JP2007233752A - Retrieval device, computer program and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To efficiently retrieve data more close to a retrieval target of a user. <P>SOLUTION: This retrieval device retrieving the retrieval target from a database 31 on the basis of a designated retrieval term or retrieval sentence, and outputting a retrieval result has: a retrieval requirement analysis means 22 analyzing inputted retrieval requirement to set a retrieval condition; a retrieval means 32 searching the database 31 on the basis of the retrieval condition set by the retrieval requirement analysis means 22, and acquiring a data set of the retrieval result; and a score calculation means 33 calculating a score for performing desired ordering to individual data of the data set acquired by the retrieval means 32. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a search device that searches electronic data stored in a storage means according to necessity or importance, a computer program executed by the search device, and a recording medium on which the computer program is recorded.

  Search technology for electronic data or search result display technology has become an increasingly important technology because of the increase in the number of search results due to an increase in the amount of information to be searched. This is because the information that is sought is buried in a large amount of search results, making it difficult to find. As for such a search, browsing and a list are functions used when a user specifies electronic data to be referred to from electronic data stored in a storage device that stores the data. Browsing is a method in which a user performs an operation that follows a stored electronic data (an electronic document—the same applies hereinafter) using a hierarchical structure or a link to reach the target electronic data. This method corresponds to, for example, an operation of tracing a folder hierarchy with an Explorer of Windows (registered trademark) or an operation of following a link from a document on the WWW by a Web browser. In this method, for example, when there is a large amount of electronic data in the same hierarchy, there is a problem that it is difficult to reach the target electronic data.

  On the other hand, a list is a method in which a user designates a condition for a set of electronic data, enumerates electronic documents that satisfy the condition, and searches for desired electronic data from the list. This method corresponds to a function provided by a search engine on the WWW or a file search in Windows (registered trademark) Explorer, for example. In this method, for example, when there is a large amount of electronic data in the same hierarchy, there is a problem that it is difficult to reach the target electronic data. A technique for solving the problems in these two methods by rearranging is generally known.

  On the other hand, as an alternative to these two methods, a method using a position where electronic data is arranged on the screen, a set of electronic data satisfying conditions specified by the user are arranged on a two-dimensional plane, There are a method in which a high document is arranged nearby, or a method in which the distribution of data is projected onto a two-dimensional plane using two of the attribute values associated with the data to be searched.

As a more advanced method, instead of simply displaying a list of information that satisfies the search conditions, the search results are ranked by relevance or displayed, or the search results are displayed in multiple dimensions for viewing. There is also a thing which raises property (refer patent documents 1 and 2). Alternatively, a search method using a plurality of ordering methods other than multi-dimensional display is also known (see Patent Document 3). Furthermore, a method of re-searching using a limited set is also known (see Patent Document 4). In this method, the degree of association between a word and a search word is calculated using the document set obtained by the first search and the document set not obtained by the first search, and the search is performed again.
JP 2002-297661 A JP 2005-56081 A Japanese Patent No. 3347088 JP 2002-288215 A

  However, in the multi-dimensional display as disclosed in Patent Documents 1 and 2, the newly added dimension is simply displayed in the order of increasing number or ascending order, with no score or with a score. However, it was fixed and was not an adequate display method. It is clear that the former is inferior in viewability compared to the scored one, and the latter is insufficient because it does not use the set of search results obtained from the first search result Met. Because this set is a set limited by a certain search condition, if scoring using this set, compared to fixed scoring, that is, scoring using the whole set, This is because such processing is not performed even though more suitable scoring is possible.

  Further, in the method disclosed in Patent Document 3, a new ordering is defined by searching by a plurality of ordering methods and combining the search results, but what set is used for each ordering method. There is no mention of what to do.

  Furthermore, in the method of re-searching using a limited set as disclosed in Patent Document 4, the obtained relevance is only used for selecting keywords for re-searching, and the like. It does not give the score of the attribute or the attribute of each document. Moreover, the object to be scored is limited to words. Therefore, the user cannot interactively specify a new attribute by looking at the first search result. What can be done is only the score of the word, not the score of the individual data or the attribute of the individual data, so it cannot be used for multidimensional display of search results.

  The present invention has been made in view of such a background, and a problem to be solved is to make it possible to efficiently search data closer to a user's search target.

  In order to solve the above-mentioned problem, the first means analyzes a search request input in a search device that searches a database based on a designated search word or search sentence and outputs a search result. Setting means for setting search conditions, acquisition means for searching the database based on the search conditions set by the setting means, and acquiring a data set of search results, and individual data of the data set acquired by the acquisition means And calculating means for calculating a score for performing a desired ordering.

  A second means is characterized in that, in the first means, the calculation means calculates a score by specifying an attribute of data to be scored in addition to individual data of the data set.

  The third means is characterized in that, in the first means, the calculation means calculates a score using a part of the data set acquired by the acquisition means.

  The fourth means is characterized in that, in the first means, a new score is calculated by using the part of the preset higher score of the data set obtained by the obtaining means by the computing means.

  The fifth means is characterized in that, in the first means, the calculation means calculates a new score using the score assigned to the data set acquired by the acquisition means.

  A sixth means is any one of the first to fifth means, wherein the computing means is designated with a score obtained by using a designated data set among the data sets obtained by the obtaining means. A new score is obtained from a calculation with a score obtained using a data set that is not included in the data set.

  A seventh means is any one of the first to sixth means, wherein the calculation means obtains a score obtained by using the data set obtained by the obtaining means and a data set that has not been obtained. The score is obtained from the calculation with the obtained score.

  The eighth means includes a user terminal, a client terminal and a server in any one of the first to seventh means, wherein the database is on the server side, the setting means is on the client terminal side, and the acquisition means is the client terminal The computing means is provided on the client terminal side or server side on the server side or server side.

  A ninth means is characterized in that, in any one of the first to eighth means, a display means for displaying the search result on a two-dimensional coordinate is provided.

  A tenth means is a computer program loaded on a computer and executed by a computer provided in the search means. The first procedure receives a search request, analyzes the search request and sets a search condition, A second procedure for searching a database based on the search condition obtained in the first procedure and obtaining a data set obtained as a result of the search, and individual data in the data set obtained in the second procedure And a third procedure for calculating a score for performing a desired ordering.

  The eleventh means is characterized in that, in the tenth means, the search request includes designation of an attribute of data to which the score is attached.

  The twelfth means is the tenth means, wherein in the third procedure, a part of the data set acquired in the second procedure, a part of the upper score set in advance of the data set, and the data set A new score is calculated using either the part that includes the specified score in the search condition and the part of the data set that is not included in the specified data set or the specified data set. It is characterized by doing.

A thirteenth means is characterized in that the computer program according to any one of the tenth to twelfth is recorded on a recording medium so as to be read and executed by a computer.

  In the embodiment described later, the database is denoted by reference numeral 31, the setting means is denoted by search request analyzing means 22, the obtaining means is denoted by retrieval means 32, the computing means is denoted by score calculating means 33, the user terminal is denoted by reference numeral 10, and the client terminal Corresponds to the reference numeral 20, the server corresponds to the reference numeral 30, and the display means corresponds to the display device 40 and the main screen 41, respectively.

  According to the present invention, since a new score is calculated using the search result set, data closer to the user's search target can be efficiently searched.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram illustrating an outline of a search system according to the first embodiment. This search system is configured as a book search system, and here, an example where the word “cooking” is used as a search word will be described. The search system according to this embodiment includes a user (terminal) 10, a client (terminal) 20, and a server 30. The user and the client are composed of personal computers. The client 20 includes an input unit 21, a search request analysis unit 22, and an output unit 23, and the server 30 includes a database 31, a search unit 32, and a score calculation unit 33.

  A search request 11 is output from the user 10 and a search result 12 is input. The search request 11 is a request from outside the system, such as a user, and includes a search word, a search sentence, a search method, a search target, and the like. The search result 12 is output (displayed) as a search result input from the client 20 side.

  The client 20 includes an input unit 21, a search request analysis unit 22, and an output unit 23. The input means 21 is an interface of this system that receives the search request 11. The search request analysis unit 22 analyzes the search request 11, requests the search unit 32 to acquire necessary data, and passes the acquired data to the output unit 23. The output means 23 is an interface of this system for outputting search results.

  The server 30 includes a database 31, search means 32, and score calculation means 33. The database 31 stores search target data. The search means 32 acquires data from the database 31 under specified conditions. If a score is required, the score calculation means 33 is requested for the score. The score calculation means 33 calculates the data score. The score calculation means 33 can receive a subset of the database from the search means 32 and perform score calculation such that the method of assigning a value changes according to the set.

  The search request analysis unit 22, the search unit 32, and the score calculation unit 33 are configured by software, and various controls are performed by the CPU (not shown) of the client 20 and the server 30 executing the software.

For example, data as shown in Table 1 is stored in the database 31.

  In this embodiment, for example, the user 10 inputs the search term “cooking” and specifies that it is included in the title. The system returns the search results to the user 10 side in the order of suitability of titles. Next, the user 10 selects an author as an attribute for rearranging the search results. The system uses the search result set to score the author. The user 10 obtains a search result arranged according to the score of the author. A processing procedure for performing such a search will be described with reference to a flowchart. FIG. 2 is a flowchart showing the processing procedure of the entire search.

  In the figure, when a search term or search target is input from the user 10 side and a search request is input (step S101), the search request analysis means 22 analyzes the user search request (step S102). Here, processing differs depending on whether or not a re-search request is made. Next, the search means 32 performs a search on the set of search targets in the database 31 (step S103), and checks whether or not there is a need for score calculation (step S104). If necessary, a score is calculated by a designated score calculation method (step S105), and a search result is output based on the calculation result (step S106). If not necessary, the search result is output as it is (step S106). If a re-search is requested, the process returns to step S101 to search again, and if not requested, the process ends in that state (step S107).

  In step S102, the user's search request is analyzed as described above. FIG. 3 is a flowchart showing the processing procedure of this analysis. In the figure, when a search request is received (step S201), it is checked whether or not it is a re-search (step S202), and if it is a re-search, the previous search request is acquired (step S203), and the acquisition target is “author”. And “author score” are additionally set (step S204), and a search request is returned to the search means 32 (step S206). On the other hand, if the search is not performed again in step S202, "cooking" is set as the search term, "title" is set as the search target, "title" and "title suitability" are set as the acquisition target (step S205), and search is performed in step S206. Returns the request.

In step S103, a search is executed. The search is executed by executing a search for obtaining data requested by the user and returning a search result set to the score calculation means 33 or the search request analysis means 22 side. It is. An example of the acquired search result set is shown in Table 2 below.

  In step S106, the search result is output. The output search result is displayed on the display screen of the user 10 here. This display screen 40 is shown in FIG. On the main screen 41 of the display screen 40, as shown in Table 2, titles are displayed side by side in the vertical direction from the descending order of suitability of titles. The first sub-screen 42 displays the search term “dishes” and the search target “title”, and the second sub-screen 43 inputs the search term and search target for further narrowing down. The screen to do is displayed. In response to this display, if there is a re-search request from the user 10 in step S107, the search is performed again. In the re-search, the user inputs a re-search specification by inputting the search request in step S101. For example, specify that the search results obtained in the first search are scored by author. Next, in step S102, the search request is analyzed. For example, in the case where a book including “cooking” in the title is scored by the author, first, acquisition of a book including “cooking” in the title is requested, and the search result Requests that each author of a book in the set be counted. In the execution of the search in step S103, for example, when a score is required and a score cannot be obtained directly from the search result, the score is calculated. In the score calculation, the score is calculated by a specified score calculation method. For example, when the score of an author is indicated by the number of books whose author is included in a certain set, the search result set is received, and the authors appearing in the search result set are counted and added to the search result.

  FIG. 5 is a flowchart showing a processing procedure of score calculation. When calculating a score, first, a search result set including a title, a title score, and an author is received (step S301), and a list for storing the number of appearances of the author is prepared (step S302). Next, one author is extracted from the retrieved result set (steps S303 and S304), and an author is further extracted from the retrieved one item to check whether it has already been acquired (step S305). If already acquired (step S306—YES), the number of appearances of the corresponding author in the list is incremented by 1, and the process returns to step S303. If not acquired (NO in step S306), the item of the author is added to the list, the appearance count is set to 1, and the process returns to step S303. This process is executed for all of the search result sets, and the number of times the author appears as an author score is added to each data of the search result set (step S309), and the search result set is returned to the search means 32 (step S310).

Table 3 is an example of a list in which the number of books of each author included in the search result set extracted in step S304 is counted.

The search result set obtained by adding the author's score in step S309 to the list in step S304 shown in Table 3 is as shown in Table 4.

  FIG. 6 is a diagram showing an output display screen of the re-search result set shown in Table 4. In this display screen, the search result is output to the main screen 41. For example, if the titles are arranged in descending order from the top in the vertical direction and arranged in descending order of the author's score in the horizontal direction, the titles are displayed at positions obtained by the title's suitability and the author's score. Incidentally, in the present embodiment, a case has been described in which a search is performed once and a search result set is obtained, and then the search results are searched again. However, it is also possible to specify all from the beginning. In this case, however, it is necessary to specify which score is to be assigned using which search result set.

  Further, in the present embodiment, it is described that the search is performed every time. However, if the difference between the previous search condition and the current search condition is only data that can be acquired by the score calculation, the already acquired search result set is used. It is also possible to omit the re-search and execute only the score calculation. However, in that case, the search means or the like needs to store the previous search condition and compare it with the current search condition.

  As described above, according to the present embodiment, the author's score is assigned using a set limited by a certain search condition, so that it is more appropriate than scoring using the entire set. Score calculation is possible. For example, in this embodiment, the score can be calculated so that a person who has written many dishes-related books has a high author score, and the author score does not increase even if many books are written in other fields.

The second embodiment is an example in which the score calculation method is different from the first embodiment.
In this embodiment, the score is calculated from the search result subset that is not limited to all the search result sets, for example, those registered within one year. FIG. 7 is a flowchart showing the processing procedure at this time. The flowchart of FIG. 7 is obtained by adding “registration date” to the search condition of the process in step 204 of the flowchart shown in FIG. 3 of the first embodiment. That is, in the flowchart of FIG. 7, after the previous search request is acquired in step S203, “author”, “author score”, and “registration date” are additionally set as acquisition targets (step S204a). A search request is returned (step S206). Other steps are the same as those in FIG.

  The score is calculated according to the processing procedure shown in the flowchart of FIG. In this processing procedure, a search result set obtained by adding a registration date to the process of step S301 in FIG. 5, that is, a search result set including a title, a title score, an author, and a registration date is received (step S301a). On the other hand, the processing from step S302 to step S310 is executed. However, the processing of steps S305a, S305b, and S305c is executed instead of step S305 due to the addition of the registration date. That is, it is checked whether or not the registration date is within one year (step S305a), and the author is taken out from those within one year (step S305b) to check whether or not it has already been acquired (step S305c). Thereafter, the processing after step S306 is executed.

The search result set processed by the processing procedure of the flowchart of FIG. 8, that is, the search result set including “dish” in the title and registration within one year is as shown in Table 5, for example.

When the number of books of each author included in the search result set limited by the registration date with respect to the results of Table 5 is counted, it is as shown in Table 6.

The search result set obtained by adding the author's score in step S309 to the list in step S304 (Table 6) in the flowchart of FIG. 8 is as shown in Table 7.

  FIG. 9 is a diagram showing an example of a search result display screen in this embodiment, in which the registration date is set on the vertical axis and the author is set on the horizontal axis. As a result, it is displayed on the two-dimensional coordinates which author has published a book related to what kind of dish at what time of the year.

  Other parts not specifically described are configured in the same manner as in the first embodiment and function in the same manner. Controls other than the flowcharts shown in FIGS. 7 and 8 are the same as those in the first embodiment.

  In the second embodiment configured as described above, a more limited set than in the first embodiment can be used, so that more appropriate score calculation can be performed. For example, in this example, as can be seen from Table 7, the score can be calculated so that a person who has written a lot of “cooking” books that have been published or received recently has a high score. Those who do not have high author scores. Therefore, a search result that gives a better understanding of the current situation can be obtained.

The third embodiment is still another example in which the score calculation method is different from the first embodiment.
In this embodiment, the score is calculated from a subset of search results restricted by, for example, the top N matching levels of titles obtained by the first search without using all the search result sets. FIG. 10 is a flowchart showing a processing procedure of this score calculation. In this processing procedure, one item having a larger title score is extracted from the processing of step S303 in the flowchart of FIG. 5 in the first embodiment (step S303a), and steps S304 to S310 are performed on the extracted title. Execute the process. At that time, the processing procedure of steps S305d and S305e is executed before the step S305. That is, in step S305d, it is checked whether the number of titles extracted in step S303a is N or less. If the number is larger than N (step S305e-NO), the process proceeds to step S309. YES), the process of step S305, that is, the author is taken out and it is checked whether or not it has already been acquired, and the process after step S306 is followed.

Table 8 is an example of a search result set including a word related to “cooking” in the title. As can be seen from this table, books whose title is “classroom” determined to be related to “cooking” are also searched.

When the search result set shown in Table 8 is limited to the top five matching degrees of titles, Table 9 shows.

Further, when the number of books of each author included in the search result set limited by the suitability of titles shown in Table 9 is counted and listed, the results are shown in Table 10.

When the score of the author is added to the results of Table 9, the search result set is as shown in Table 11.

  FIG. 11 is a diagram showing an example of a search result display screen in this embodiment, in which the registration date is set on the vertical axis and the author is set on the horizontal axis. As a result, the search result subset of the author and the book ranked higher in the degree of suitability of the title obtained in the first search is displayed on the two-dimensional coordinates.

  Other parts not specifically described are configured in the same manner as in the first embodiment and function in the same manner. Controls other than the flowcharts shown in FIGS. 7 and 8 are the same as those in the first embodiment.

  In the third embodiment configured as described above, a more limited set than that in the first embodiment can be used in a limiting method different from that in the second embodiment, so that more appropriate score calculation can be performed. For example, in this embodiment, it is assumed that a book including “related words” in the title is searched for in the first search, not a book including “dishes” in the title. In this case, although “cooking” is not included, a book including “cooking” or “dinner” in the title can be searched.

  However, since an appropriate related word is not always selected, a book that does not seem to be related to cooking is also included in the search result set. In this example, it is “English conversation class”. However, books that do not seem to be related to such dishes often have low titles, so if you restrict the titles to only the top ones when assigning the author's score, these books Can be calculated so that a person who has written many books related to “cooking” has a high score.

The fourth embodiment is another example in which the score calculation method is further different from the first embodiment.
In this embodiment, the score is calculated by weighting the number of counts according to the score obtained in the first search. FIG. 12 is a flowchart showing a score calculation processing procedure in this embodiment. This flowchart differs from that of the first embodiment in the processing of steps S307, S308, and S309 of the flowchart of FIG. That is, after the processing from step S301 to step S306, if the author is acquired, the weighted appearance count of the corresponding author in the list is increased by the title score (step S307a), and if not, the author is added to the list. The title score is set as the weighted appearance count (step S308a). After performing this operation on all search result sets (step S304), the author's weighted appearance count is added to each data of the search result set (step S309a), and the process of step S310 is executed. To do.

An example of the initial search result set in which the scores are calculated in this way is shown in Table 12.

Further, the number of weighted books of each author included in the search result set is as shown in the list shown in Table 13, for example.

A set of search results obtained by adding the author's score to the list of Table 12 is as shown in Table 14, for example.

  Other parts not specifically described are configured in the same manner as in the first embodiment and function in the same manner. Controls other than the flowchart shown in FIG. 12 are the same as those in the first embodiment.

  According to this embodiment configured and processed in this way, score calculation reflecting the contents of the initial search result set can be performed, and thus more appropriate score calculation can be performed. For example, if the total suitability of titles as shown in Table 13 is used instead of the number of cases, the score of authors with many books with high title suitability will be high, and the score of authors with only a large number of books will be low. Become. If the result is displayed on the main screen 41 of the display screen 40 as described above, the search result can be displayed on the screen using the weighted author score as a parameter.

  The fifth embodiment is an example in which the score calculation method is further different from the first embodiment.

  In this embodiment, scoring is performed using the entire set, but the score is calculated by weighting the set obtained in the first search. For example, when a book of a certain author is included in the search result set, it is counted as two, and a book not included in the search result set is counted as one and the score is calculated. FIG. 13 is a flowchart showing a processing procedure of score calculation in this embodiment. In this flowchart, steps S308b and S308c are provided after step S308 in the flowchart of FIG. That is, when an author name appears for the first time (step S306-NO), an item of the author is added to the list in step S308, and the number of appearances is set to 1, and then the author name to be searched as a search term and the author as the search target Then, a search result is acquired by searching for the number of books to be acquired (step S308b), the obtained number is set as the number of appearances (step S3008c), the process returns to step S302, and the processes after the step are repeated.

  Other parts not specifically described are configured in the same manner as in the first embodiment and function in the same manner. Controls other than the flowchart shown in FIG. 12 are the same as those in the first embodiment.

  According to this embodiment configured and processed in this way, an appropriate score can be calculated even when the set is too limited. For example, if there are few search result sets and most authors have written only one, there is no difference in author scores. Therefore, by taking into account information other than the search result set, the author's score difference can be calculated.

It is a block diagram which shows the outline | summary of the search system in Example 1. FIG. 6 is a flowchart illustrating a processing procedure for the entire search in the first embodiment. It is a flowchart which shows the detail of the process sequence of step S102 in FIG. It is a figure which shows the display screen of the search result of step S106 in FIG. 6 is a flowchart illustrating a score calculation processing procedure according to the first embodiment. It is a figure which shows the output display screen of the re-search result set shown in Table 4. FIG. It is a flowchart which shows the process sequence of step S102 in Example 2. FIG. 10 is a flowchart illustrating a processing procedure of score calculation in the second embodiment. FIG. 10 is a diagram illustrating a search result display screen according to the second embodiment. 12 is a flowchart illustrating a processing procedure of score calculation in the third embodiment. FIG. 10 is a diagram illustrating a search result display screen according to the third embodiment. 12 is a flowchart illustrating a score calculation processing procedure according to the fourth embodiment. 10 is a flowchart illustrating a score calculation processing procedure according to the fifth embodiment.

Explanation of symbols

10 users (terminals)
20 clients (terminals)
22 Search request analysis means 30 Server 31 Database 32 Search means 33 Score calculation means

Claims (13)

In a search device that searches a search object from a database based on a specified search word or search sentence and outputs a search result,
A setting means for analyzing the input search request and setting search conditions;
An acquisition unit that searches the database based on the search condition set by the setting unit and acquires a data set of search results;
Computing means for calculating a score for performing a desired ordering on individual data of the data set obtained by the obtaining means;
A search apparatus comprising:   The search device according to claim 1, wherein the arithmetic means calculates a score by specifying an attribute of data to be scored in addition to individual data of the data set.   The search device according to claim 1, wherein the calculation unit calculates a score using a part of the data set acquired by the acquisition unit.   The search device according to claim 1, wherein the calculation unit calculates a new score using a preset upper score portion of the data set acquired by the acquisition unit.   The search device according to claim 1, wherein the calculation unit calculates a new score using a score given to the data set acquired by the acquisition unit.   The calculation means includes a score obtained using a specified data set among the data sets obtained by the obtaining means, and a score obtained using a data set not included in the designated data set. The search device according to claim 1, wherein a new score is obtained from the calculation.   The said calculating means calculates | requires a score from the calculation of the score acquired using the data set acquired by the said acquisition means, and the score acquired using the data set which was not acquired. 7. The search device according to any one of items 6 to 6.   Including a user terminal, a client terminal and a server, the database is provided on the server side, the setting means is provided on the client terminal side, the acquisition means is provided on the client terminal side or server side, and the computing means is provided on the client terminal side or server side. The search device according to claim 1, wherein the search device is provided.   The search device according to claim 1, further comprising display means for displaying the search result on a two-dimensional coordinate. In a computer program loaded on a computer and executed by a computer provided in the search means,
A first procedure for receiving a search request, analyzing the search request and setting search conditions;
A second procedure for searching a database based on the search condition obtained in the first procedure and acquiring a data set obtained as a result of the search;
A third procedure for calculating a score for performing a desired ordering on each piece of data of the data set obtained in the second procedure;
The computer program characterized by including.   11. The computer program according to claim 10, wherein the search request includes specification of an attribute of data to be scored.   In the third procedure, a part of the data set acquired in the second procedure, a part of a higher-order score set in advance of the data set, a part including a score assigned to the data set as a search condition, and 11. The computer according to claim 10, wherein a new score is calculated using any one of a designated data set and a data set portion not included in the designated data set among the data sets. program.   13. A recording medium on which the computer program according to claim 10 is recorded so as to be read and executed by a computer.

Images