JP6564910B2 - CONVERSION DEVICE, CONVERSION METHOD, AND PROGRAM - Google Patents

Description

  The present invention relates to a conversion device, a conversion method, and a program.

  With the spread of the Internet, various information has come to be obtained. One way to use the Internet is to browse web content. The web content includes an HTML document, an image, and the like, and displays text and an image on a browser as described in the HTML document. If you use the Internet, you can send information all over the world. On websites of Japanese companies, in addition to transmitting information in Japanese, information is transmitted in languages other than Japanese such as English and Chinese. In recent years, automatic translation technology has advanced, and automatic translation has become possible.

  This is also related to accessibility assistive technologies that allow website information to be communicated to users with various characteristics. Various studies have been conducted on accessibility support technologies.

JP 2010-250776 A

  However, if the language in the web content is translated into another language so as not to change the structure of the web content, the translation of the language may greatly change the impression of the web content and make it difficult to transmit information. In addition to language translation, the same problem can arise when rephrasing difficult words easily.

  For example, in the navigation menu for web content as shown in FIG. 1, the size of the character may change when the character is modified to another language without changing the size of the menu frame. This is because even words that have the same meaning have different lengths depending on the language. As a result, it is necessary to reduce or increase the font size when attempting to express within the same menu area. For example, when a Japanese word is translated into an English word, if the English word becomes longer than the original Japanese word, it is necessary to make the characters smaller as shown in FIG. As a result, there is a problem that it is difficult to see because the characters are small.

  In addition to the size of the frame, the website conveys information in an easy-to-understand manner by balancing the navigation menu, headline, and font size of the text. For example, the heading character is made different from the text representing the body text, and the font size is made larger to make the heading stand out. As a result, a user who uses web content can understand the outline of the content by looking at the headline and read the text. In addition, the heading has a level, and a higher level heading is displayed larger so that the information structure can be easily understood.

  However, when translating languages or converting difficult words into easy words, if the headline becomes smaller than the main text, it will be difficult to figure out where the headline is. The problem arises that browsing the site becomes difficult.

  The present invention has been made in view of the above points, and an object of the present invention is to reduce a change in balance before and after conversion when a character string in content data is converted.

  Therefore, in order to solve the above-described problem, the conversion device refers to a storage unit that stores one or more conversion candidates and priorities in association with each character string for each character string for each element based on the display format in the content data. A conversion unit that selects conversion candidates, and the priority is for determining whether the conversion is adopted as a conversion result of each conversion candidate when there are a plurality of conversion candidates for the same original word. It is determined for each element based on the display format with respect to one conversion candidate as a numerical value, and for the character string having a plurality of conversion candidates, the conversion unit relates to the element based on the display format related to the character string. A conversion candidate is selected based on the priority set for each of the plurality of conversion candidates.

  When the character string in the web content is converted, a change in the balance before and after conversion can be reduced.

It is a figure which shows the example of a display when a Japanese web content is displayed with a browser. It is a figure which shows the example of a display when the web content translated from Japanese into English is displayed with a browser. It is a figure which shows the system configuration example in 1st Embodiment. It is a figure which shows the hardware structural example of the conversion apparatus in 1st Embodiment. It is a figure which shows the function structural example of the converter in 1st Embodiment. It is a flowchart for demonstrating the process sequence which the converter in 1st Embodiment performs. It is a figure which shows an example of the HTML document which comprises web content. It is a figure which shows an example of the work table in 1st Embodiment. It is a figure which shows an example of the data for translation word selection in 1st Embodiment. It is a figure which shows an example of the weighting data of the score of a translation word. It is a figure which shows the 1st example of the work table after translation of each extraction character string in 1st Embodiment. It is a figure which shows the 2nd example of the work table after translation of each extraction character string in 1st Embodiment. It is a figure which shows the example of a display of the web content after translation. It is a figure which shows the example of the work table after translation of each extraction character string in 2nd Embodiment. It is a figure which shows an example of a character modification conversion table. It is a flowchart for demonstrating the process sequence which the converter in 3rd Embodiment performs. It is a figure which shows an example of the data for translation word selection in 3rd Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 3 is a diagram illustrating a system configuration example according to the first embodiment. In FIG. 3, the conversion apparatus 10 is connected to one or more web servers 20 via a network such as the Internet or a LAN (Local Area Network).

  The web server 20 is one or more computers that provide web content.

  The conversion device 10 is an information processing device that receives and displays web content provided by the web server 20. For example, a PC (Personal Computer), a smartphone, a tablet terminal, or the like may be used as the conversion device 10.

  FIG. 4 is a diagram illustrating a hardware configuration example of the conversion device according to the first embodiment. 4 includes a drive device 100, an auxiliary storage device 102, a memory device 103, a CPU 104, an interface device 105, a display device 106, an input device 107, and the like, which are mutually connected by a bus B.

  A program for realizing processing in the conversion apparatus 10 is provided by a recording medium 101 such as a CD-ROM. When the recording medium 101 storing the program is set in the drive device 100, the program is installed from the recording medium 101 to the auxiliary storage device 102 via the drive device 100. However, the program need not be installed from the recording medium 101 and may be downloaded from another computer via a network. The auxiliary storage device 102 stores the installed program and also stores necessary files and data.

  The memory device 103 reads the program from the auxiliary storage device 102 and stores it when there is an instruction to start the program. The CPU 104 realizes functions related to the conversion device 10 in accordance with a program stored in the memory device 103. The interface device 105 is used as an interface for connecting to a network. The display device 106 displays a GUI (Graphical User Interface) or the like by a program. The input device 107 includes a keyboard and a mouse, and is used for inputting various operation instructions.

  FIG. 5 is a diagram illustrating a functional configuration example of the conversion device according to the first embodiment. In FIG. 5, the conversion device 10 includes a control unit 11, a transmission / reception unit 12, an information analysis unit 13, an information conversion unit 14, and the like. Each of these units is realized by processing in which one or more programs (for example, a web browser program) installed in the conversion apparatus 10 are executed by the CPU 104. The conversion device 10 also uses the information storage unit 15. The information storage unit 15 can be realized using the auxiliary storage device 102 or the like, for example.

  The control unit 11 controls the entire Web content display process executed by the conversion apparatus 10. The transmission / reception unit 12 transmits a request to the web server 20 and receives web content returned in response to the request. The information analysis unit 13 analyzes the web content received by the transmission / reception unit 12. For example, attribute information or the like is acquired for each HTML (HyperText Markup Language) element included in the web content. The information conversion unit 14 performs conversion processing on the character string included in the web content received by the transmission / reception unit 12. In this embodiment, an example in which a Japanese character string is converted (translated) into an English character string will be described. However, the language of the translation destination is not limited to English. Moreover, the language of the translation source is not limited to Japanese. Furthermore, the character string conversion is not limited to translation. For example, this embodiment may be applied to a case where a difficult expression word / phrase is converted into a gentle expression word / phrase. For example, when converting the contents of web content to make it larger for elderly people, or when converting Japanese with less vocabulary into content for non-native language people and elementary school students. May be.

  Hereinafter, a processing procedure executed by the conversion apparatus 10 will be described. FIG. 6 is a flowchart for explaining a processing procedure executed by the conversion apparatus according to the first embodiment.

  In step S <b> 101, the control unit 11 receives an input of a web content display instruction from the user via the input device 107. For example, the URL (Uniform Resource Locator) of the web content is input by the user.

  Subsequently, the control unit 11 acquires (for example, downloads or acquires as a response to an HTTP request) web content from the web server 20 corresponding to the input URL via the transmission / reception unit 12 (step S102). . For example, web content including an HTML document as shown in FIG. 7 is acquired.

  Subsequently, the control unit 11 records the acquired web content in the information storage unit 15 (S103).

  Subsequently, the information analysis unit 13 analyzes the web content and records the analysis result in the work table (step S104). For example, HTML elements are extracted from the web content, and the information shown in FIG. 8 is recorded for each HTML element.

  FIG. 8 is a diagram illustrating an example of a work table according to the first embodiment. In FIG. 8, the work table can store ID, element name, extracted character string, font size, number of characters, width, height, translated word, font size of translated word, and the like for each HTML element. The work table is stored in the information storage unit 15, for example.

  The ID is an identifier for identifying each extracted HTML element. However, since the ID is given for convenience of explanation, it may not be recorded as an analysis result. The element name is an HTML element name (tag name). The extracted character string is a character string extracted from the HTML element (a character string included as a value of the HTML element). The font size is the font size of the extracted character string. The number of characters is the number of characters in the extracted character string. The horizontal width and height are the horizontal width and height of the display area of the extracted character string. The font size, width, height, and the like may be specified based on, for example, a style sheet that is referenced from an HTML document. The translated word is a translation result of the extracted character string and is recorded in the subsequent processing. The font size of the translated word is the font size of the translation result of the extracted character string, and is recorded in the subsequent processing.

  Subsequently, the information conversion unit 14 determines a translation word for the extracted character string for each HTML element (selects the translation word so as to maintain the size relationship of the characters in the original language), and the font size of the translation word. Is determined (step S105).

  Specifically, the translation words are determined in order from the extracted character string having the largest font size. Therefore, in FIG. 8, a translated word is first determined for “Yokosuka International Airport” of the h1 element (ID = 1) whose font size is 18 pt. The information conversion unit 14 performs morphological analysis on “Yokosuka International Airport”, and divides “Yokosuka International Airport” into a proper noun that is a place name “Yokosuka” and a general noun “International Airport”. “Yokosuka” is a proper noun and is translated into “Yokosuka”. Since “international airport” is a general noun, a translation word is selected using data for selecting a translation word stored in the information storage unit 15 (hereinafter referred to as “translation word selection data”). .

  FIG. 9 is a diagram illustrating an example of translation word selection data according to the first embodiment. In FIG. 9, translation word candidates and priority levels are recorded in correspondence with each translation target word (original word) in the translation word selection data. A translation word that is a plurality of conversion words may be associated with one translation target word (original language).

  A translated word is an English expression of the original language. The priority is a numerical value for determining the superiority or inferiority of each candidate when there are a plurality of translation word candidates for the same original word. In the present embodiment, a smaller value has a higher priority. In FIG. 9, a priority is set for each HTML element. That is, the priority of a certain translation word varies depending on the HTML element from which the original word is extracted.

  For example, in FIG. 9, the records having “International Airport” as the original language are the record with the number 4 and the record with the number 5. “International Airport” and “Int. Airport”, which are translations of these records, are candidates for translation of “International Airport”. Therefore, there are two translation candidates for “Yokosuka International Airport”: “Yokosuka International Airport” and “Yokosuka Int. Airport”. From FIG. 8, “Yokosuka International Airport” is an extracted character string from the h1 element (that is, heading). Therefore, the information conversion unit 14 specifies the priority of the translation word corresponding to the h1 element with reference to the translation word selection data (FIG. 9). In the translation word selection data, when the number of characters of the translation word (word after conversion) is small relative to the original word, the priority of the upper heading in the web content is higher than the priority of the lower heading, and The priority of the heading is set to be higher than the priority of the body (for example, p element or a element). On the other hand, when the number of characters in the translated word is larger than the original word, the priority of the text is higher than the heading, and the priority at the lower heading is set to be higher than the priority at the upper heading. Yes.

  In FIG. 9, the priority of “Int. Airport” is 22 and the priority of “International Airport” is 53 for the h1 element. On the other hand, the translated word of the proper noun “Yokosuka” is not included in the translation word selection data, but in that case, it may be matched with the priority of “International Airport” and the priority is 0. It may be said. That is, the priority between “Yokosuka International Airport” and “Yokosuka Int. Airport” follows the priority between “Int. Airport” and “International Airport”. Accordingly, “Yokosuka Int. Airport” having a high priority is selected as the translated word for “International Airport”. The information conversion unit 14 records “Yokosuka Int. Airport” as a translated word of the record with ID = 1 in the work table of FIG.

  The “international airport” may be further divided into “international” and “airport” for translation. In this case, translation candidates of “international” are “International” in the record of number 1 in FIG. 9 and “Int.” In the record of number 2. Regarding the h1 element, the priority of “International” is 120, and the priority of “Int.” is 28. Therefore, “Int.” Is selected. On the other hand, in FIG. 9, the translation word for “airport” is only “Airport”. Therefore, the translated word of “International Airport” may be determined as “Int. Airport”. In this case, the translation of “Yokosuka International Airport” is determined as “Yokosuka Int. Airport”.

  The priority may be determined in advance based on, for example, the closeness of the font size to the word (original language) before translation and the ease of understanding the translated word. The closeness of the font size to the original word is a difference in font size between the original word and the translated word when the original word and the translated word are stored in the same display area (for example, the original word display area). For example, “Int. Airport” has a character string length that is closer to “International Airport” than “International Airport”, so the font size is close to the original language but is inferior in terms of ease of understanding.

When each of the closeness of the font size and the ease of understanding the translated word is a score x and a score y, the priority p may be calculated by the following equation (1), for example.
p = (fx + cy) / (f + c) (1)
Note that x is a score (numerical value) that becomes smaller as the font size is closer. Further, y is a score (numerical value) that decreases as the value becomes easier to understand. F is weighting data for the closeness of the font size, and c is weighting data for easy understanding of the translated word. The weighting data may be different for each HTML element. For example, in the h1 element, the proximity of the font size to the original word is more important than the ease of understanding the translated word. Although the font size balance between HTML elements is important, the h1 element having the largest font size serves as a reference for balance, and therefore the font size is particularly important. Therefore, for the h1 element, f may be made larger than c. FIG. 10 shows an example of the weighted data of the translation word score.

  In FIG. 10, the value of f regarding the p element is 0 and the value of c is 10. The p element indicates a paragraph, but there are cases in which the length of a character string included in a paragraph is not limited in web content. In such a case, even if the translation result becomes longer than before translation, only the display area of the paragraph is expanded, and the display balance of the web content is not adversely affected. Thus, the p element may use the same font size as the original word. Then, for the p element, the font size of the translated word is always the same as the font size of the extracted character string. That is, the priority of the translation word for the p element depends only on the ease of understanding. Therefore, in the example of FIG. 10, the value of f regarding the p element is 0, and the value of c is 10.

  Next, the h2 element having the second largest font size is translated. That is, the “flight information” with ID = 2 in FIG. 8 is translated. According to the translation word selection data of FIG. 9, as translation word candidates corresponding to the “flight information” of the h2 element, “Flight Info” with a priority of 51 and “Flight Information” with a priority of 82 It is. Therefore, “Flight Info” having a high priority is selected.

  By selecting the converted word (converted word) based on the priority in the translation word selection data, the heading of the original word can be easily converted to a short (small number of characters) word, and the text is long (number of characters) Easily converted to words. As a result, a conversion word with a small number of characters is likely to be displayed with relatively large characters, and a body with a large number of characters is likely to be displayed with relatively small characters, so that the size relationship of the characters in the original language can be easily maintained.

  Similarly, the information conversion unit 14 determines a translated word of each extracted character string, and adds the determined translated word to the work table. For example, “Flight Information” included in “Today's flight information is ...” that is the extracted character string of ID = 3 in FIG. 8 is given priority as “Flight Info” with a priority of 42 as a translation word candidate. “Flight Information” having a degree of 15 is selected, but “Flight Information” having a high priority is selected. Since the extracted character string with ID = 3 is extracted from the p element, the priority may be determined without considering the closeness of the font size as described above. Good.

  The translation word selection data is basically used for selecting a translation word for a word. When the extracted character string is not a word but a sentence, the translation according to the grammar of the translation destination language may be performed using a known translation technique.

  When the translation is completed for all the extracted character strings, the work table is updated as shown in FIG. 11, for example. In FIG. 11, the translated word and the font size of the translated word are added to each extracted character string. As the font size of the translation word, for example, the maximum font size is selected in a range in which the translation word fits in a rectangular area specified by the horizontal width and height of the extracted character string. However, for the p element, as described above, the font size before and after translation may be the same.

  In addition, the translation word of the extracted character string extracted from the p element with the restriction | limiting in the length of a paragraph may be determined as follows.

  It is desirable that the length of the translation result of the p element, which is limited in the length of the paragraph (the size of the display area), be close to the length before translation. On the other hand, since the content of the p element is a sentence, various words may be used. Therefore, as a result of selecting a translated word having a high priority for each word, it is conceivable that the entire paragraph has a larger display area than before translation. In this case, the length of the entire paragraph may be adjusted by replacing a translation word having a relatively low priority with another translation word having a common original word.

  For example, if the translation result paragraph contains “International Airport” and “Access Information” and if one of them is shortened, it will be closer to the length of the pre-translation paragraph. Word priorities are compared. Regarding the p element, the priority of “International Airport” is 23, and the priority of “Access Information” is 12. Therefore, “Access Information” with a high priority may be adopted, and “International Airport” with a low priority may be replaced with “Int. Airport”.

  In addition, the information conversion unit 14 corrects the font size of each translated word so that the size relationship of each character string before translation is maintained. For example, the information conversion unit 14 groups the HTML elements before translation for each font size of the extracted character string. The information conversion unit 14 performs adjustment so that the difference in font size between groups after translation is as close as possible to the difference in font size between groups before translation. By doing so, the balance between HTML elements before translation can be maintained even after translation.

  In this case, first, the maximum font size among the font sizes of the translated words may be used as a reference. For example, in FIG. 11, the maximum font size of the translated word is 16 pt. In this case, the font size of the HTML element classified into the same group as the HTML element related to the font size (hereinafter referred to as “first group”) is corrected to 16 pt. Next, the font size of the translated word of each HTML element belonging to the group with the next largest font size (hereinafter referred to as “second group”) is the font size before translation of the first group and the font before translation of the second group. It is corrected to a value obtained by subtracting the difference from the size from 16 pt. For example, in FIG. 11, the font size of the h1 element before translation is 18 pt, the font size of the h2 element before translation is 14 pt, and the difference is 4 pt. Therefore, in this case, the font size of the translation word of the h2 element is 16pt-4p5 = 12pt. An example of the work table in this case is shown in FIG. In the work table shown in FIG. 12, a column of font size difference from the upper group is added. The upper group is a previous group when the groups are sorted in descending order of font size. In FIG. 12, the font size value of the translated word is corrected by the font size difference from the upper group.

  Note that the font size of the translated word may be corrected so that the ratio of the font sizes between groups is maintained, not the difference in font size between groups.

  Further, the font size may be corrected within a limit range in which the translated word fits in the display area before translation (the width and height of the extracted character string). For example, when the font size of the second group is corrected to 12 pt, if the translated word of any HTML element belonging to the second group does not fit in the width and height of the HTML element, the translated word of the HTML element is The maximum font size within a range that fits in the horizontal width and height of the HTML element may be the second group font size.

  Subsequently, the information conversion unit 14 rewrites the HTML document of the web content recorded in the information storage unit 15 based on the work table shown in FIG. 11 or FIG. 12 (step S106). Specifically, the value (character string) of each HTML element is replaced with the translated word. In addition, an attribute indicating a font size is added to each HTML element. At this time, the font size value of the translated word in the work table is used as the font size value.

  Subsequently, the control unit 11 displays the web content recorded in the information storage unit 15 on the display device 106 (step S107). As a result, the web content after translation as shown in FIG. 13 is displayed. Note that without replacing step S106 (that is, without rewriting the HTML document), replacement with a translated word, change of the font size, or the like may be performed in the process of display processing in step S107.

  As described above, according to the first embodiment, translation (conversion) is performed so that the magnitude relationship between HTML elements before translation (before conversion) is maintained. Therefore, when the character string in the web content is converted, the change in the balance before and after conversion can be reduced.

  In the above example, one or more converted word (word) candidates are stored in association with the storage unit for the conversion target word (word). Is not limited to this method. For example, the conversion device 10 may transmit a word (word) to be converted to an external conversion device via the transmission / reception unit 12 and receive one or more converted words (words) as a response. good.

  In the above example, an input of a web content display instruction is received from the user via the input device 107, but another method may be used to acquire information on the web content to be converted. For example, when a user terminal (not shown) transmits a content browsing request to the web server, the conversion device sets the conversion device as a proxy so that the conversion device transmits and receives an HTTP request transmitted by the user terminal. By sending an HTTP request to the URL obtained from the user terminal, the web content is obtained from the web server, and the obtained web content is converted to the user terminal (to the HTTP request). As a response).

  In the above example, the acquired web content, work table, and translation word selection data are stored in the information storage unit 15. However, the translation word selection data that has a large amount of data and is read by the main access is as follows. It is also possible to use different storage units, such as storing web contents and work tables that are stored in a database and have a smaller amount of information than the translated word selection data and read / written at a high frequency in a memory.

  In the above example, the size relationship between characters is determined using a tag attached to web content, but if a tag (element name such as h1, p) is not attached, existing characters are used. It is also possible to estimate the size of each character (eg, font size) using a clipping technique and use the font size information as an alternative to HTML tag information. Further, when the font size type is extracted more than a predetermined threshold value, the font sizes may be clustered according to a predetermined rule (an example of a predetermined rule: a font size of 32 points or more corresponds to the HTML tag h1, The font size from 24 to 31 points corresponds to h1).

  Further, in the above example, the priority in the translation word selection data is given a smaller value as the priority should be given, but it is also possible to take a larger value as the priority should be given (in this case, the priority is given). The evaluation formula is also changed correspondingly).

  Also, the conversion device 10 converts the conversion type (example 1: translation from Japanese into English, example 2: converts a technical document in Japanese into easy-to-understand Japanese), and destination information of each external conversion device. Are stored in the storage unit in association with each other, receive an instruction for specifying the conversion type, specify a conversion device according to the conversion type, transmit the conversion target word to the conversion device, and convert the result. By obtaining the above, various types of conversion may be realized.

  Next, a second embodiment will be described. In the second embodiment, differences from the first embodiment will be described. Points that are not particularly mentioned in the second embodiment may be the same as those in the first embodiment.

  In the first embodiment, based on the font size of the translated word for the extracted character string having the largest font size, the font size of the translated word having a smaller font size is sequentially determined. However, even if the font size is the same, it is possible to make the characters look larger by modifying the characters, such as using bold or different fonts. In the second embodiment, an example in which the magnitude relationship between HTML elements is maintained using a change (modification) of character expression will be described. For example, when converting a character string with a different font size into a translated word with a similar font size, the character string with a relatively large font size is more prominent than a character string with a relatively small font size in the content data. Conversion of the column representation is performed.

  In the second embodiment, at the end of step S105, for example, the work table is updated as shown in FIG.

  FIG. 14 is a diagram illustrating an example of a work table after translation of each extracted character string in the second embodiment. In the work table shown in FIG. 14, columns such as font modification of the translation word and corrected font size are added to FIG.

  Font modification is a modification method for the extracted character string. The font modification of the translation word is a modification method determined to be given to the translation word. The corrected font size is the corrected font size for the translated word. For translated words that are modified, the effect of the modification is also added to the corrected font size.

  The values in the respective columns of the font modification and the corrected font size of the translated word are recorded when the information conversion unit 14 corrects the font size of each translated word in step S105. That is, in the second embodiment, correction of the font size of the translated word is performed not only by changing the font size itself but also by modifying the translated word.

  The modification or the like given to each translated word may be performed with reference to a table as shown in FIG.

  FIG. 15 is a diagram illustrating an example of the character modification conversion table. In the character modification conversion table shown in FIG. 15, a font size conversion value and a priority order are set for each character modification method. The font size conversion value is a value indicating the magnitude of the influence or action on the font size by the corresponding modification method, and has a format of + N (or -N). + N indicates that the font size has an effect of increasing Npt. The priority order indicates a priority order when selecting a modification method for a translated word. In the priority order, a modification method having a relatively large font size conversion value (a relatively large font size enlargement effect) is ranked high. The character modification conversion table is stored in the information storage unit 15 in advance, for example.

  First, the information conversion unit 14 first fills in the difference between the font size of the translated word of the first group (the group having the largest font size before translation) and the font size before translation. Determine the modification method for.

  For example, in FIG. 14, the font size of “Yokosuka International Airport”, which is an extracted character string from the h1 element, is 18 pt, and the font modification is Gothic. The font size of “Yokosuka Int. Airport”, which is a translated word of “Yokosuka International Airport”, is 14 pt. Therefore, the information conversion unit 14 searches the character modification conversion table for a modification method that can fill in the difference of 18pt-14pt = 4pt. At this time, application to the translation word is attempted in order from the modification method having the highest priority.

  First, “bold”, which is a modification method with a priority of 1, is selected. The font size conversion value of the modification method is +3, which is less than 1 pt for the required 4 pt. Therefore, “Gothic”, which is a modification method with a priority of 2, is selected. The font size conversion value of the modification method is +1. Then, by combining “bold” and “gothic”, an effect of 4 pt can be obtained. Therefore, the font modification value of the translated word of the h1 element is “bold + gothic”.

  In the second group and thereafter, the correction font size and the modification method are determined based on the correction font size of the upper group. At this time, it is assumed that the font size of the translated word cannot be larger than the font size of the translated word recorded in the work table. This is because the maximum font size that can be displayed in the range of the display area of the original word is selected as the font size of the translated word. When the font size of a translation word of a certain group and the corrected font size and the difference of the upper group are less than the font size difference between the groups before translation, the translation of the group is modified. For example, the difference in font size before translation between the h1 element of the first group and the h2 element of the second group is 4 pt. Here, the font size of the translated word of the h2 element is 14 pt. In this case, by adopting 14pt as it is, the difference from the corrected font size of the translated word of the h1 element can be set to 4pt. Therefore, for the h2 element, the font modification of the translated word is “none”, and the correction font size is 14 pt.

  The difference in font size before translation between the h2 element of the second group and the h3 element of the third group is 2 pt. Here, the font size of the translated word of the h3 element is 12 pt. In this case, by adopting 12pt as it is, the difference from the corrected font size of the translated word of the h2 element can be set to 2pt. Therefore, for the h3 element, the font modification of the translated word is “none”, and the correction font size is 12 pt.

  On the other hand, the difference in font size before translation between the h3 element of the third group and the p element of the fourth group is 1 pt. Here, the font size of the translated word of the p element is 9 pt. In this case, if 9pt is adopted as it is, the difference between 12pt, which is the third group correction font size, cannot be made 1. That is, it is necessary to increase the font size of the p element by 2 pt. Therefore, the information conversion unit 14 searches for a modification method for 2 pt with reference to the character modification conversion data. As a result, “gothic” with a priority of 2 and “capital letters” with a priority of 3 are selected. This is because the total font size conversion value of the two selection methods is +2 pt. Accordingly, “gothic + capital letters” is recorded in the font modification of the translated word of the p element. The corrected font size of the p element is 9 + 2 = 11 pt. Note that the capital letter means, for example, “FLIHT INFO” for “Flight Info”.

  As described above, according to the second embodiment, the characters of the translated word are given so that the HTML element belonging to the first group as a reference can be given an impression of the same font size as the font size before translation. Qualification is done on the column. For groups below the second group, the font size of the translated word is corrected based on the first group. Even if the translated word font size cannot be increased to fit in the display area of the extracted character string, the font size can be increased in a pseudo manner by font modification. Therefore, in the second embodiment, the change in the balance before and after translation can be further reduced as compared with the first embodiment.

  Note that the font size conversion value is not limited to a positive value, and changes in expression that make it inconspicuous (for example, changing the color of a character to gray, thinning, etc.) are associated with a negative conversion value. Also good.

  Next, a third embodiment will be described. In the third embodiment, differences from the first embodiment will be described. Points not particularly mentioned in the third embodiment may be the same as those in the first embodiment.

  In the first embodiment and the second embodiment, the method of selecting a translation word according to the tagged HTML element has been described. However, there are many web contents that give meaning to character strings even though they are not semantically tagged. For example, some web contents are made to appear as a headline by visually specifying where the h1 element should be used for the headline and specifying the font size in the style sheet using the div element or the like. It is difficult to apply the first embodiment to such web content. Originally, all web content must be marked up with the correct elements to support accessibility, but in the third embodiment, it is necessary to support web content that does not consider accessibility. aim. Specifically, the translation word is determined based on the font size without depending on the HTML element.

  FIG. 16 is a flowchart for explaining a processing procedure executed by the conversion apparatus according to the third embodiment. In FIG. 16, the same steps as those in FIG. 6 are denoted by the same step numbers, and the description thereof is omitted.

  In FIG. 16, step S105 is replaced with step S105a. In step S105a, the information conversion unit 14 determines translated words in order from the extracted character string having the largest font size. At this time, the information conversion unit 14 refers to translation word selection data as shown in FIG. That is, in the third embodiment, the configuration of the translation word selection data is different from that of the first embodiment (FIG. 9).

  FIG. 17 is a diagram illustrating an example of translation word selection data according to the third embodiment. In FIG. 17, the translation word selection data includes translation word candidates and priorities for each word (original language) to be translated. A significant difference from FIG. 9 is that the priority is defined for each font size range, not for each HTML element. In the expression of the font size range, “Xpt to Ypt” indicates Xpt or more and less than Ypt. In FIG. 17, the priority is defined for each font size range, but may be further subdivided to define the priority for each font size.

  Here, determination of the translation word for the extracted character string (“flight information”) with ID = 2 in FIG. 8 will be described. Note that since the font size of the extracted character string with ID = 2 is second, the translated word of the extracted character string is determined second. The work table (FIG. 8) according to the third embodiment may not include the element name column.

  According to FIG. 17, the translation word candidates of “flight information” are “Flight Info” and “Flight Information”. On the other hand, according to FIG. 8, the font size of the extracted character string with ID = 2 (“flight information”) is 14 pt. Therefore, the priority of each candidate is specified with reference to the column “14pt to 16pt” corresponding to 14pt which is the font size of the extracted character string in FIG. As a result, it is specified that the priority of “Flight Info” is 42 and the priority of “Flight Information” is 55. Since “Flight Info” has a higher priority, “Flight Info” is selected as a translated word of “Flight Information”. Further, the font size of “Flight Info” is determined based on the width and height of “Flight Information”.

  Next, the determination of the translation word of “flight information” included in “Today's flight information is ...” that is the extracted character string of ID = 3 in FIG. 8 will be described. According to FIG. 17, the translation word candidates of “flight information” are “Flight Info” and “Flight Information”. On the other hand, according to FIG. 8, the font size of the extracted character string with ID = 3 is 12 pt. Accordingly, the priority of each candidate is specified with reference to a column of “less than 14 pt” corresponding to 12 pt which is the font size of the extracted character string in FIG. As a result, it is specified that the priority of “Flight Info” is 44 and the priority of “Flight Information” is 23. Since “Flight Information” has a higher priority, “Flight Information” is selected as a translated word of “Flight Information”.

  Similarly, the translated word of each extracted character string is determined, and the work table is updated as shown in FIG. However, the element name column is not necessary. Further, the font size of each translated word is corrected based on the difference in font size between groups based on the font size.

  Subsequently, the translated web content is displayed as in the first embodiment (step S107).

  As described above, according to the third embodiment, the same effects as those of the first embodiment can be obtained even for web content that is not marked up with correct elements and does not support accessibility. be able to.

  Note that the second embodiment may be combined with the third embodiment.

  Each of the above embodiments may be applied to content data other than web content.

  In the present embodiment, the information conversion unit 14 is an example of a conversion unit. A translation word candidate is an example of a conversion candidate. The font size is an example of a display size.

  As mentioned above, although the Example of this invention was explained in full detail, this invention is not limited to such specific embodiment, In the range of the summary of this invention described in the claim, various deformation | transformation・ Change is possible.

DESCRIPTION OF SYMBOLS 10 Conversion apparatus 11 Control part 12 Transmission / reception part 13 Information analysis part 14 Information conversion part 15 Information storage part 20 Web server 100 Drive apparatus 101 Recording medium 102 Auxiliary storage apparatus 103 Memory apparatus 104 CPU
105 Interface device B bus

Claims (8)

For each character string for each element based on the display format in the content data, it has a conversion unit that selects conversion candidates with reference to a storage unit that stores one or more conversion candidates and priorities in association with each original word,
The priority is displayed for one conversion candidate as a numerical value for determining whether the conversion is adopted as a conversion result of each conversion candidate when there are a plurality of conversion candidates for the same source language. It is defined for each element based on the format,
The conversion unit selects a conversion candidate for a character string having a plurality of conversion candidates based on a priority set for each of the plurality of conversion candidates with respect to an element based on the display format related to the character string. ,
A conversion device characterized by that. For conversion candidates with a small number of characters relative to the original language, the priority of the element representing the upper heading in the content data is higher than the priority of the element representing the lower heading, and the priority of the element representing the heading is the text. It is set to be higher than the priority of the element representing, and for other conversion candidates, the priority of the element representing the body is higher than the element representing the heading, and the element representing the lower heading The priority is set to be higher than the priority of the element representing the top headline,
The conversion device according to claim 1. For each character string for each element based on the display format in the content data, it has a conversion unit that selects conversion candidates with reference to a storage unit that stores one or more conversion candidates and priorities in association with each original word,
The priority is a display size for one conversion candidate as a numerical value for determining whether or not to adopt as a conversion result of each conversion candidate when there are a plurality of conversion candidates for the same source language. For each range of
For the character string having a plurality of conversion candidates, the conversion unit selects a conversion candidate based on the priority set for each of the plurality of conversion candidates with respect to the range including the display size of the character string.
A conversion device characterized by that. The conversion unit has an overall display area of the plurality of conversion candidates selected for each of the plurality of character strings of the source language larger than the display area of the plurality of character strings of the source language, and the plurality of characters of the source language For any one or more character strings in the column, when another conversion candidate having a smaller number of characters than the selected conversion candidate is stored in the storage unit, the conversion candidate selected for each of the one or more character strings Replace a conversion candidate with a relatively low priority with another conversion candidate.
The conversion device according to claim 1, wherein the conversion device is characterized in that When the display size of the source language character string is different from the display size of the conversion candidate selected for the source language character string, the conversion unit includes at least a font, a character thickness, and a character color. For each of the modification methods, a second value that stores a conversion value indicating the influence or effect of the modification method on the display size and a priority order in which the modification method having a relatively large conversion value is higher is stored. With reference to the storage unit, for the conversion candidates selected for the original language character string, one or more such that the total of the converted values is the difference between the display size of the original language character string and the display size of the conversion candidate Selecting the modification method based on the priority,
The conversion device according to claim 1 or 3, wherein Computer
For each character string for each element based on the display format in the content data, execute a conversion procedure for selecting a conversion candidate with reference to a storage unit that stores one or more conversion candidates and priorities in association with each original word,
The priority is displayed for one conversion candidate as a numerical value for determining whether the conversion is adopted as a conversion result of each conversion candidate when there are a plurality of conversion candidates for the same source language. It is defined for each element based on the format,
In the conversion procedure, for a character string having a plurality of conversion candidates, a conversion candidate is selected based on a priority set for each of the plurality of conversion candidates with respect to an element based on the display format related to the character string. ,
A conversion method characterized by that. Computer
For each character string for each element based on the display format in the content data, execute a conversion procedure for selecting a conversion candidate with reference to a storage unit that stores one or more conversion candidates and priorities in association with each original word,
The priority is a display size for one conversion candidate as a numerical value for determining whether or not to adopt as a conversion result of each conversion candidate when there are a plurality of conversion candidates for the same source language. For each range of
In the conversion procedure, for a character string having a plurality of conversion candidates, a conversion candidate is selected based on a priority set for each of the plurality of conversion candidates with respect to the range including the display size of the character string.
A conversion method characterized by that.   A program that causes a computer to function as the conversion unit according to any one of claims 1 to 5.

Images