JP2009080675A - Retrieval device, retrieval method, program, and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To extend a retrieval term considering the meaning of the term and to achieve retrieval further matched with a condition while suppressing the extension of the retrieval term to extremely general upper terms or the extension of retrieval term to abstract terms in retrieval using a thesaurus dictionary. <P>SOLUTION: A thesaurus dictionary retrieval means 101 extends a retrieval term by using a thesaurus dictionary DB 103 and a human resource retrieval means 102 retrieves a human resource DB 104 by using the extended retrieval terms. In the thesaurus dictionary DB 103, a stop mark indicating a general term, an abstract flag indicating an abstract term, or the like can be set in a registered term. The thesaurus dictionary retrieval means 101 is provided with a thesaurus dictionary retrieval end point determination means 1011 for setting a term in which the stop mark is set as an end point of upper retrieval of the thesaurus dictionary and a thesaurus dictionary abstract term excluding means 1012 for excluding a term in which the abstract flag is set from the extension of retrieval terms. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to control when database search is performed by expanding a search term using a thesaurus dictionary.

Patent Document 1 discloses a search device that performs synonym expansion of a search target term by searching from the search target term to the highest word (or the lowest word) of the thesaurus dictionary. This search device calculates the similarity in all the extended synonyms, and narrows down the search target terms based on the similarity threshold. Alternatively, this search device specifies a hierarchical range of a thesaurus dictionary and searches for terms to be searched when searching.
JP 2003-345824 A

  However, in general, a thesaurus dictionary varies in terms of synonyms depending on terms. For example, a synonym hierarchy of a certain term is fine, and a synonym hierarchy of a certain term is roughly created. This is because, for example, because a person who creates a thesaurus dictionary has different specialized abilities, a dictionary having many layers is created in a technically detailed field.

  As described above, when a term registered in the thesaurus dictionary has a difference in the hierarchy of synonyms depending on the term, the similarity threshold value and the number of times are used when searching for synonyms as in the search device described in Patent Document 1. In the method of specifying the search range, there is a possibility that only a very relevant word is expanded or a word that is not very relevant is expanded depending on a search term to be expanded.

  In general, a thesaurus dictionary is managed without considering the meaning of synonyms registered in the dictionary. For this reason, in the method of specifying the search range by the threshold value and the number of times of similarity as in the search device of Patent Document 1, since all of the search range is a search target, the search target is a meaningless synonym. There is a risk of expanding the search term.

  The present invention has been made to solve the above-described problems, and the object of the present invention is to expand search terms to extremely general terms and search for abstract terms in a search using a thesaurus dictionary. It is to provide a mechanism that can suppress the expansion of the search term to the search term, extend the search term in consideration of the meaning of the term, and perform a search that matches the condition more.

  The present invention provides a thesaurus dictionary search means for searching for synonyms of a search term using a thesaurus dictionary and extending the search term, and a database search means for searching a database using the search term extended by the thesaurus dictionary search means. The thesaurus dictionary can set either a stop mark indicating a general term or an abstract flag indicating an abstract term, or both of the registered terms in the registered term. The means includes a thesaurus dictionary search end point determination unit that uses the term set with the stop mark as the end point of the upper search of the thesaurus dictionary, and the thesaurus dictionary abstract term exclusion that excludes the term set with the abstract flag from the expansion of the search term. Means.

  According to the present invention, in a search using a thesaurus dictionary, a search that considers the meaning of a term by suppressing the expansion of a search term to a very general high-level term or the expansion of a search term to an abstract term. Terms can be expanded, and effects such as a more suitable search can be performed.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In this description, the case where it is assumed that the search device according to the present invention is used for human resource search will be described as an example.

  FIG. 1 is a system configuration diagram showing an example of a configuration of a human resource search apparatus to which the search apparatus according to the present invention can be applied.

  In FIG. 1, reference numeral 100 denotes a human resource search apparatus, which includes a thesaurus dictionary search unit 101 that performs a thesaurus dictionary search, a human resource search unit 102 that performs a human resource search, a thesaurus dictionary DB 103 that manages a thesaurus dictionary, and a human resource that manages human resource data. It consists of a DB (human resource database) 104.

  The thesaurus dictionary search means 101 also determines a thesaurus dictionary search end point determination means 1011 that determines the end point of the higher level search of the thesaurus dictionary (a term that is meaningless even if a higher level search is performed because it is a very general term). Thesaurus dictionary abstract term exclusion means 1012 that excludes from the expansion of the search term because the term is abstract among the terms hit in the thesaurus dictionary search, thesaurus dictionary search that calculates the usefulness of the term hit in the thesaurus search It has hit term usefulness calculating means 1013.

  The human resource search unit 102 includes a human resource search score calculation unit 1021 that calculates an experience value of a human resource hit in the human resource search.

  These thesaurus dictionary search means 101 and human resource search means 102 can be realized by a general Web application technology such as a Java (registered trademark (hereinafter omitted)) servlet.

  Further, the client communication unit 105 can use a function provided by a Web application server serving as a Web application platform. For example, these functions can be realized as a Java servlet program that runs on the Web application server Tomcat.

  Reference numeral 111 denotes a client device that is communicably connected to the human resource search device 100 via a predetermined communication medium. The client device 111 is configured by a personal computer or the like, and can communicate with the human resource search device 100 using, for example, a Web browser.

  The human resource search device 100 has a configuration in which search processing is performed based on a search condition input from the client device 111 and a search result is transmitted to the client device 110.

  FIG. 2 is a block diagram illustrating an example of a hardware configuration of the human resource search apparatus 100 illustrated in FIG.

  As shown in FIG. 2, the human resource search apparatus 100 includes a central processing unit (hereinafter referred to as CPU) 201, a ROM (Read Only Memory) 202, a RAM (Random Access Memory) 203, an HDD (Hard Disk Drive) 204, It consists of a network interface 205.

  The CPU 201 can implement various functions to be described later by loading a program stored in the HDD 204 onto the RAM 203 and executing the program.

  The thesaurus dictionary search means 101 (including thesaurus dictionary search end point determination means 1011, thesaurus dictionary abstract term exclusion means 1012, thesaurus dictionary search hit term usefulness calculation means 1013) and the personnel search means 102 (personnel search) shown in FIG. (Including score calculation means 1021) corresponds to a function realized by the CPU 201 loading a program stored in the HDD 204 into the RAM 203 and executing it.

  Further, the thesaurus dictionary DB 103 and the human resource DB 104 shown in FIG. 1 are realized on the HDD 204, for example.

  First, the configuration of the thesaurus dictionary DB 103 used in this embodiment will be described with reference to FIGS.

  FIG. 3 is a schematic diagram schematically showing an example of the thesaurus dictionary DB 103 shown in FIG. Here, an example of the thesaurus dictionary DB 103 taking “software” as an example is shown.

  As indicated by 301 in the figure, a line (−) without an arrow indicates a higher order and a lower order.

  In addition, as shown by 302 in the figure, words connected by arrows (→) indicate related words. Note that a word with an arrow is a related word when viewed from a word without an arrow (the reverse is not true).

  Furthermore, as indicated by 303 in the figure, a line (← →) with arrows at both ends indicates that the words connected by the arrows (← →) are related words.

  In addition, as shown by 304 and 305 in the figure, the underlined words have the following meanings.

  First, as indicated by reference numeral 304 in the figure, a normal underlined word is a word having an abstract attribute (a term excluded from expansion of a search term because it is abstract).

  A word with a wavy line (not shown) indicates a word having an upper end attribute (a term used as an end point of upper search in the thesaurus dictionary because it is a very general term).

  Furthermore, as indicated by reference numeral 305 in the figure, a word with a double line indicates that it is a word having an upper end attribute and an abstract attribute.

  FIG. 4 is a schematic diagram schematically showing an example of weights assigned to the respective words registered in the thesaurus dictionary DB 103 shown in FIG. Here is an example of tracing from "Java1.3".

  The numerical value in FIG. 4 shows the weight (similarity) given between each term. For example, 401 indicates the weight of the broader term “Java” as viewed from “Java 1.3”, and here, the weight is “0.8”.

  On the other hand, 402 indicates the weight of the low-level word “Java1.3” as viewed from “Java”, and here, the weight is “1.0”.

  FIG. 5 is a data configuration diagram showing an example of the data configuration of each word registered in the thesaurus dictionary DB 103 shown in FIGS.

  In FIG. 5, 500 is a word registered in the thesaurus dictionary DB 103, and includes a heading 501, a synonym list (synonym list) 502, a broader word 503, a narrower word list (lower word List) 504, a related word list (related word List). ) 505, an abstract flag 506, and a stop mark (Stop mark) 507.

  The heading 501 stores a character string. The synonym list 502 is a list of synonyms. Each synonym stores a character string.

  The broader word 503 is composed of information indicating the broader word (higher word information). The low-order word list 504 is a list of information (low-order word information) indicating low-order words. The related word list 505 is a list of information (related word information) indicating related words. The data structure of the broader word information, the lower word information (individual items 1 to j), and the related word information (individual items 1 to k) is shown in FIG.

  The abstract flag 506 is a flag indicating that the word is excluded as a search word when the term is expanded using the thesaurus dictionary DB 103 because the word is an extremely abstract term. When the abstract flag 506 is “TRUE”, the word is excluded as a search term. That is, a term in which “TRUE” is set in the abstract flag when executing a thesaurus dictionary search when expanding a term is excluded from the expansion of the search term.

  The stop mark 507 is a mark indicating that a search higher than the word (upper extension) is not performed when the term is expanded using the thesaurus dictionary DB 103 because the word is a very general term. When the stop mark 507 is “TRUE”, a search higher than that word is not performed. That is, a term for which a stop mark is set when executing a thesaurus dictionary search for expanding a term is the end point of a higher-level search of the thesaurus dictionary.

  FIG. 6 is a data configuration diagram showing a detailed structure of the broad word information, the low word information (individual items 1 to j), and the related word information (individual items 1 to k) shown in FIG.

  As shown in FIG. 6, the broader word information, the narrower word information, and the related word information are composed of a partner headline 601 and a similarity 602. The partner headline 601 stores a character string corresponding to a headline 501 (FIG. 5) of a broader word, a lower word, or a related word (that is, a word of another node in the thesaurus). The similarity 602 calculates and stores the cumulative weight (numerical value) of the partner headline word from the currently focused word (a calculation method will be described later).

  FIG. 7 is a diagram illustrating an example of words in the thesaurus dictionary DB 103.

  7A shows an example of a word in the thesaurus dictionary DB 103 when the heading is “Oracle (registered trademark (hereinafter omitted))”. Further, (b) shows an example of a broader term of “Oracle”. This example corresponds to the case where the weight of the broader term “RDB” viewed from “Oracle” is “0.5”.

  Hereinafter, a search process in the search device of the present invention will be described with reference to a flowchart.

  FIG. 8 is a flowchart showing an example of the first control processing procedure in the present invention, and corresponds to the search processing in the human resource search device 100. Note that the processing of this flowchart corresponds to a function realized by the CPU 201 illustrated in FIG. 2 loading a program stored in the HDD 204 into the RAM 203 and executing the program.

  First, in step S <b> 801, the CPU 201 receives an input of search conditions transmitted from the client device 111.

  In the client device 111, the search condition input screen 900 (FIG. 9) is displayed on the display, and the search condition input screen 900 accepts input of the search condition.

  FIG. 9 is a diagram showing an example of a search condition input screen 900 in the present embodiment. The search condition input screen 900 is configured to be read from the hard disk or the like of the client device 111 and displayed on the display even if the search condition input screen 900 is transmitted from the human resource search device 100 and displayed on the display of the client device 111. There may be.

  In FIG. 9, reference numeral 901 denotes an indispensable check box, and a search condition checked in this check box is an indispensable search condition.

  Reference numeral 902 is a technical / environmental condition input field, 903 is a process condition input field, 904 is a role condition input field, and 905 is an experience time condition input field.

  Reference numeral 906 denotes a cancel button. When this button is clicked (pressed), the search condition input on the search condition input screen 900 is discarded and the search process is cancelled.

  Reference numeral 907 denotes a search start button. When this button is clicked (pressed), the search condition input on the search condition input screen 900 is transmitted from the client device 111 to the human resource search device 100.

  In the human resource search device 100, when the search condition is received from the client device 111, the CPU 201 of the human resource search device 100 stores the received search condition in the RAM 203 in the data structure shown in FIGS. 10 to 13 and advances the process to step S1302. .

  FIG. 10 is a data configuration diagram illustrating an example of a data configuration of search conditions held by the human resource search device 100.

  As shown in FIG. 10, the search condition is held in a list structure (search condition list 1000) of a plurality of search conditions (condition 1, condition 2, condition 3,..., Condition n).

  Each search condition in the search condition list 1000 is input in a mandatory check box 901, a technology / environment condition input field 902, a process condition input field 903, a role condition input field 904, and an experience time condition input field 905 on the search condition input screen. The set of search conditions (that is, the input conditions in one horizontal row in FIG. 10) are stored.

  FIG. 11 is a data configuration diagram showing an example of the data configuration of each search condition in the search condition list 1000 shown in FIG.

  As shown in FIG. 11, the search condition includes a technology / environment condition 1101, a process condition 1101, a role condition 1103, and a required experience level 1104.

  The technology / environment condition 1101 includes a condition heading 1105, an extended list 1106, and an essential flag 1107.

  The condition heading 1105 of the technology / environment condition 1101 stores the condition input in the technology / environment condition input field 902 of the search condition input screen 900.

  The process condition 1102 stores the condition input in the process condition input field 903 of the search condition input screen 900. The condition heading 1105 of the role condition 1103 stores the condition input in the role condition input field 904 of the search condition input screen 900. In the condition heading 1105 of the required experience level 1104, the experience time value input in the experience time input field 904 of the search condition input screen 900 is stored.

  Further, the mandatory flag 1107 of the technology / environment condition 1101 stores “TRUE” when the mandatory check box 901 is checked, and “FALSE” when the mandatory check box 901 is not checked.

  Note that when the skill condition (search condition) is input in step S801 in FIG. 8, no data is stored in the extended list 1106 of the technical / environmental condition 1101, and information on the extended word extended in step S802 described later. Is stored.

  FIG. 12 is a data configuration diagram showing an example of the data configuration of each extended word in the extended list 1106 shown in FIG.

  As shown in FIG. 12, the extended word is composed of an extended headline 1201, an extended weight 1202, and an abstract flag 1203, and information on the extended word extended in step S802 of FIG.

  The extended heading 1201 stores the extended word heading character of the extended word. The extension weight 1202 stores an extension weight of the extension word (which indicates the usefulness of the extended search term and will be described in detail later). The abstract flag 1203 stores the true / false value of the extended word abstract flag.

  FIG. 13 is a diagram illustrating an example of the search condition list illustrated in FIG.

  The example illustrated in FIG. 13 corresponds to the case where “Oracle” and “Java 1.3” are input in the technology / environment condition input field 902 of the search condition input screen 900.

  Returning to the flowchart of FIG.

  In step S802, the CPU 201 (thesaurus dictionary search unit 101) executes a search word expansion process (keyword expansion process) using synonyms. Details of this search term expansion processing will be described later with reference to FIG.

  Next, in step S803, the CPU 201 (human resource search means 102) executes a human resource skill profile DB search process using the keyword expansion process result of step S802. Details of the human resource skill profile DB search process will be described later with reference to FIG.

  Next, in step S804, the CPU 201 (human resource search meter detection / calculation means 1021) adds a point (scoring process of searched human resources) for each human resource search result in step S803 that is closer to the search condition. ). The details of scoring of the searched human resources will be described later with reference to FIG.

  Next, in step S805, the CPU 201 transmits the human resource search result list (an example shown in FIG. 25 described later) searched in step S803 and scored in step S804 to the client device 111 that has requested the human resource search. . The client device 111 that has received this personnel search result list displays the personnel search result list on a display.

  Hereinafter, a search word expansion process for expanding the search condition to words having similar meanings using the synonym dictionary shown in step S802 of FIG. 8 will be described with reference to the flowchart of FIG.

  FIG. 14 is a flowchart showing an example of a second control processing procedure according to the present invention, and corresponds to the search word expansion processing shown in step S802 of FIG. Note that the processing in this flowchart corresponds to the function of the thesaurus dictionary search unit 101 realized by the CPU 201 shown in FIG. 2 loading the program stored in the HDD 204 into the RAM 203 and executing it.

  First, in step S1401, the CPU 201 determines whether an unextended search condition (skill condition) exists in the search condition list 1000 held in the RAM 203 input in step S801, and the unextended search condition is determined. While it is determined that it exists, the processes of steps S1402 to S1408 are executed. In this embodiment, only the technical / environmental condition 1101 in the search condition is expanded for the search term. However, the process condition 1102 and the role condition 1103 may be configured to extend the search term, respectively. Good.

  On the other hand, if it is determined in step S1401 that there is no unextended search condition, the CPU 201 ends the search word expansion process.

  In step S1402, the CPU 201 extracts a search condition (skill condition) from the search condition list 1000.

  In step S1403, the CPU 201 extracts a keyword (condition heading 1105 (FIG. 11)) from the search condition (skill condition) extracted in step S1402.

  In step S1404, the CPU 201 executes a synonym control process for controlling the keyword extracted in step S1403 to a representative word using the thesaurus dictionary DB103. Details of the synonym control process will be described later with reference to FIG.

  In step S <b> 1405, the CPU 201 executes a high-order search process for expanding the broad word of the keyword using the keyword controlled to the representative word in step S <b> 1404 and the thesaurus dictionary DB 103. The details of the upper search process will be described later with reference to FIG.

  In step S1406, the CPU 201 executes low-order search processing for performing low-order word expansion of keywords using the broader word expanded in step S1405 and the thesaurus dictionary DB103. The details of the lower search process will be described later with reference to FIG.

  In step S <b> 1407, the CPU 201 executes related word search processing for expanding keywords to related terms using the thesaurus controlled by the keyword and the thesaurus dictionary DB 103 in step S <b> 1404. Details of the related word search processing will be described later with reference to FIG.

  In step S 1408, the CPU 201 (thesaurus dictionary abstract term excluding means 1012) executes post-abstraction exclusion processing for deleting the extended word whose abstract flag 1203 is “FALSE” from the keyword expansion list 1106. As a result, it is possible to exclude very abstract terms from the extended words as excluded from the search terms.

  The CPU 201 ends the processing of this flowchart when there is no unextended search condition (skill condition) in the search condition list 1000 held in the RAM 203 input in step S801 of FIG.

  The synonym control process shown in step S1404 in FIG. 14 will be described below with reference to the flowchart in FIG.

  FIG. 15 is a flowchart showing an example of a third control processing procedure in the present invention, and corresponds to the synonym control processing shown in step S1404 of FIG. Note that the processing in this flowchart corresponds to the function of the thesaurus dictionary search unit 101 realized by the CPU 201 shown in FIG. 2 loading the program stored in the HDD 204 into the RAM 203 and executing it.

  First, in step S1501, the CPU 201 controls the keyword extracted in step S1403 in FIG. 14 (that is, the keyword input as the search condition (skill condition)) as a representative word. For example, when the keyword input as the search condition is “Java” or “JAVA (registered trademark (hereinafter abbreviated))”, the search condition is converted to a character (representative word) “Java”.

  In the term management means, representative words and those that are not representative words can be discriminated from the detailed structure of each word in the thesaurus dictionary DB 103 shown in FIG.

  The representative word does not hold information in the synonym list 502 of the detailed structure of each word in the thesaurus dictionary DB 103 in FIG.

  On the other hand, information that is not a representative word holds information as a representative word in the synonym list 503 of the detailed structure of each word in the thesaurus dictionary DB 103 of FIG.

  That is, in step S1501, the CPU 201 searches the thesaurus dictionary DB 103 of FIG. 5 using the keyword input as the search condition. When the information that becomes the representative word is held in the synonym list 503 of the searched word, the CPU 201 rewrites the keyword (that is, the condition heading 1105 in FIG. 11) to the representative word ( Convert).

  Hereinafter, the broader term search processing shown in step S1405 of FIG. 14 will be described with reference to the flowchart of FIG.

  FIG. 16 is a flowchart showing an example of a fourth control processing procedure in the present invention, and corresponds to the broader term search processing shown in step S1405 of FIG. Note that the processing in this flowchart corresponds to the function of the thesaurus dictionary search unit 101 realized by the CPU 201 shown in FIG. 2 loading the program stored in the HDD 204 into the RAM 203 and executing it.

  First, in step S1601, the CPU 201 determines whether or not the keyword (hereinafter, keyword) controlled to the representative word in step S1404 in FIG. 14 is a keyword set as an essential condition. If the required flag 1107 (FIG. 11) in the skill condition is “TRUE”, the CPU 201 determines that the keyword is set as an essential condition.

  If it is determined in step S1601 that the keyword has been set as an essential condition, the CPU 201 advances the process to step S1608.

  In step S1608, the CPU 201 sets the keyword as the top element of the extended list 1106 (FIG. 11) in the skill condition with the keyword as the condition heading 1105. That is, an extension word is added to the head of the extension list 1106, and a character string corresponding to the keyword is set in the extension heading 1201 of the added extension word. Thereafter, the CPU 201 advances the processing to step S1609.

  In step S1609, the CPU 201 inputs “1” as the extension weight 1202 of the extension word and “TRUE” as the abstract flag 1203, and ends the processing of this flowchart.

  On the other hand, if it is determined in step S1601 that the keyword is not set as an essential condition, the CPU 201 advances the process to step S1602.

  In step S1602, the CPU 201 sets the keyword as the head element of the extended list 1106 (FIG. 11) in the skill condition with the keyword as the condition heading 1105. Further, the CPU 201 stores the keyword in the RAM 203 as “a term of interest” and advances the process to step S1603.

  Next, in step S1603, the CPU 201 inputs “1” to the extension weight 1202 of the extension word. Further, the CPU 201 searches the thesaurus dictionary DB 103 for the “term of interest” and sets the true / false value of the abstract flag 506 set in the “term of interest” in the thesaurus dictionary DB 103 to the extension word. The abstract flag 1203 is input, and the process proceeds to step S1604. The reason why the extension weight 1202 is set to “1” is because the “term of interest” is not yet expanded.

  In step S <b> 1604, the CPU 201 determines whether the “term of interest” is the highest level in the thesaurus dictionary DB 103 (there is no information in the broader term 503).

  If it is determined in step S1604 that the “term of interest” is the highest level of the thesaurus (the broader term 503 has no information), the CPU 201 ends the processing of this flowchart.

  On the other hand, if it is determined in step S1604 that the “term of interest” is not the highest level of the thesaurus (the higher word 503 has information), the CPU 201 advances the process to step S1605.

  In step S1605, the CPU 201 (thesaurus dictionary search end point determination means 1011) determines whether the “term of interest” is a term with a stop mark (“TRUE” of the stop mark 507).

  If it is determined in step S1605 that the “term of interest” is a term with a stop mark (“TRUE” of the stop mark 507), the CPU 201 (thesaurus dictionary search end point determination means 1011) The process of the flowchart is terminated.

  On the other hand, if it is determined in step S1605 that the “term of interest” is not a term with a stop mark (“FALSE” in the stop mark 507), the CPU 201 (thesaurus dictionary search end point determining means 1011) determines whether the term is “stopped 507”. Proceed with the process.

  In step S <b> 1606, the CPU 201 acquires the broader term 503 of “the term of interest” from the thesaurus dictionary DB 103 and sets it as the head element of the extended list 1106. Further, the CPU 201 sets the acquired broader term as a new “term of interest” (stores it in the RAM 203 as “term of interest”). Referring to FIG. 3, when the “term of interest” is “Java1.3”, “Java” is acquired as a broader term, and “Java” is set as the new “term of interest” To do. Then, the CPU 201 advances the process to step S1607.

  Next, in step S1603, the CPU 201 (thesaurus dictionary search hit term usefulness calculating means 1013) sets a cumulative weight (described later) as the extended weight 1202 of the extended word. Further, the CPU 201 sets the true / false value of the “term of interest” set in the thesaurus dictionary DB 103 in the extended word abstract flag 1203, and the process returns to step S1604.

  The “cumulative weight” here refers to the next element of the new term of interest and the new term of interest in the extended list 1106 set in the thesaurus dictionary DB 103. The weight between words (similarity 602) is multiplied by the cumulative weight of the next element, and indicates the usefulness of the expanded search term. Referring to FIG. 4, for example, when the new “term of interest” is “object-oriented language”, an inter-word space between “Java” and “object-oriented language” which is the next element in the expanded list "0.56" is multiplied by the weight of (here "0.7") and the cumulative weight of "Java", which is the next element in the expanded list (here "0.8 (= 0.8 x 1.0)"). Calculated as “weight”.

  Hereinafter, referring to the flowchart of FIG. 17, the narrower word search process shown in step S <b> 1406 of FIG. 14 (the process of extending the representative word to the broader word in the broader word expansion flow and expanding to the lower word) Will be described.

  FIG. 17 is a flowchart showing an example of a fifth control processing procedure in the present invention, and corresponds to the narrower word search processing shown in step S1406 of FIG. Note that the processing in this flowchart corresponds to the function of the thesaurus dictionary search unit 101 realized by the CPU 201 shown in FIG. 2 loading the program stored in the HDD 204 into the RAM 203 and executing it.

  First, in step S1701, the CPU 201 uses the expanded list 1106 in the skill condition with the keyword, which is the expanded result of the representative word, as the condition heading 1105 (the expanded list added in the upper search process in step S1405 (FIG. 16) in FIG. ) Are repeated (until there are no lower-word unsearched extended words), the processes in steps S1702 to S1709 are repeated.

  In step S1702, the CPU 201 selects one extension word from the extension list 1106.

  Next, in step S1703, the CPU 201 determines whether the extended word selected in step S1702 has low-order word information. This determination is performed by searching the thesaurus dictionary DB 103 based on the extended word to obtain the extended word low-order word list 504 (FIG. 5) and determining whether there is low-order word information in the extended list 504.

  If it is determined in step S1703 that the extended word has low-order word information, the CPU 201 advances the process to step S1704.

  In step S1704, the CPU 201 repeats the processing in steps S1705 to S1709 for all the lower-words in the extended-word lower-word list 504 selected in step S1702 (until there are no unprocessed lower-words).

  In step S1705, the CPU 201 (thesaurus dictionary search hit term usefulness calculation means 1013) selects one low-order word from the low-word list 504 of the extended word selected in step S1702, and the extended word selected in step S1702 and the above-mentioned low-order word. Calculate the "cumulative weight" between words. This “cumulative weight” is calculated as a product of the cumulative weight of the extended word (extended weight 1202 in FIG. 12) and the weight between words of the lower word (similarity 602 in FIG. 6).

  Next, in step S1706, the CPU 201 determines whether or not the narrower word selected in step S1705 exists in the extended list 1106.

  If it is determined in step S1706 that the narrower word selected in step S1705 does not exist in the extended list 1106, the CPU 201 controls to perform the process in step S1709.

  In step S1709, the CPU 201 adds the narrower word selected in step S1705 to the end of the extended list 1106. Specifically, the CPU 201 adds an extended word to the end of the extended list 1106, adds the extended heading 1201 of the added extended word to the partner heading 601 of the lower-order word and the extended weight 1202 (usefulness of the extended search term). The “accumulated weight” calculated in step S1705 and the true / false value of the above-mentioned narrower term set in the thesaurus dictionary DB 103 are stored in the abstract flag 1203, and the process returns to step S1704.

  On the other hand, if it is determined in step S1706 that the narrower word selected in step S1705 exists in the extended list 1106, the CPU 201 controls to perform the processes in steps S1707 and S1708.

  In step S 1707, the CPU 201 (thesaurus dictionary search hit term usefulness calculating means 1013), the “cumulative weight” (extended weight 1202) of the above-mentioned low-order word already in the extended list 1106 and the “cumulative weight calculated in step S 1705. "Compare.

  In step S1707, if the “cumulative weight” calculated in step S1705 is larger (closer to the keyword), the CPU 201 advances the process to step S1708.

  In step S 1708, the CPU 201 rewrites the “cumulative weight” (extended weight 1202) of the lower-order word already in the extended list 1106 with the “cumulative weight” calculated in step S 1705, and returns the process to step S 1704.

  On the other hand, in step S1707, if the “cumulative weight” calculated in step S1705 is not greater (if it is farther from the keyword), the CPU 201 returns the process to step S1704 as it is. In the present embodiment, when there are a plurality of the same low-order words, the “cumulative weight” (greater usefulness) closer to the keyword is left.

  In step S1704, the CPU 201 executes the processing in steps S1705 to S1709 for all the lower-order words in the extended-word lower-word list 504 selected in step S1702 (until there are no unprocessed lower-order words). If it is determined, the process returns to step S1701.

  In step S1701, the CPU 201 executes the processing of steps S1702 to S1708 for all the extended words in the keyword extended list 1106, which is the extended result of the representative word (until there are no lower word unsearched extended words). If it is determined that the process has been performed, the process of this flowchart ends.

  In the following, the related word search process (expansion process from the representative word controlled and displayed in the synonym expansion to the related word) shown in step S1407 of FIG. 14 will be described with reference to the flowchart of FIG.

  FIG. 18 is a flowchart showing an example of the sixth control processing procedure in the present invention, and corresponds to the related word search processing shown in step S1407 of FIG. Note that the processing in this flowchart corresponds to the function of the thesaurus dictionary search unit 101 realized by the CPU 201 shown in FIG. 2 loading the program stored in the HDD 204 into the RAM 203 and executing it.

  First, in step S1801, the CPU 201 refers to the word related word list 505 (FIG. 5) in the thesaurus dictionary DB 103 corresponding to the keyword (hereinafter referred to as the keyword) controlled to the representative word in step S1404 in FIG. It is determined whether or not there is even one related word in the word list 505. For example, referring to the thesaurus diagram of FIG. 4, if the representative word is “Java”, the related words are “JSP”, “Servlet”, and “Swing”.

  If it is determined in step S1801 that there is no related word in the keyword related word list 505, the CPU 201 ends the process of this flowchart.

  On the other hand, if it is determined in step S1801 that there is at least one related word in the keyword related word list 505, the CPU 201 advances the process to step S1802.

  In step S1802, the CPU 201 performs the processing of steps S1803 to S1806 for all the related word lists 505 possessed by the keyword.

  In step S1803, the CPU 201 selects one related item from the related word list 505 that the keyword has, and determines whether or not the selected related word exists in the extended list 1106 that the keyword has.

  If it is determined in step S1803 that the related word is present in the extended list 1106, the CPU 201 controls to perform the process in step S1804.

  In step S1804, the CPU 201 (thesaurus dictionary search hit term usefulness calculating means 1013) calculates the “cumulative weight” (usefulness of the extended search term) of the related word. The “cumulative weight” of the related word is calculated by multiplying the “cumulative weight” of the keyword by the weight between the keyword and the related word. Then, the “cumulative weight” of the related word is compared with the “cumulative weight” (that is, the extended weight 1202) of the related word already existing in the extended list 1106, and the numerical value is larger (more representative words and words The one having the closest meaning is registered in the expansion weight 1202 (usefulness of the expanded search term) of the related word already existing in the expansion list 1106, and the process returns to step S1802.

  On the other hand, if it is determined in step S1803 that the related word does not exist in the extended list 1106, the CPU 201 controls to perform the process in step S1805.

  In step S1805, the CPU 201 adds the related word to the end of the extended list 1106. More specifically, the CPU 201 adds an extended word to the end of the extended list 1106, adds the related word partner heading 601 to the extended heading 1201 of the added extended word, and the abstract flag 1203 stores the related word in the thesaurus dictionary DB 103. The true / false value of the abstract flag 506 is registered.

  Next, in step S 1806, the CPU 201 (thesaurus dictionary search hit term usefulness calculating means 1013) calculates the “cumulative weight” (usefulness of the extended search term) of the related word. The “cumulative weight” of the related word is calculated by multiplying the “cumulative weight” of the keyword by the weight between the keyword and the related word. The CPU 201 registers the “cumulative weight” of the calculated related word in the extended weight 1202 of the added extended word, and returns the process to step S1802.

  If it is determined in step S1802 that the processes in steps S1803 to S1806 have been performed on all the related word lists 505 included in the keyword, the process of this flowchart ends.

  Hereinafter, the search term expansion processing results described in detail in FIGS. 14 to 18 with reference to FIGS. 19 to 22 will be exemplified.

  FIG. 19 is a diagram illustrating an example of an extended processing result after search in Article 1 of the search condition list 1000 illustrated in FIG.

  The example here corresponds to an extended processing result when the search condition condition header 1105 is “Oracle” and the mandatory flag 1107 is “TRUE”.

  In this case, since the essential flag 1107 is “TRUE”, it is not expanded to a broader word. Further, as shown in FIG. 3, since the low-order word “Oracle” is not registered in the thesaurus dictionary DB 103, it is not substantially expanded to a low-order word. However, since “SQL” is registered as a related word of “Oracle” in the thesaurus dictionary DB 103 as shown in FIG. 3, the related word “SQL” is registered in the extended list 1106.

  FIG. 20 is a diagram illustrating an example of an extended processing result after searching in Article 2 of the example of the search condition list 1000 illustrated in FIG.

  The example here corresponds to the extended processing result when the condition header 1105 of the search condition is “Java1.3” and the mandatory flag 1107 is “FALSE”.

  In this case, since the mandatory flag 1107 is “FALSE”, “Java1.3” is first registered with “Java” and “object-oriented language” as the broader words at the top of the extended list 1106 sequentially. Next, “Java1.4”, “C ++”, and “SmallTalk” are sequentially registered at the end of the extended list 1106 as low-order words. As shown in FIG. 3, the related word “Java1.3” is not registered in the thesaurus dictionary DB 103 and is not expanded to the related word.

  FIG. 21 is a diagram illustrating the extension word “Java1.3” in the extension list shown in FIG.

  FIG. 22 is a diagram illustrating an extension word “Java” in the extension list shown in FIG.

  Hereinafter, the human resource profile DB search process (a process of searching the human resource DB 104 for human resources corresponding to the expanded keyword input as a search condition) shown in step S803 of FIG. 8 will be described with reference to the flowchart of FIG.

  FIG. 23 is a flowchart showing an example of the seventh control processing procedure in the present invention, and corresponds to the human resource profile DB search processing shown in step S803 of FIG. Note that the processing of this flowchart corresponds to the function of the human resource search unit 102 realized by the CPU 201 shown in FIG. 2 loading the program stored in the HDD 204 into the RAM 203 and executing it.

  First, in step S2301, the CPU 201 stores the initialized search expression (that is, no search condition) in the RAM 203, and advances the process to step S2301.

  Next, in step S2302, the CPU 201 executes processing (steps S2303 to S2307) for creating a search expression for searching the human resource DB 104 for all the search conditions in the search condition list 1000.

  First, in step S2303, the CPU 201 selects one search condition for which a search expression has not yet been created from the search condition list 1000.

  In step S2304, the CPU 201 creates a “conditional expression” by connecting all of the search elements of the search condition selected in step S2303 (that is, extended words in the extended list 1106) with OR.

  Next, in step S2305, the CPU 201 determines whether the search condition selected in step S2303 is an essential condition.

  If it is determined in step S2305 that the search condition selected in step S2303 is an essential condition, the CPU 201 advances the process to step S2306.

  In step S2306, the CPU 201 creates (updates) a search expression. Specifically, when creating a retrieval formula for an essential condition, the CPU 201 creates (updates) the retrieval formula by connecting the retrieval formula stored in the RAM 203 with “AND” in the conditional formula created in step S2304. . For example, ““ search expression ”=“ search expression ”AND“ conditional expression ”” is created. Then, the CPU 201 returns the process to step S2302.

  On the other hand, if it is determined in step S2305 that the search condition selected in step S2303 is not an essential condition, the CPU 201 advances the process to step S2307.

  In step S2307, the CPU 201 creates (updates) a search expression. Specifically, when creating a search expression for a non-essential condition, the CPU 201 connects the search expression stored in the RAM 203 with the condition expression created in step S2304 with “OR” to create (update) the search expression. To do. For example, ““ search expression ”=“ search expression ”OR“ conditional expression ”” is created. Then, the CPU 201 returns the process to step S2302.

  If it is determined in step S2302 that a search expression for DB search has been created from all search conditions, the CPU 201 advances the process to step S2308.

  In step S2308, the CPU 201 searches the human resource DB 104 (for example, FIG. 24 described later) using the search formula created in step S2306 or step S2307.

  In step S2309, the CPU 201 creates the search results performed in step S2308 as a list (for example, FIG. 25 described later), and ends the process of this flowchart.

  FIG. 24 is a diagram illustrating an example of the human resource DB 104 illustrated in FIG.

  FIG. 25 is a diagram showing an example of a list of human resource search results created in step S2309 of FIG.

  In FIG. 25, reference numeral 2501 denotes a searched human resource. Reference numeral 2503 denotes each skill condition (search condition) used for the search.

  Reference numeral 2502 denotes a total point.

  FIG. 25 shows a human resource search result for which all scoring has been completed, but when the human resource profile DB search process shown in FIG. 23 is completed, total points 2502 have not yet been calculated. The total points 2502 are calculated by the following human resource scoring process.

  Hereinafter, the retrieved human resource scoring process shown in step S804 of FIG. 8 will be described with reference to the flowchart of FIG.

  FIG. 26 is a flowchart showing an example of the eighth control processing procedure in the present invention. The retrieved human resource scoring process (research result of human resource skill profile DB search process shown in step S804 of FIG. 8) is shown. Corresponding to the skill scoring of each human resource). Note that the processing of this flowchart corresponds to the function of the human resource search score calculation unit 1021 realized by the CPU 201 illustrated in FIG. 2 loading the program stored in the HDD 204 into the RAM 203 and executing the program.

  First, in step S2601, the CPU 201 performs scoring processing (steps S2602 to S2606) for all human resources included in the search result list (for example, FIG. 25) of the human resource skill profile DB search processing illustrated in FIG.

  First, in step S2602, the CPU 201 selects one human resource from the search result of the human resource skill profile DB search process shown in FIG. 23 and sets it as the focused human resource, and stores the score of the focused human resource stored in the RAM 203. Is initialized to 0 (total score).

  Next, in step S2603, the CPU 201 performs scoring (steps S2604 to S2606) for all skills possessed by the focused human resource.

  First, in step S2604, the CPU 201 selects one skill from the skills of the focused personnel, sets the focused skill, the experience time of the focused skill, and the extension weight (within the search condition) of the focused skill. The experience time is recalculated by multiplying the extension weight (usefulness) 1202 (corresponding to the score in the skill condition 2503 in FIG. 25) of the extension word corresponding to the skill under attention held in FIG.

  Next, in step S2605, the CPU 201 performs an attenuation process of the experience value using the experience time recalculated in step S2604 for the selected skill to obtain an assumed experience month number. Details of the attenuation process will be described later with reference to FIG.

  Next, in step S2606, the CPU 201 calculates a score for the skill subjected to the attenuation process in step S2605 using the following equation (1), adds the score to the total score initialized in step S2602, and performs the process in step S2603. return.

* Formula (1) is given a maximum of 100 points for convenience.
* Abs represents an absolute value.

  If it is determined in step S2603 that scoring has been completed for all skills of the focused human resource, the CPU 201 advances the process to step S2607.

  In step S2607, the CPU 201 executes bonus point addition processing from all skills of the focused human resource. The bonus point addition process will be described later with reference to FIG.

  Next, in step S2608, the CPU 201 adds the bonus points calculated in step S2607 to the total score, saves them in the RAM 203 as the total points 2501 (FIG. 25) of the focused human resource, and returns the process to step S2601. .

  When the CPU 201 determines in step S2601 that the scoring process (steps S2602 to S2606) has been completed for all the human resources included in the search result list of the human resource skill profile DB search process illustrated in FIG. The process of this flowchart is complete | finished.

  Hereinafter, the attenuation process shown in step S2605 of FIG. 26 will be described with reference to the flowchart of FIG.

  FIG. 27 is a flowchart showing an example of the ninth control processing procedure in the present invention, and corresponds to the attenuation processing shown in step S2605 of FIG. Note that the processing of this flowchart corresponds to the function of the human resource search score calculation unit 1021 realized by the CPU 201 illustrated in FIG. 2 loading the program stored in the HDD 204 into the RAM 203 and executing the program.

  First, in step S2701, the CPU 201 determines the skill of the target for calculating the attenuation of the experience value (hereinafter, the skill of interest) among the skills possessed by the target human resource for calculating the attenuation of the experience value (hereinafter, the target person of interest). ) Is acquired from the human resource DB 104, the rating date (ie, the end date 2401 shown in FIG. 24) indicating the date (date) determined. Then, using the acquired rating date (end date 2401) and the conversion date (the date previously entered by the system administrator or user and stored in the RAM 203) representing the date for calculating the attenuated experience value of each skill is used. Then, the number of blank months (that is, the elapsed period from the time when the skill of the searched human resource is determined) is calculated from the following equation (2), and the process proceeds to step S2702.

“Number of blank months = (year of conversion date−year of rating date) × 12 + month of conversion date−month of rating date” (2)
Next, in step S2702, the CPU 201 sets an attenuation start period (2801 in FIG. 28) and an attenuation end period (2802 in FIG. 28) of the skill of interest set in advance by the system administrator or user and stored in the HDD 204. ) To get. Then, the CPU 201 calculates an attenuation period (2803 in FIG. 28) from the acquired attenuation start period and attenuation end period using the following equation (3), and the process proceeds to step S2703.

“Attenuation period = Attenuation end period−Attenuation start period” (3)
FIG. 28 is a diagram for explaining the attenuation rate according to the number of blank months.

  Next, in step S2703, the CPU 201 compares the number of blank months calculated in step S2501 with the decay start period of the skill being noticed. As a result of the comparison, if the CPU 201 determines that the number of blank months is smaller than the decay start period, the CPU 201 advances the process to step S2706.

  In step S2706, the CPU 201 sets “1” (no attenuation) as the attenuation rate of the skill being noticed because the number of blank months is shorter than the attenuation start period, that is, attenuation has not started, and the process proceeds to step S2510. .

  On the other hand, if it is determined in step S2703 that the number of blank months is equal to or greater than the attenuation start period, the CPU 201 advances the process to step S2704.

  In step S2704, the CPU 201 compares the number of blank months calculated in step S2701 with the attenuation end period of the skill being noticed. As a result of the comparison, if the CPU 201 determines that the number of blank months is greater than the decay end period, the CPU 201 advances the process to step S2707.

  In step S2707, since the blank months have passed the decay end period, the CPU 201 sets the maximum decay rate of the skill being focused on the decay rate (2804 in FIG. 28; preset by the system administrator or user in the HDD 204). Stored) and the process proceeds to step S2510.

  On the other hand, if it is determined in step S2704 that the number of blank months is equal to or less than the attenuation end period, the CPU 201 advances the process to step S2705.

  In step S2705, the CPU 201 compares the decay start period and decay end period of the skill under consideration. As a result of the comparison, if it is determined that the attenuation start period and the attenuation end period are equal (that is, the attenuation period is “0”), the CPU 201 advances the process to step S2707.

  On the other hand, if it is determined in step S2704 that the attenuation start period and the attenuation end period are not equal (that is, the attenuation period is not “0”), the CPU 201 advances the process to step S2708.

  In step S2708, the CPU 201 calculates an attenuation slope from the attenuation opening calculated in step S2702 and the maximum attenuation rate according to the following formula (4), and calculates the attenuation blank months from the attenuation months of the blank months and the skill of interest ( 2805) of FIG. 28 is calculated from the following equation (5), and the process proceeds to step S2709.

“Attenuation slope = (1−maximum attenuation rate) / attenuation period” (4)
“Attenuation blank months = blank months−attenuation start period” (5)
In step S2709, an attenuation rate is calculated from the following equation (6) from the attenuation slope and attenuation blank months calculated in step S2708, and the process proceeds to step S2510.

“Attenuation rate = 1−Attenuation slope × Attenuation blank months” (6)
In step S2710, the CPU 201 determines the number of months of experience of the skill being focused on (experience time calculated in step S2604 in FIG. 26) and the maximum number of months of experience of the skill being focused (preset by the system administrator or user in the HDD 204). Compare).

  If it is determined in step S2710 that the experience month of the skill being focused on is equal to or less than the maximum experience month, the CPU 201 advances the process to step S2711.

  In step S <b> 2511, the CPU 201 replaces the experience month of the skill being focused with the maximum experience month, and advances the process to step S <b> 2512.

  On the other hand, if it is determined in step S2710 that the number of months of experience of the skill being focused on is greater than the maximum number of months of experience, the CPU 201 proceeds directly to step S2712.

  In step S <b> 2512, the CPU 201 calculates an assumed experience month from the calculated attenuation rate of the skill under consideration and the experience month by the following formula (7), and ends the experience value attenuation flow process.

"Assumed months of experience = decay rate x months of experience" (6)
Hereinafter, the bonus point processing (processing for adding bonus points to the human resource search result as the human resource skill profile DB search processing result) shown in step S2607 of FIG. 26 will be described with reference to the flowchart of FIG.

  FIG. 29 is a flowchart showing an example of the tenth control processing procedure in the present invention, and corresponds to the bonus point processing shown in step S2607 of FIG. Note that the processing of this flowchart corresponds to the function of the human resource search score calculation unit 1021 realized by the CPU 201 illustrated in FIG. 2 loading the program stored in the HDD 204 into the RAM 203 and executing the program.

  First, in step S2901, the CPU 201 calculates bonus points (processing in steps S2902 to S2907) for all the human resources in the human resource search result (FIG. 25) that is the human resource skill profile DB search processing result.

  In step S2902, the CPU 201 selects one human resource from the human resource search result and sets it as the focused human resource, and the number of “skills” and “search condition condition header 1105” held by the focused human resource are completely matched ( Hereinafter, the number (hereinafter referred to as “P check count number”) in which either “skill” or “synonym of search condition” possessed by the human resource of interest matches is counted.

  For example, Mr. A holds the skills of "Java1.3", "C language", and "SQL", and Mr. B holds the skills of "C ++" and "Windows (registered trademark) XP". Suppose that “C ++” is a synonym for “C language”.

  At this time, if the search conditions are “Java1.3” and “C language”, the representative word count of Mr. A is “2”, the extended word count is “2”, and the representative word count of Mr. B is “0” and the expanded word count are “1”.

  Next, in step S 2903, the CPU 201 initializes the bonus point of the focused human resource stored in the RAM 203 with 0.

  In step S2904, the CPU 201 performs determination based on the count number. If it is determined that the count number is not “0” or “1”, the CPU 201 advances the process to step S2905.

  In step S2905, the CPU 201 calculates a bonus point for the human resource from the counter number and the reference point (preset by the system administrator or user and stored in the HDD 204) using the formula “(count number−1) × reference point = bonus”. The calculation is performed based on “point”, and the process proceeds to step S2906.

  FIG. 30 is a diagram illustrating bonus points based on the number of matching with the search condition, and corresponds to an example in which the number of search conditions is 10.

  As indicated by 3001 in FIG. 30, bonus points are awarded according to the count number (matching number).

  That is, the human resource search score calculation unit 1021 adds bonus points corresponding to the number of skills that the search human resource has that completely matches any of the search terms to the total score of the searched human resource data.

  Returning to the flowchart of FIG.

  On the other hand, if it is determined in step S2904 that the count number is “0” or “1”, the CPU 201 proceeds to step S2906 as it is.

  In step S2906, the CPU 201 performs determination based on the P count number. If it is determined that the P count number matches the search number, the CPU 201 advances the process to step S2907.

  In step S2907, as shown in FIG. 30, the CPU 201 adds a perfect bonus point 3002 (preset by the system administrator or user and stored in the HDD 204) to the bonus point 3001 calculated in step S2905 and bonus. Points (final result value) are calculated. That is, “bonus points + perfect bonus points = bonus points (final result value)” is calculated.

  For example, to explain using Mr. A and Mr. B exemplified above, the search condition is “Java1.3” and “C language”, so the number of search conditions is “2”. Since Mr. A has a P count of “2”, a perfect bonus point is added. Because B has a P count of “1”, no perfect bonus points are added.

  That is, the human resource search score calculation means 1021 has a perfect bonus for the total score of the human resources when the skills of the searched human resource data match all of the search terms and the extended words expanded from the search terms. Add points.

  Then, when the process of step S2907 ends, the CPU 201 returns the process to step S2901.

  On the other hand, if it is determined in step S2906 that the P count number matches the search number, the CPU 201 returns the process to step S2901 as it is.

  If it is determined in step S2901 that bonus points have been calculated (steps S2902 to S2907) for all human resources in the human resource search result, the CPU 201 ends the process of this flowchart.

  The flow of scoring will be specifically described for the following skill conditions and the case of “Taro Yamada” in the human resource DB 104 of FIG.

  Search skill condition (conversion date: April 2007), skill condition 1 (skill "Java1.4", process "implementation", role "TM" experience time "54" months), skill condition 2 (skill "RDB" , Process “implementation”, role “TM” experience time “24” months), skill condition 3 (skill “UML2.0”, process “design”, role “TM” experience time “24” months).

  Note that “Java1.4” in the search skill condition 1 is extended to “Java1.3” or the like, and the extension weight of the extended word “Java1.3” is “0.8”. Further, “RDB” in the search skill condition 2 is expanded to “Oracle10.g” or the like, and the expansion weight of the expanded word “Java1.3” is “1”. Furthermore, “UML2.0” of search skill condition 3 is expanded to “UML1.5”, “UML2.0”, etc., and the extension weights of the expanded words “UML1.5” and “UML2.0” are “ 0.9 ”. Note that any of the extension weights may differ depending on the thesaurus dictionary of the constructed system, and the extension weights can be reviewed as appropriate.

  From the human resource DB 104 in FIG. 24, in the case of the human resource “Taro Yamada”, “Java1.3” is “60” months, “Oracle10.g” is “24” months, “UML1.5” is “32” months Is held.

  Taking the product of the extended weight of each search condition and the experience time and recalculating the experience time, for the human resource "Taro Yamada", "Java1.3" is "48" months, "Oracle10.g" is "24" months "UML1.5" will be "28.8" months of experience.

  In addition, the skills “Java1.3”, “Oracle10.g” and “UML1.5” of “Taro Yamada” have the end date of “2007/04”, so the number of blank months becomes “0” and decays There is no empirical decay in the process.

  If the score of each skill is calculated from the recalculated experience time using Equation (1), “Java1.3” is “88”, “Oracle10.g” is “100”, “UML1.5” is “ 80 ”.

  Therefore, the total score based only on the skills held by Taro Yamada is “268”.

  Since “Taro Yamada” has “0” as the number (count number) that matches the search conditions “Java”, “RDB”, and “UML”, no bonus points are added.

  However, Taro Yamada matches all the synonyms of the search condition, so perfect bonus points (here, “150”) are added to the total score. The perfect bonus points can be changed by the constructed system, and the bonus points can be appropriately reviewed.

  From the above, Taro Yamada's total score (final result value) is total score “268” + perfect bonus point “150” = “418”.

  As described above, the human resource search apparatus according to the present embodiment performs a search by flexibly using a thesaurus (higher / lower terms, synonyms, related terms) and expanding the search terms. When the term is extended to a very general term in the broader term, it is excluded from the search term because it is a “stop mark” indicating that no further superior search is performed and the thesaurus is extremely abstract. The search term is expanded in consideration of the meaning of the term by using the “abstract flag” indicating “.”

  And by using the above stop mark and abstract flag, it is possible to narrow down the terms in consideration of the meaning of the terms, rather than simply expanding the terms by the threshold or the number of times, and search for human resources that meet the conditions more It can be performed.

  In the above embodiment, the case where the search device according to the present invention is assumed to be used for human resource search has been described as an example. However, the present invention is not limited to the human resource search device, but other than the human resource DB. The present invention can also be applied to a search device for searching general databases.

  As described above, when the present invention is used for searching a general database, numerical data corresponding to the experience time 2402 of the human resource DB 104 (that is, the experience value for each skill of the human resource) is obtained as a search result depending on the configuration of the database. May not be included. In such a case, the numerical data corresponding to the experience time 2402 are all set to default numerical values (for example, “1”), and the scoring process shown in FIGS. 26 to 30 is executed.

  It should be noted that the configuration and contents of the various data described above are not limited to this, and it goes without saying that the various data and configurations are configured according to the application and purpose.

  Although one embodiment has been described above, the present invention can take an embodiment as, for example, a system, apparatus, method, program, or recording medium, and specifically includes a plurality of devices. The present invention may be applied to a system including a single device.

  As described above, the search device of the present invention is a system that flexibly uses a thesaurus (synonyms such as broader terms, narrower terms, and related terms) and expands search terms, and is a broader term of the thesaurus, A "stop mark" indicating that no higher-level search is performed when extended to a very general term, and an "abstract flag" indicating that the search term is not included because of a very abstract term in the thesaurus ”To expand the search term in consideration of the meaning of the term, so that the search term can be expanded to a very general high-level term or a search term to an abstract term in the search using the thesaurus dictionary It is possible to expand the search term in consideration of the meaning of the term while suppressing the expansion of the term, and it is possible to perform a search that matches more conditions.

  Therefore, unlike the conventional method of searching for synonyms, the search range is specified by the threshold value and the number of similarities, and only highly relevant ones are expanded depending on the search terms to be expanded. Search under appropriate conditions, suppressing the occurrence of differences such as expanding to unrelated words and expanding search terms to meaningless synonyms for search. It can be carried out.

  The configuration of a data processing program that can be read by the human resource search device 100 according to the present invention will be described below with reference to the memory map shown in FIG.

  FIG. 31 is a diagram illustrating a memory map of a recording medium (storage medium) that stores various data processing programs readable by the human resource search device 100 according to the present invention.

  Although not specifically shown, information for managing a program group stored in the recording medium, for example, version information, creator, etc. is also stored, and information depending on the OS on the program reading side, for example, a program is identified and displayed. Icons may also be stored.

  Further, data depending on various programs is also managed in the directory. In addition, when a program or data to be installed is compressed, a program to be decompressed may be stored.

  The functions shown in FIGS. 8, 14, 15, 16, 17, 18, 23, 26, 27, and 29 in the present embodiment are performed by a host computer by a program installed from the outside. It may be. In this case, the present invention is applied even when an information group including a program is supplied to the output device from a recording medium such as a CD-ROM, a flash memory, or an FD, or from an external recording medium via a network. Is.

  As described above, a recording medium in which a program code of software for realizing the functions of the above-described embodiments is recorded is supplied to the system or apparatus, and the computer (or CPU or MPU) of the system or apparatus is stored in the recording medium. It goes without saying that the object of the present invention can also be achieved by reading and executing the program code.

  In this case, the program code itself read from the recording medium realizes the novel function of the present invention, and the recording medium storing the program code constitutes the present invention.

  As a recording medium for supplying the program code, for example, a flexible disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, DVD-ROM, magnetic tape, nonvolatile memory card, ROM, EEPROM, A silicon disk or the like can be used.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an OS (operating system) or the like running on the computer based on the instruction of the program code. It goes without saying that a case where the function of the above-described embodiment is realized by performing part or all of the actual processing and the processing is included.

  Furthermore, after the program code read from the recording medium is written in a memory provided in a function expansion board inserted in the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that the case where the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing and the functions of the above-described embodiments are realized by the processing.

  Further, the present invention may be applied to a system composed of a plurality of devices or an apparatus composed of a single device. Needless to say, the present invention can be applied to a case where the present invention is achieved by supplying a program to a system or apparatus. In this case, by reading a recording medium storing a program represented by software for achieving the present invention into the system or apparatus, the system or apparatus can enjoy the effects of the present invention.

  Furthermore, by downloading and reading out a program represented by software for achieving the present invention from a server, database, etc. on a network using a communication program, the system or apparatus can enjoy the effects of the present invention. It becomes.

  In addition, all the structures which combined each embodiment mentioned above and its modification are also included in this invention.

  By providing another DB instead of the human resource DB 104 and inputting the search condition of the DB as the search condition, the DB can also be applied to search uses other than human resources.

1 is a system configuration diagram showing an example of a system configuration to which a search device according to the present invention can be applied. It is a block diagram which shows an example of the hardware constitutions of the personnel search apparatus 100 shown in FIG. It is the schematic diagram which showed typically an example of the thesaurus dictionary DB103 shown in FIG. It is the schematic diagram which showed typically an example of the weight attached | subjected to each word registered into the thesaurus dictionary DB103 shown in FIG. FIG. 5 is a data configuration diagram illustrating an example of a data configuration of each word registered in the thesaurus dictionary DB 103 illustrated in FIGS. 3 and 4. It is a data block diagram which shows the detailed structure of the high-order word information shown in FIG. 5, low-order word information (individual thing of 1-j), and related word information (individual thing of 1-k). It is a figure which shows an example of the word in the thesaurus dictionary DB103. It is a flowchart which shows an example of the 1st control processing procedure in this invention. It is a figure which shows an example of the search condition input screen 900 in this embodiment. 3 is a data configuration diagram illustrating an example of a data configuration of search conditions held in the human resource search device 100. FIG. It is a data block diagram which shows an example of the data structure of each search condition in the search condition list | wrist 1000 shown in FIG. It is a data block diagram which shows an example of a data structure of each extended word in the extended list 1106 shown in FIG. It is a figure which shows an example of the search condition list | wrist shown in FIG. It is a flowchart which shows an example of the 2nd control processing procedure in this invention. It is a flowchart which shows an example of the 3rd control processing procedure in this invention. It is a flowchart which shows an example of the 4th control processing procedure in this invention. It is a flowchart which shows an example of the 5th control processing procedure in this invention. It is a flowchart which shows an example of the 6th control processing procedure in this invention. It is a figure which shows an example of the extended process result after the search in the example 1 of the search condition list | wrist 1000 shown in FIG. It is a figure which shows an example of the extended process result after the search in the example 2 of the search condition list | wrist 1000 shown in FIG. FIG. 21 is a diagram illustrating an extension word “Java1.3” in the extension list shown in FIG. 20. FIG. 21 is a diagram illustrating an extension word “Java” in the extension list shown in FIG. 20. It is a flowchart which shows an example of the 7th control processing procedure in this invention. It is a figure which shows an example of human resource DB104 shown in FIG. It is a figure which shows an example of the list of the human resource search result produced by step S2309 of FIG. It is a flowchart which shows an example of the 8th control processing procedure in this invention. It is a flowchart which shows an example of the 9th control processing procedure in this invention. It is a figure for demonstrating the attenuation factor by the blank months. It is a flowchart which shows an example of the 10th control processing procedure in this invention. It is a figure explaining the bonus point by the number of matching with search conditions. It is a figure explaining the memory map of the recording medium (storage medium) which stores the various data processing program which can be read by the human resource search apparatus 100 which concerns on this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Personnel search apparatus 101 Thesaurus dictionary search means 1011 Thesaurus dictionary search end point determination means 1012 Thesaurus dictionary abstract term exclusion means 1013 Thesaurus dictionary search hit term usefulness calculation means 102 Personnel search means 1021 Personnel search meter detection calculation means 103 Thesaurus dictionary DB
104 Human Resources DB
105 Client communication means 201 Central processing unit (CPU)
202 ROM
203 RAM
204 HDD
205 Network interface

Claims (10)

A thesaurus dictionary search means for searching for synonyms of search terms using a thesaurus dictionary and extending the search terms;
Database search means for searching a database using search terms expanded by the thesaurus dictionary search means,
The thesaurus dictionary can set one or both of a stop mark indicating a general term and an abstract flag indicating an abstract term as a registered term.
The thesaurus dictionary search means includes a thesaurus dictionary search end point determination means that uses the term set with the stop mark as an end point of the upper search of the thesaurus dictionary, and a thesaurus that excludes the term set with the abstract flag from the search term expansion. And a dictionary abstract term excluding unit. The thesaurus dictionary is set with the similarity between registered terms,
The thesaurus dictionary search means includes a thesaurus dictionary search hit term usefulness calculation means for calculating the usefulness of the search term expanded by the thesaurus dictionary search from the similarity set between the terms in the thesaurus dictionary. Yes,
Search score calculation means for calculating a score based on the usefulness of the search term calculated by the thesaurus dictionary search hit term usefulness calculation means for each data searched by the database search means, The search device according to claim 1.   3. The search device according to claim 1, wherein the thesaurus dictionary search unit does not extend the search term to a broader term when the search term is set as an essential condition. 4. The database includes information indicating a point in time when data registered in the database is confirmed,
The search score calculation means calculates an attenuation rate according to an elapsed period from the time when the data searched from the database by the search means is determined, and attenuates the calculated score according to the attenuation rate. The personnel search device according to claim 2 or 3, wherein   3. The search score calculation unit adds a predetermined point to a score of data that completely matches one of the search terms among the data searched from the database by the search unit. 5. The search device according to any one of items 4 to 4.   The search score calculation means adds a predetermined point to the score of data that matches all of the search terms and the extended words expanded from the search terms among the data searched from the database by the search means. The search device according to any one of claims 2 to 5, wherein: The database is a human resource database in which experience values for each skill of human resources and information indicating the point in time when the experience values for each skill of the human resources are determined are registered,
The search score calculation means includes, for each human resource searched by the database search means, an experience value for each skill of the human resource and the usefulness of the search term calculated by the thesaurus dictionary search hit term usefulness calculation means. The search device according to any one of claims 2 to 6, wherein a score based on the above is calculated. A thesaurus dictionary search means for searching for synonyms of search terms using a thesaurus dictionary and expanding the search terms;
A database search means for searching a database using a search term expanded by the thesaurus dictionary search means, and
The thesaurus dictionary can set one or both of a stop mark indicating a general term and an abstract flag indicating an abstract term as a registered term.
The thesaurus dictionary search step includes a thesaurus dictionary search end point determination step in which the term set with the stop mark is set as the end point of the upper search of the thesaurus dictionary, and a thesaurus that excludes the term set with the abstract flag from the search term expansion. And a dictionary abstract term exclusion step. Computer
A thesaurus dictionary search means for searching for synonyms of search terms using a thesaurus dictionary and extending the search terms;
A program for functioning as database search means for searching a database using search terms extended by the thesaurus dictionary search means;
The thesaurus dictionary can set one or both of a stop mark indicating a general term and an abstract flag indicating an abstract term as a registered term.
The thesaurus dictionary search means includes a thesaurus dictionary search end point determination means that uses the term set with the stop mark as an end point of the upper search of the thesaurus dictionary, and a thesaurus that excludes the term set with the abstract flag from the search term expansion. A program characterized by comprising dictionary abstract term exclusion means. Computer
A thesaurus dictionary search means for searching for synonyms of search terms using a thesaurus dictionary and extending the search terms;
A program for functioning as database search means for searching a database using search terms extended by the thesaurus dictionary search means;
The thesaurus dictionary can set one or both of a stop mark indicating a general term and an abstract flag indicating an abstract term as a registered term.
The thesaurus dictionary search means includes a thesaurus dictionary search end point determination means that uses the term set with the stop mark as an end point of the upper search of the thesaurus dictionary, and a thesaurus that excludes the term set with the abstract flag from the search term expansion. A computer-readable recording medium on which a program having dictionary abstract term excluding means is recorded.

Images