a target language sentence by translating the input sentence into the target language, and generates a second reverse-translated sentence by translating the generated target language sentence into the source language;
a detecting unit that detects a difference portion between the first reverse-translated sentence and the second reverse-translated sentence; and
an output unit that outputs the difference portion.

### Claim 8 (depends on 7)

The apparatus according to claim 7, further comprising a generating unit that generates an output sentence by adding the difference portions to the input sentence, wherein
the output unit outputs the output sentence to which the difference portions are added.

### Claim 9 (depends on 7)

The apparatus according to claim 7, further comprising a selection receiving unit that receives a selection of one of the output difference portions, wherein
the second translating unit further generates the target language sentence by translating the input sentence to which the selected one of the difference portion are added into the target language, and
the output unit further outputs the target language sentence.

### Claim 10 (depends on 9)

The apparatus according to claim 9, wherein
the example storage unit further stores priority levels in correspondence with the target language examples, the priority levels indicating degrees with which the target language examples are prioritized, respectively,
the searching unit conducts a search by giving a higher priority to some of the target language examples having higher priority levels than other target language examples having lower priority levels, and
the apparatus further comprises an updating unit that raises, by a predetermined value, the priority level of the target language example from which the reverse-translated sentence containing the selected difference portion are generated.

### Claim 11 (depends on 7)

The apparatus according to claim 7, wherein
the input receiving unit further receives an input of one or more words to be added to the input sentence,
the second translating unit further generates the target language sentence by translating the input sentence to which the received one or more words are added into the target language, and
the output unit further outputs the target language sentence.

### Claim 12 (depends on 7)

The apparatus according to claim 7, wherein
the example storage unit stores the target language examples and the source language examples in correspondence with one another, one or more words contained in a corresponding one of the target language examples being capable of omitting from each of the source language examples, and
the detecting unit detects, as one of the difference portions, the one or more words omitted from the one of the source language examples that either matches or is similar to the input sentence.

### Claim 13

A machine-translation method comprising:
receiving an input sentence in a source language;
conducting a search in an example storage unit that stores target language examples in a target language and source language examples in the source language in correspondence with one another, each of the target language examples having a meaning equivalent to a meaning of a corresponding one of the source language examples, for one of the target language examples corresponding to one of the source language examples that either matches or is similar to the input sentence;
generating a target language sentence by translating the input sentence into the target language;
generating a reverse-translated sentence by translating the one of the target language examples found in the search into the source language;
detecting a difference portion between the reverse-translated sentence and the input sentence; and
outputting the difference portion.

### Claim 14

A machine-translation method comprising;
receiving an input sentence in a source language;
conducting a search in an example storage unit that stores target language examples in a target language and source language examples in the source language in correspondence with one another, each of the target language examples having a meaning equivalent to a meaning of a corresponding one of the source language examples, for one of the target language examples corresponding to one of the source language examples that either matches or is similar to the input sentence;
generating a first reverse-translated sentence by translating the one of the target language examples found in the search into the source language;
generating a target language sentence by translating the input sentence into the target language and generating a second reverse-translated sentence by translating the generated target language sentence into the source language;
detecting a difference portion between the first reverse-translated sentence and the second reverse-translated sentence; and
outputting the difference portion.
