JP5537124B2 - Attribute analysis method, program and system by first name and last name - Google Patents

Description

  The present invention relates to a method, a program, and a system for analyzing an attribute of an inputted first and last name as an analysis target.

  There are many membership-based services available on the Internet, whether paid or free. Normally, membership registration is required in order to receive the service provided by such a membership system service. When registering a member, for example, as shown in FIG. 12, a screen prompting input of information such as name, gender, address, telephone number, and mail address is displayed. Those who wish to register can input such information as appropriate and receive services for registered members.

  In the so-called job information site, by registering his / her professional skills, work history, educational background, etc., there is a possibility that the frequency of matching with job offers will increase.

  On the other hand, for example, in the above-described job information site, information that is disadvantageous for job hunting activities by registering is not necessarily positively input if it is an optional item. In addition, items that are prohibited from being collected for the purpose of deciding whether or not to accept (for example, parental occupation, family composition, permanent affiliation, etc.) from the request of the Employment Stabilization Law (Law No. 141, November 30, 1947) It cannot be set as a mandatory input item.

  As a recruiting company, from the viewpoint of labor saving in the selection process, primary selection (filtering) is performed according to attributes other than the items entered at the member registration stage of the job search information site or at the subsequent additional profile registration stage. There is a need to do it. Some of these attributes can be grasped to some extent from the name that is an essential input item.

  A surname dictionary that stores surname data consisting of surname words, their readings, their types, and frequencies Search all the names that match the name data input from the storage means and the word or part of the reading at the same time, and if there is a match, the name data input is either the last name or An analysis method for determining whether other than the above is disclosed (for example, Patent Document 1). In addition, as a character recognition processing device used when recognizing first and last names using an optical character reader (OCR), a partition unique to the first and last names according to the number of characters in order to correctly recognize the separation between first and last names A split processing unit that divides the first half into the latter half of the first half, and a first and last name recognition unit that compares the first half with the last name dictionary and compares the second half with the first name dictionary and recognizes the first name and last name, A character recognition processing device is disclosed that includes a recognition control unit that changes a partition of the division processing when recognition is impossible (for example, Patent Document 2).

JP 2004-318699 A Japanese Patent No. 2892376

  However, in Patent Document 1, what can be analyzed is whether the name is a corporation or an individual, and the attribute of the personal name cannot be analyzed from the inputted first and last names. Further, Patent Document 2 only shows a control for performing recognition processing again even when the separation of how far is a last name and where is a first name cannot be recognized by the original division processing. Also, it is impossible to analyze an attribute that can be recognized by associating it with a certain level from the first and last names.

  The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide a method, a program, and an analysis system for analyzing an attribute having an association with a first name and last name inputted.

  The attribute analysis method according to claim 1 is inputted from a step of accepting input of nominal data based on an external operation, and a last name dictionary storage unit storing last name data having at least last name and last name attribute information. The first search step for searching for the last name data that matches forward with the nominal data, and the name that stores the name data having at least the name and the name attribute information, and the name that matches backward with the input nominal data When there is no matching surname data or name data in any of the second search step for searching for data and the first search step and the second search step, (a) the nominal data is (B) all the nominal data is hiragana, (c) the nominal data is a predetermined number of characters or less, (d) the nominal data. And (e) determining whether it falls under any of special information that contains alphanumeric characters in the nominal data (claims). Item 1).

The steps after the second search step are performed only when the last matching first surname data is not searched in the first search step, and it is determined whether it does not correspond to any of (a) to (e). The step can be configured to be performed only when name data that matches backward in the second search step is not searched.

Moreover, it can also comprise as a program which makes a computer perform these attribute analysis methods (Claim 3).

The attribute analysis system according to claim 4 is an attribute analysis system for analyzing an attribute of nominal data from input nominal data, wherein the surname data having at least last name and last name attribute information is recorded. From the dictionary storage means, name dictionary storage means in which name data having at least a name and name attribute information is recorded, means for inputting nominal data, and last name data recorded in the last name dictionary storage means, First name processing means for searching for last name data that matches the entered nominal data first, and name data that matches the input nominal data from the name data recorded in the name dictionary storage means. A second processing means; (a) the nominal data is all katakana; (b) the nominal data is all hiragana; (c) the nominal data is No more than a fixed number of characters, (d) whether the nominal data includes at least a predetermined number of characters in the nominal data, and (e) whether the nominal data includes alphanumeric characters. Three processing means, and a judgment means for judging an attribute of the inputted nominal data based on processing results of the first processing means, the second processing means, and the third processing means. It is characterized (claim 4).

According to the first aspect of the present invention, part of the first and last names constituting the nominal data to be analyzed matches the last name data included in the last name dictionary or the name data included in the name dictionary, or corresponds to special information Depending on whether or not, it is possible to determine the attribute of the name holder of the name data appearing in the first and last names, for example, the place of birth. Accordingly, when an individual having a predetermined profile is extracted, primary filtering can be easily performed.

  According to the second aspect of the present invention, the attribute determination is completed without performing the subsequent processing when any one of the processing steps is satisfied, so that it is possible to perform a quicker determination processing. Become. These inventions may be executed by general-purpose hardware as a program for realizing the functions other than those realized by dedicated hardware (claim 3).

  According to the invention described in claim 4, a part of the surname that constitutes the nominal data to be analyzed matches the last name data included in the last name dictionary or the name data included in the name dictionary, or corresponds to special information Depending on whether or not, it is possible to determine the attribute of the name holder of the name data appearing in the first and last names, for example, the place of birth. Accordingly, when an individual having a predetermined profile is extracted, primary filtering can be easily performed.

1 is a block diagram showing an overall configuration of an attribute analysis system 100 according to the present invention including a service providing server 10. FIG. 2 is a block diagram illustrating a configuration of a service providing server 10. FIG. It is an image figure of the registration screen in which a job seeker registers as a member. It is an image figure of the registration screen where a job seeker also performs member registration. It is an image figure of the registration screen when a recruiting company registers a job offer. It is a conceptual diagram of job seeker data recorded in the job seeker information storage means. It is a conceptual diagram of the job offer item data recorded in the job offer information storage means. It is a flowchart which shows the flow of the attribute analysis based on nominal data. It is a figure which shows the aspect of the pre-processing of nominal data. It is a figure which shows the process of a front matching search with the structure of a last name dictionary, and nominal data. It is a figure which shows the structure of a name dictionary, and the process of a backward matching search with nominal data. It is a figure which shows the structure of a feature information storage part, and the discrimination | determination process of feature information applicable property. It is an image figure of the member registration screen in member system service.

  The form for implementing this invention is demonstrated in detail based on one Example. Here, a description will be given using an example in which analysis based on nominal data is applied to a job information site.

  FIG. 1 is a block diagram showing an overall configuration of an attribute analysis system 100 including a service providing server 10 according to an embodiment of the present invention. As shown in FIG. 1, the attribute analysis system 100 includes a service providing server 10, a plurality of user (job seeker) terminal devices 20 ... 20, and a plurality of user (job seeking company) terminal devices 30 ... 30. Are configured to be connectable to the network N.

  The service providing server 10 is an information processing apparatus that provides a job information service. Specifically, as will be described later, the job seeker registers his / her resume information in the job seeker information storage means 111 in the server 10 at the same time or after the member registration, Registered in the job information storage means 112 of

  The user (job seeker) terminal device 20 ... 20 is an information processing device having a communication function via the network N, such as a personal computer (PC), a personal digital assistant (PDA), or a mobile phone. The user (job seeker) terminal devices 20... 20 can be connected to the service providing server 10 via the network N.

  The user (recruitment company) terminal devices 30... 30 are typically personal computers (PCs), but the form is not limited as long as it has a communication function via the network N. User (recruitment company) terminal devices 30... 30 can also be connected to the service providing server 10 via the network N.

  The network N is a communication line for transmitting and receiving information between the service providing server 10 and user (job seeker) terminal devices 20... 20 and user (job seeking company) terminal devices 30. For example, it may be any of the Internet, LAN, WAN, dedicated line, telephone line, other communication line, and combinations thereof, and it does not matter whether it is wired or wireless.

  Here, regarding the configuration of the service providing server 10, a configuration particularly related to the present embodiment will be described by paying attention to its function.

FIG. 2 is a block diagram illustrating a configuration of the service providing server 10 according to the present embodiment. As shown in the figure, the service providing server 10 controls the operation and processing of storage means 11 for storing information (as a hardware configuration, for example, a memory such as a ROM and a RAM, and a storage device such as an HDD), and each unit. Control means 12 (for example, a processor such as a CPU), a communication means 13 for transmitting / receiving information to / from other nodes in the system via the network N, and these components are connected to each other to exchange information. A bus 14 is provided.

  The storage means 11 includes job seeker information storage means 111, job offer information storage means 112, last name dictionary storage means 113, name dictionary storage means 114, feature information storage means 115, and input format storage means 116. The control unit 12 includes a nominal data pre-processing unit 121, a surname data search unit 122, a name data search unit 123, a feature information determination unit 124, and an attribute determination unit 125.

  The job seeker information storage unit 111 records profile information of job seekers who have registered as members. 3A and 3B are screens displayed on the display means of the user (job seeker) terminal device 20 when the job seeker inputs his / her profile information from the user (job seeker) terminal device 20. These pieces of screen information are recorded in the input format storage means 116 and transmitted from the service providing server 10 to each user (job seeker) terminal device 20... 20 via the network N. Member information includes bibliographic information such as name, gender, date of birth, address, telephone number, e-mail address, and mobile e-mail address, as well as items described in the resume such as final educational background, work history, and self-promotion. It is. These pieces of information are recorded as job seeker data together with a member ID given at the time of member registration and a password set at the time of member registration.

  The job seeker inputs such information as a user (job seeker) terminal device 20 using his / her PC connected to the network, and transmits it to the service providing server 10 via the network N. The sent information is recorded as a database in the job seeker information storage unit 111. FIG. 5 is a conceptual diagram of job seeker data recorded in the job seeker information storage means. In order to contribute to applicant filtering, which will be described later, each item input at the time of member registration is recorded in a form arranged in a table.

  The job offer information storage unit 112 records job offer information registered by the job offer company. FIG. 4 is a screen for setting an applicant filter, which is a primary selection condition for applicants, in the registration screen for job offer information. Here, in addition to the essential conditions that must be met for selection (for example, whether you have reached the age at which you can work), the “filter” includes conditions that are positively or passively evaluated for selection (for example, whether or not you have part-time work experience). included. In this example, the filtering conditions for the recruiting job category “WEB designer” include age “under 18 years”, blank period (= not working) “within 3 months”, part-time job experience “assessment”, foreign applicants A filtering condition of “Accept” is set.

  In the registration of job posting information in the present embodiment, primary selection can be performed based on “experienced job type”, “number of job changes”, and “enrollment period of the most recent job”. In addition, the text of the automatic reply mail when filtering is applicable in the primary selection can be set in advance. The job offer company records job offer information from the user (job offer company) terminal device 30 together with the filtering conditions as described above in the job offer information storage unit 112 of the server 10 via the network N.

  FIG. 6 is a conceptual diagram of job offer item information recorded in the job offer information storage unit 112. For example, “Recruiting Company” Co., Ltd. Δ △ Design with “Company ID” 0002 has registered a Web designer as “Recruitment Type”. As filtering conditions, “age is under 18 years old”, the number of job changes is “unquestioned”, and foreign applicants are “present”. If the profile information of the job seeker who has applied for the job posting is determined not to be recruited as a result of these filtering, a preset automatic reply mail is distributed to the job seeker.

  Among the filtering conditions exemplified above, regarding data directly recorded in the job seeker's profile information, whether or not filtering can be executed is determined based on the data. On the other hand, there is no registration item for the region of origin in the job applicant's profile information. Therefore, the attributes related to the region of origin are analyzed based on the name in the profile information. FIG. 7 is a flowchart showing the flow of attribute analysis based on nominal data.

  When there is a request from a job seeker to apply for a job posting, the name data pre-processing means 121 performs pre-processing for analyzing the name data of the job seeker, that is, “name” and “kana” of profile information. (Step S701). Specifically, the space is removed from the character string inputted as the name and kana of the job seeker. As shown in FIG. 8, both full-width and half-width spaces are deleted to form a series of first and last names.

  Next, the last name data search unit 122 performs a forward matching search on the given character string using the data stored in the last name dictionary storage unit 113 (step S702). FIG. 9 is a conceptual diagram showing the process of forward matching search. The last name data search means 122 creates a plurality of search keys in which a given character string is added character by character from the beginning. In the case of “Sato Ichiro”, four patterns of search keys “S”, “Sato”, “Sato Ichi”, and “Sato Ichiro” are created. Similarly, multiple search keys are created for kana.

Next, a complete match search with the last name data recorded in the last name dictionary storage unit 113 is executed. The surname information storage means 113 accumulates surname data existing as Japanese surnames. In the case of “Ichiro Sato”, “Sato” and “Sato” exist in the last name dictionary (step S703: YES).

  As in the above example (“member ID” 0001 = Ichiro Sato), when there is a matching last name in the last name dictionary, the attribute determination unit 125 does not perform the subsequent search and discrimination process, “Japanese” is determined (step S708). However, for example, in the case of “member ID” 0004 = test Taro, there is no matching last name in the last match search for the last name dictionary (step S703: NO). In this case, the name data search means 123 performs a name dictionary backward match search (step S704). FIG. 10 is a conceptual diagram showing the process of backward matching search. The name data search means 123 creates a plurality of search keys in which a given character string is added character by character from the end. In the case of “Test Taro”, five patterns of search keys of “Taro”, “Taro”, “Taro”, “Sto Taro”, and “Test Taro” are created. Similarly, multiple search keys are created for kana.

  Next, an exact match search with the name data recorded in the name dictionary storage means 114 is executed. In the name information storage means 114, name data existing as Japanese names are accumulated. In the case of “Test Taro”, “Taro” and “Taro” exist in the name dictionary, respectively (step S705: YES).

  When there is a matching name in the name dictionary as in the above example (“member ID” 0004 = test Taro), the attribute determination means 125 does not perform subsequent determination processing, and the attribute determination means 125 converts the name data to “Japanese”. Is determined (step S708). However, for example, “member ID” 0005 = James. In the case of Johansen, there is no matching name in the name dictionary backward matching search (step S705: NO). In this case, a feature information appropriate determination process is performed by the feature information determination unit 124 (step S706). FIG. 11 is a conceptual diagram showing the feature information corresponding determination process.

  The feature information discriminating unit 124 counts the number of characters for each character type constituting the character string of the given nominal data. In the case of “James Johansen”, it is counted that Kanji = 0, Katakana = 10, Hiragana = 0, Alphanumeric = 0, and Symbol = 1.

On the other hand, in the present embodiment, the following five items are defined as feature information.
(A) All character strings are katakana (b) All character strings are hiragana (c) Three or fewer characters are included (d) There are eight or more katakana characters in the character string (e) Alphanumeric characters in the character string Is included

  The feature information discriminating unit 124 discriminates whether or not the character string of the nominal data given in light of the defined feature information corresponds to these. When it corresponds to any one of the feature information (step S707: YES), it is determined to be a foreigner (step S709), and the determination result is output (step S710). In the case of “James Johansen”, since it includes 10 katakana characters, it is determined to be applicable to the feature information (a), and is determined to be a foreigner. According to the output result, when the foreign applicant in FIG. 4 is targeted for filtering, it corresponds to filtering, and the automatic reply mail is sent from the service providing server 10 to the corresponding user (job seeker) terminal device 20. Sent.

  In addition to what is implement | achieved by exclusive hardware, this invention may be performed by general purpose hardware as a program for implement | achieving the function.

  As described above, by using the attribute analysis method according to the present invention, an attribute that is not registered as registered individual attribute information can be analyzed based on the registered nominal data. Thereby, primary selection in the case where it is necessary to select desired information from a large number of personal registration information can be facilitated.

  Although the above has described an example of an attribute analysis method suitable for filtering whether or not Japanese, this is not a limitation. For example, in a system that requires registration of genuine first and last names, It can also be applied to a filter for whether it is a stage name or a handle name. In this case, feature information such as “a symbol (♂, ♪, etc.) is included in the character string” can be defined as feature information defined in the feature information storage means. Further, in the present embodiment, the application example has been described as an application example in the recruitment site, but the application scene of the attribute analysis method according to the present invention is not limited to this, and the system has a need for discrimination as to whether it is a Japanese or not. It can be applied to anything.

  Furthermore, the accuracy of the last name dictionary information and the name dictionary information can be improved by increasing the data. In addition, the accuracy of erroneous determination can be improved by redefining the surname dictionary, name dictionary, or feature information afterwards.

DESCRIPTION OF SYMBOLS 10 Service provision server 11 Storage means 12 Control means 13 Communication means 14 Bus 20 User (job seeker) terminal device 30 User (job seeking company) terminal device 100 Attribute analysis system 111 Job seeker information storage means 112 Job offer information storage means 113 Last name dictionary storage Means 114 Name dictionary storage means 115 Feature information storage means 116 Input format storage means 121 Nominal data preprocessing means 122 Last name data search means 123 Name data search means 124 Feature information determination means 125 Attribute determination means

Claims (2)

An attribute analysis system that analyzes attributes of nominal data from input nominal data,
Last name dictionary storage means in which last name data is recorded,
Name dictionary storage means in which name data is recorded;
Means for entering nominal data;
First processing means for searching for last name data that matches the entered nominal data from the last name data recorded in the last name dictionary storage means;
Second processing means for searching for name data that matches the inputted nominal data from the name data recorded in the name dictionary storage means;
(A) All the nominal data is katakana.
(B) All the nominal data is hiragana,
(C) the nominal data is a predetermined number of characters or less;
(D) Katakana is included in the nominal data by a predetermined number of characters or more.
(E) Alphanumeric characters are included in the nominal data,
A third processing means for determining whether any of the special information is not applicable,
An attribute analysis system comprising: determination means for determining an attribute of the inputted nominal data based on processing results of the first processing means, the second processing means, and the third processing means. . The second processing means is performed only when the last matching last name data is not searched in the first processing means,
2. The attribute analysis system according to claim 1, wherein the third processing unit is performed only when name data that matches backward is not searched in the second processing unit. 3.

Images