JP2000276393A - Method and device for collecting bilingual data and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a method and device for automatically collecting bilingual data from a document acquired from a computer network such as an Internet without executing key entry or the like and a recording medium recording a computer program for collecting the bilingual data. SOLUTION: The bilingual data collection device is provided with an input part 101 for inputting a URL and information of two language sorts, an HTML document acquisition part 103 for acquiring a document from an Internet 20 in accordance with the inputted URL, a bilingual HTML document acquisition part 102 for allowing the acquisition part 103 to acquire another document from the Internet 20 when a hyperlink to another document including the information of either one of the two inputted language sorts exists in the document concerned, a format judging part 104 for judging whether the formats of two documents coincide with each other or not, and a bilingual data output part 105 for collecting both data whose formats correspond to each other in the two documents as the bilingual data and outputting the collected bilingual data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to HTML obtained from a computer network such as the Internet.
The present invention relates to a method and apparatus for automatically collecting bilingual data from a document such as a document, and a recording medium on which a computer program for collecting bilingual data is recorded.

[0002]

2. Description of the Related Art Conventionally, bilingual data, such as a bilingual dictionary used for machine translation and a bilingual example sentence used for supporting the creation of English sentences, have been prepared by a provider of such software. On the side that provides the software, the operator inputs bilingual data such as a bilingual dictionary and a bilingual example sentence by key.
Also, with these software, the user can add the translation data. Also in this case, the user inputs the bilingual data such as the bilingual dictionary and the bilingual example sentence by a key.

[0003]

By the way, the number of documents provided by the Internet has increased due to the spread of the Internet. Since users in many countries access the Internet, an increasing number of these documents include hyperlinks to their native language documents, such as Japanese, and translated versions of these documents in foreign languages. are doing. It is conceivable to collect bilingual data from a document in which such an original sentence and a translated sentence exist, but no means has been taken.

The present invention has been made to solve such a problem, and a URL or a character string of a link destination embedded in a document such as an HTML document is input in advance. If it is a string that is related to the language type, it is presumed that the original document is likely to be a translated document, and this document is obtained, and if the format of these documents matches, the document is translated By judging that the document is a document and collecting the data of the two documents having the corresponding formats as bilingual data, it is possible to automatically convert a document obtained from a computer network such as the Internet without performing key input or the like. It is an object of the present invention to provide a method and apparatus for collecting bilingual data and a recording medium on which a computer program for collecting bilingual data is recorded.

[0005]

According to the present invention, there is provided a method for collecting bilingual data from a document obtained from a computer network.
Obtaining a first document from the computer network according to L, including in the first document information relating to one of a first language type and a second language type for which bilingual data is to be collected; , Determine whether a hyperlink to the second document exists, and if the hyperlink exists, obtain the second document from the computer network;
And determine whether the format of the second document matches,
If the formats match, the first and second documents
It is characterized in that data having a corresponding format is collected as bilingual data.

A bilingual data collection device according to the present invention is a device for collecting bilingual data from a document obtained from a computer network, wherein the bilingual data collecting device relates to a URL and a first language type and a second language type for which bilingual data is to be collected. Means for inputting information, document acquisition means for acquiring a first document from a computer network in accordance with the input URL, and a first language type and a first language type which are input in the first document. It is determined whether or not there is a hyperlink to the second document including information related to any one of the two language types. Means for acquiring the second document, means for determining whether the formats of the first and second documents match, and, if the formats match, the first and second sentences Of the format is characterized in that a means for collecting data with each other corresponds as bilingual data.

[0007] The recording medium of the present invention stores a computer program for collecting bilingual data from a document obtained from a computer network according to a URL.
On a computer-readable recording medium,
The computer may include a hyperlink to a second document including information relating to either the first language type or the second language type for which bilingual data is to be collected in the first document obtained from the computer network. Program code means for determining whether a link exists; program code means for, if the hyperlink exists, causing a computer to obtain a second document from a computer network; Program code means for determining whether or not the formats of the documents match, and if the formats match, the computer sends the data of the first and second documents having the corresponding format to each other. A computer program including program code means for collecting bilingual data is recorded.

According to the present invention, a document existing at a designated URL is obtained from a computer network such as the Internet, and a character string related to a language type is input in the obtained document in advance. That is, it is determined whether or not there is a hyperlink including a character string that can be presumed that the linked document is a translated document of the original document. If such a hyperlink exists, the linked document is obtained and the formats of the two documents are compared. If the formats match, the data in these documents whose formats correspond Is collected as bilingual data. The collected bilingual data can be output to a display, a printer, a portable recording medium, another computer, or the like.
Accordingly, bilingual data is automatically collected from a document obtained from a computer network such as the Internet, without the user performing a complicated operation such as key input.

[0009]

FIG. 1 is a functional block diagram of a bilingual data collecting apparatus according to the present invention. In the following description, the case where the computer network is the Internet will be described, but the computer network is not limited to the Internet. The bilingual data collection device 10 is realized by, for example, a personal computer and a computer program for bilingual data collection, and is connected to the Internet 20 via a public line, a dedicated line, or the like.

An input unit 101 composed of a keyboard, a mouse, etc.
Is a document obtained by translating an original document such as “English”, “Japanese”, “English”, “Japanese”, etc., related to the URL and the language type of the source language and the partner language from which the bilingual data is to be collected. This is a means for inputting a character string that can be estimated. A URL input by the input unit 101;
Translated information related to source language and language type of partner language H
This is provided to the TML document acquisition unit 102.

The bilingual HTML document acquisition unit 102 includes an input unit 10
The HTML document existing at the URL given from
The TML document acquisition unit 103 is acquired from the Internet 20. Further, the bilingual HTML document acquisition unit 102 includes an input unit 101
A character string table for determining information related to the language type of the source language and the partner language from which the bilingual data is to be collected, ie, "English", "Japanese", "English", "Japanese", etc. (FIG. 3) To be stored. Translation HT
The ML document acquisition unit 102 refers to the character string table for determination and refers to the first HTML acquired by the HTML document acquisition unit 103.
It is determined whether a hyperlink including these character strings exists in the document.

If the bilingual HTML document acquisition unit 102 determines that such a hyperlink exists in the first document, the bilingual HTML document acquisition unit 102 determines whether the second HT exists in the URL of the hyperlink.
The ML document is sent to the HTML document acquisition unit 103 via the Internet.
Then, the first and second HTML documents are passed to the format determination unit 104.

The HTML document acquisition unit 103 is made of software such as a browser, accesses the Internet 20, and receives the UR provided from the bilingual HTML document acquisition unit 102.
The HTML document existing in L is acquired and passed to the bilingual HTML document acquisition unit 102.

The format determining unit 104 determines whether or not the formats of the two HTML documents passed from the bilingual HTML document acquiring unit 102 match, as described later, by checking the type of the tag and the type of the option at the corresponding position. It is determined based on whether they are the same. The bilingual data output unit 105 collects, as the bilingual data, the data of the two HTML documents that the format judging unit 104 has determined that the formats match, as bilingual data, and displays the collected bilingual data on a display, a printer, a CD-ROM, or the like. , MO and other portable recording media, and other personal computers.

Next, the procedure of the bilingual data collection method performed by the bilingual data collection device of the present invention will be described with reference to the flowcharts of FIGS. 2 and 4, the conceptual diagram of the character string table for determination of FIG. The description will be made based on the principle diagram and a diagram showing an example of the format in FIG. First, the procedure by which the bilingual HTML document acquisition unit 102 determines whether a hyperlink to the second document obtained by translating the first document exists in the first document is described in the flowchart of FIG. 3 will be described based on a conceptual diagram of the determination character string table. It is determined whether or not the head of the tag of the HTML document is the head of a tag for providing a hyperlink to another page or another site, "<A HREF =" (step S2-1).

If the head of the tag is "<A HREF=", the character string following "<A HREF=", that is, the character string following ">" is described with reference to the determination character string table. Character string indicating link destination, file name of link destination, URL of link destination
Etc. or the character string from the following “>” to the tag “</A>”
"English""Japanese""English""Japan
It is determined whether or not a character string that can be presumed to be a document in the partner language translated from the source language document, such as "se", is included (step S2-2). If a character string is included, the bilingual HTML document acquisition unit 10
2 causes the HTML document acquisition unit 103 to acquire the linked HTML document from the Internet 20 (step S2-
3).

Next, the procedure in which the format determining unit 104 determines whether or not the formats of the two HTML documents acquired by the bilingual HTML document acquiring unit 102 match will be described with reference to the flowchart of FIG. The first tag (other than annotation tag “<!”) In each document is extracted (step S4-1),
It is determined whether there is no tag in both documents, that is, whether the document has reached the end (step S4-2).

If any document has a tag (NO in step S4-2), it is determined whether any document has a tag (step S4-3). If both documents have tags (NO in step S4-3), the types and options of both retrieved tags (auxiliary specifications such as font size)
It is determined whether or not is the same (step S4-4). If the type and option of both tags are different (NO in step S4-4), it is determined that the formats do not match (step S4-5). If there is no tag in either document (YES in step S4-3), it is determined that the formats do not match (step S4-5).

If the type and option of both tags are the same (YES in step S4-4), the next tag (other than the annotation tag) in each document is extracted (step S4).
-6), and return to step S4-2. Steps S4-2 to S4-4, S4-6
Is repeated as a result, the tag disappears in both documents, that is, when the document reaches the end (step S4-2
YES), determine that the formats match (step S4-
7).

Next, the principle of the present invention is described in Japanese and English.
A description will be given based on a principle diagram of the present invention in FIG. 5 and a diagram showing an example of a format in FIG. 6, taking a TML document as an example. URL
The first document, Japanese HTM, from “http://www.xxx”
Acquire an L document. When displaying a character string “English” that can jump to the English version (URL “http: //www.yyy”) of the Japanese HTML document, the Japanese HTML
In the L document, <A HREF="http://www.yyy"> English </A> will be described.

As shown in FIG. 6, the link source Japanese HT
ML document and English HTML of hyperlink destination
As a result of comparing each tag and option with the document,
If it can be determined that the formats match, two HTML
Data between documents enclosed in corresponding tags, "What's new" and "What's new", and "New product" and "Newpro"
ducts ”as bilingual data.

The computer program for collecting bilingual data as described above can be provided by being preinstalled in a computer, or can be provided on a portable recording medium such as a CD-ROM or MO. Further, it can be provided via a line.

[0023]

As described above, according to the bilingual data collection method, apparatus and recording medium of the present invention, the link destination URL or character string embedded in a document such as an HTML document is
If the character string is related to the type of language that has been input in advance, it is presumed that the original document is likely to be a translated document, and this document is obtained. If they match, it is determined that the document is a translated document, and 2
Since the data of two documents that have the same format are collected as bilingual data, bilingual data is automatically collected from documents obtained from a computer network such as the Internet without key input. It works.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a bilingual data collection device of the present invention.

FIG. 2 is a flowchart of a procedure for determining a hyperlink.

FIG. 3 is a conceptual diagram of a character string table for determination.

FIG. 4 is a flowchart of a format determination procedure.

FIG. 5 is a diagram illustrating the principle of the present invention.

FIG. 6 is a diagram showing an example of a format.

[Explanation of symbols]

 10 bilingual data collection device 101 input unit 102 bilingual HTML document acquisition unit 103 HTML document acquisition unit 104 format determination unit 105 bilingual data output unit 20 Internet

Claims (3)

[Claims] 1. A method for collecting bilingual data from a document acquired from a computer network, comprising: acquiring a first document from a computer network in accordance with a specified URL; Including information related to either the first language type and the second language type to be collected;
Determining whether a hyperlink to the second document exists; if the hyperlink exists, obtaining the second document from the computer network; And if the formats match, the first and second documents
A bilingual data collection method, wherein data whose formats correspond to each other is collected as bilingual data. 2. An apparatus for collecting bilingual data from a document obtained from a computer network, comprising means for inputting a URL and information relating to a first language type and a second language type for which bilingual data is to be collected. A document acquisition unit for acquiring a first document from a computer network according to the input URL; and a first language type and a second language input in the first document.
It is determined whether or not there is a hyperlink to the second document including information related to any of the language types. If the hyperlink exists, the document acquisition unit informs the document acquisition unit of the second from the computer network. Means for obtaining the second document, means for determining whether the formats of the first and second documents match, and, if the formats match,
Means for collecting data having corresponding formats as bilingual data. 3. A computer program for collecting bilingual data from a document obtained from a computer network in accordance with a URL, wherein the computer-readable recording medium stores the computer program in a computer. It is determined whether there is a hyperlink to the second document including information related to one of the first language type and the second language type for which the bilingual data is to be collected in the document. Program code means, and if the hyperlink exists, a computer
Program code means for obtaining a second document from a computer network; program code means for causing a computer to determine whether the formats of the first and second documents match; A computer-readable storage medium storing a computer program including program code means for causing a computer to collect data of formats corresponding to the first and second documents as bilingual data.