JP2020107005A - Information providing device, system and program - Google Patents

Abstract

To provide an information providing device, system and program which can be easily installed even in circumstances where information is habitually exchanged via e-mail.SOLUTION: In accordance with this invention, the information providing device aggregates and provides information included in a plurality of e-mails and comprises a mail acquisition unit, a mail analysis unit, a mail distribution unit, and a provision information generation unit. The mail acquisition unit acquires e-mails destined for a preliminarily determined address, and the mail analysis unit analyzes the e-mails acquired by the mail acquisition unit to identify words included in the e-mails, and the mail distribution unit distributes the e-mails acquired by the mail acquisition unit, by preliminarily determined conditions on the basis of words identified by the mail analysis unit, and the provision information generation unit uses at least a part of the words identified by the mail analysis unit as an item to generate information resulting of listing the e-mails distributed by the mail distribution unit.SELECTED DRAWING: Figure 1

Description

The present invention relates to an information providing device, system and program.

Conventionally, there are many opportunities to perform information matching. For example, in real estate transactions, information on sellers and lenders is matched with information on buyers and borrowers. Also in the staffing business, it is necessary to match a project for recruiting human resources with a human resource suitable for the project.

Since such matching requires a great deal of labor, a device for supporting the matching has been proposed, for example, as described in Patent Document 1.

JP, 2005-327190, A

By the way, in recent years, many devices that can be connected to a computer network have become widespread, and e-mail is often used for exchanging information. In such a situation, by introducing a new device or system, it is often necessary to register the same information in the system or the like in addition to the conventional exchange of information by electronic mail.

The present invention has been made in view of such circumstances, and provides an information providing device, system, and program that can be easily introduced even in a situation where information exchange by e-mail is customarily performed. The purpose is to do.

According to the present invention, there is provided an information providing device that aggregates and provides information included in a plurality of electronic mails, and includes a mail acquisition unit, a mail analysis unit, a mail distribution unit, and a provided information generation unit, and acquires the mail. The unit acquires an email of a predetermined destination, the mail analysis unit analyzes the email acquired by the mail acquisition unit to identify words included in the email, and the mail distribution unit. Is based on the word specified by the mail analysis unit, sorts the electronic mail acquired by the mail acquisition unit for each predetermined condition, the provision information generation unit, the word specified by the mail analysis unit There is provided an information providing device that generates information in which electronic mails sorted by the mail sorting unit are listed with at least a part of the items as items.

FIG. 1 is a diagram showing an overall configuration of a system including an information providing system 1 and an information providing device 11 according to an embodiment of the present invention. 3 is a diagram showing a configuration of an information providing device 11. FIG. 6 is a flowchart showing a flow of processing of the information providing device 11. It is the figure which showed the example of the state which selected the case list on the screen displayed on the terminal 32 grade. It is the figure which displayed the detail of the case information from the state of FIG. It is the figure which displayed the detail of the personnel information from the state of FIG. It is the figure which showed the example of the state which selected the human resources list|wrist on the screen displayed on the terminal 32 grade. It is the figure which displayed the detail of the personnel information from the state of FIG. It is the figure which displayed the detail of the case information from the state of FIG. It is an example of a display screen by an automatic matching function. It is an example of an offer management screen 208. 9 is a flowchart showing a flow of operations of the offer management unit 120. 13 is a flowchart showing the flow of operations in S207 shown in FIG.

Embodiments of the present invention will be described below with reference to the drawings. The various features shown in the embodiments described below can be combined with each other.

1. Overall Configuration FIG. 1 is a diagram showing an overall configuration of a system including an information providing system 1 and an information providing apparatus 11 according to an embodiment of the present invention.

The information providing system 1 includes an information providing device 11, a mail server 12 and a web server 13, and the mail server 12 and the web server 13 are connected to a network 2. The mail server 12 and the web server 13 do not necessarily have to be dedicated to the information providing system 1, and those used in other systems can be used.

The network 2 is, for example, the Internet, and systems such as the dispatch company 3, the dispatch company 4, the dispatch company 5, and the dispatch company 6 are also connected.

The dispatching company 3 has a mail server 31, terminals 32 to 34, etc., each of which is connected to the network 2. Although not shown, the dispatching company 4, the dispatching company 5, and the dispatching company 6 also have mail servers and terminals, like the dispatching company 3. Note that the systems of the dispatching companies 3 to 6 may be different from those shown in the figure, and if the person in charge (employee) of each company can send and receive e-mail and use a web browser, he or she can May use a mobile terminal. Here, the information providing system 1 according to the present embodiment can be used by other users, not limited to the dispatching companies 3 to 6. For example, a system operated by SIer or a small-to-medium-sized system development company may be connected to the information providing system 1.

In such a configuration, the person in charge of each company, together with another person in charge of the company or the person in charge of another company, information on the ordered project (hereinafter referred to as “project information”) or information on human resources that can be introduced (hereinafter referred to as “personal information”) , Personnel information) will be exchanged by e-mail.

The information providing system 1 collects the case information and the personnel information included in the emails exchanged between the persons in charge. The information providing system 1 issues an electronic mail address for using the information providing system 1 to a person in charge of the dispatching company 3 and transfers the mail received by the mail server 31 to the electronic mail address concerned, thereby You can get mail. Note that the acquisition route of the electronic mail is not limited to this, and for example, the broadcast destination of the electronic mail exchanged between the persons in charge includes an address that the information providing device 11 can receive via the mail server 12, or The mail server 31 or the like may be set to be transferred to the address. Hereinafter, these mail addresses are collectively referred to as a designated address.

2. Configuration of Information Providing Device 11 FIG. 2 is a diagram showing the configuration of the information providing device 11. The information providing device 11 aggregates and provides information included in a plurality of electronic mails. As shown in FIG. 2, the information providing device 11 includes a mail acquisition unit 111, a mail analysis unit 112, a mail division unit 113, a mail distribution unit 114, a resend determination unit 115, a commercial flow depth determination unit 116, and a matching element extraction unit. 117, a matching processing unit 118, a provision information generation unit 119, an offer management unit 120, a learning unit 121, and a storage unit 122.

The information providing apparatus 11 operates a general-purpose computer according to a program, and acquires a mail acquisition unit 111, a mail analysis unit 112, a mail division unit 113, a mail distribution unit 114, a resend determination unit 115, and a commercial flow depth determination unit 116. The functional units of the matching element extraction unit 117, the matching processing unit 118, the provision information generation unit 119, the offer management unit 120, and the learning unit 121 can be realized. The storage unit 122 can be realized by using a storage device such as a hard disk drive included in the computer.

The mail acquisition unit 111 acquires electronic mail of a predetermined destination. Here, the predetermined e-mail is, for example, a designated address for using the information providing system 1. In the present embodiment, the mail acquisition unit 111 acquires the electronic mail of the designated address from the mail server 12, and stores the acquired electronic mail in the storage unit 122. Here, it is preferable that the e-mail on the mail server 12 is deleted when the mail acquisition unit 111 acquires the e-mail.

The mail analysis unit 112 analyzes the electronic mail acquired by the mail acquisition unit 111 and identifies the words included in the electronic mail. In the present embodiment, the mail analysis unit 112 analyzes the electronic mail acquired by the mail acquisition unit 111 and stored in the storage unit 122. The analysis of the electronic mail is performed on the title or the text of the electronic mail. For example, the text of the electronic mail is divided into a plurality of words by morphological analysis using a dictionary prepared specially. At the time of this division, words located before and after a character or symbol that is often used as a separator such as a colon or a space are associated. For example, if “place: Akasaka Mitsuke” is included in the e-mail, the “place” and “Akasaka Mitsuke” are associated. The divided words are stored in the storage unit 122 in association with the original electronic mail. The words stored in the storage unit 122 may be only nouns or may include words other than nouns. The analysis may use a method other than the morphological analysis. Furthermore, the analysis of the electronic mail by the mail analysis unit 112 may reflect the result of machine learning by the learning unit 121 described later.

When the e-mail acquired by the mail acquisition unit 111 includes a plurality of item information or a plurality of personnel information, the mail dividing unit 113 divides the information included in the e-mail for each item information or each personnel information. To do. In the present embodiment, it is determined whether or not the electronic mail acquired by the mail acquisition unit 111 includes a plurality of item information or a plurality of personnel information, and a plurality of item information or a plurality of personnel information is included. When it is determined that these are divided into individual item information or personnel information. The determination as to whether a plurality of item information or a plurality of personnel information is included is made based on the analysis result by the mail analysis unit 112.

Specifically, the mail dividing unit 113 determines a delimiter in the electronic mail and divides the electronic mail text into some blocks. Next, machine learning is executed by a learning unit 121, which will be described later, and it is determined whether a plurality of item information or personnel information is described in the email based on the number of appearances of the word. Then, if the degree to which it is estimated that a plurality of item information or personnel information is described in the e-mail exceeds a predetermined value, a “plurality flag” may be set.

For example, in the case of an email containing item information, if multiple words such as “reward” and “place” are included, it is highly possible that the email contains multiple item information. Even if a plurality of words such as “Java (registered trademark)” are included, a plurality of item information items are not necessarily included. In this way, the mail dividing unit 113 counts the number of appearances of each word, and the count of the number of appearances of a specific word that rarely appears multiple times in one item information or one personnel information statistically is calculated. When the number is equal to or more than a predetermined number, it can be determined that the email contains a plurality of item information or a plurality of personnel information. Further, when a plurality of flags are set, the mail dividing unit 113 calculates, for example, tf-idf for each word in each block, and if the total tf-idf of the block is large, one item information or human resource information. You may count as a block (collection) regarding.

The mail distribution unit 114 distributes the electronic mail acquired by the mail acquisition unit 111 based on the words specified by the mail analysis unit 112, according to predetermined conditions. In the present embodiment, the mail distribution unit 114 introduces the e-mail acquired by the mail acquisition unit 111 based on the analysis result by the mail analysis unit 112, including the case information indicating the case requiring human resources. It is divided into one that includes human resource information indicating possible human resources. In this distribution, it is determined whether the e-mail contains the case information or the personnel information based on the characteristics of the words contained in the e-mail. For example, if the e-mail contains words such as "project name" and "request skill", the probability that the e-mail contains the project information is high, and words such as "name" and "desired unit price" are included in the e-mail. Is included, the probability that the e-mail contains personnel information is high. Further, the e-mail determined not to include both the case information and the personnel information may be deleted from the storage unit 122. Furthermore, the result of machine learning by the learning unit 121 described later may be reflected in the distribution of electronic mail by the mail distribution unit 114. Further, the e-mail may be distributed by not considering the number of appearances of each word, but by distinguishing the feature of the word by, for example, a neural network.

Further, the mail distribution unit 114 further distributes each of the e-mails determined to include the case information and the e-mails determined to include the personnel information, for each job type such as development and infrastructure construction.

The resend determination unit 115 determines whether the email acquired by the email acquisition unit 111 includes the same information as the information already included in the received email, based on the word identified by the email analysis unit 112. To judge. In the present embodiment, the resend determination unit 115 measures the distance of each electronic mail as a document between the electronic mails. For the measurement of this distance, for example, a technique called a Levenshtein distance for measuring a distance between documents can be adopted. When the distance between the emails is measured to be closer than a predetermined value, the resend determination unit 115 determines that these emails have been resent, that is, mediation has not been established, Set the resend flag for the corresponding email. Further, the resend determination unit 115 uses the natural language processing technique or the like instead of measuring the distance between the emails, and the email acquired by the email acquisition unit 111 completely matches the content of the email already received. In this case, a resend flag may be set in the email. Further, the result of machine learning by the learning unit 121 described below may be reflected in the setting of the retransmission flag by the retransmission determination unit 115.

The commercial flow depth determination unit 116 determines the commercial flow depth for the item or the human resource based on the word specified by the mail analysis unit 112. In the present embodiment, the commercial flow depth determination unit 116 measures the document distance between each electronic mail. For the measurement of this distance, for example, a technique called a Levenshtein distance for measuring a distance between documents can be adopted. When the distance between emails is measured to be shorter than a predetermined value, the commercial flow depth determination unit 116 may compare the reward amounts of the item information or the personnel information included in these emails. Here, in general, the deeper the commercial flow depth of a project, the larger the amount of payment to the intermediary, while apparently reducing the reward presented to the human resources side for the project. On the other hand, the deeper the distribution depth of a human resource, the more the commission paid to the intermediary will be added to the remuneration for the human resource, so that the remuneration required by the human resource for the project side is apparently increased. If there is a difference in the reward presented by the case or the reward required by the human resources as compared with the case where the depth of business flow is shallow, these are cases in which multiple intermediaries intervene, that is, the commercial flow (the number of intermediaries ) Assuming that the items have different depths, the commercial flow depth is set in each item information or personnel information. Further, instead of measuring the distance between emails, the commercial flow depth determination unit 116 performs the same processing as the matching processing performed by the matching processing unit 118 described below between item information or personnel information, and the matching score You may make it perform the determination process of a commercial flow depth with respect to the case information or personnel information higher than a predetermined value. In addition, when the analysis result by the mail analysis unit 112 includes “commercial flow”, that is, the case information or the personnel information includes the depth of “commercial flow” (for example, “one company business partner human resource”). If so, the commercial flow depth determination unit 116 may set the commercial flow depth included in the item information or the personnel information as it is.

Further, the result of machine learning by the learning unit 121 described later may be reflected in the setting of the commercial flow depth by the commercial flow depth determination unit 116.

Furthermore, if an email containing the item information or personnel information for which the distribution depth is to be determined is sent from your company, the depth of the distribution will be lower than when such email is sent from another company. The commercial flow depth may be set so that it becomes one step shallower. For example, when the dispatching company 4 sends an e-mail (e-mail including manpower information) of a personnel proposal to the dispatching company 3 for the engineer A, who is an employee of the company, when the e-mail text says “commercial flow”. : A regular employee of our company” is used. The depth of commercial flow in this case is “one company ahead” from the perspective of the person in charge of the dispatching company 3. Next, when the dispatching company 3 sends an e-mail (e-mail including manpower information) of a personnel proposal to another company such as the dispatching company 4 for the engineer B who is an employee of the company, he/she designates it as a broadcast destination When the address is included, the personnel information included in the electronic mail is input to the information providing system 1. At this time, it is assumed that an expression such as “commercial flow: a regular employee of our company” is used in the body of the e-mail, similar to the e-mail sent by the temporary agency 4. In this case, even if the expressions regarding commercial flow (commercial flow: regular employees belonging to our company) are the same, the depth of commercial flow seen from the person in charge of the dispatching company 3 is set as "0 company ahead" (that is, the company). You may. As a result, it becomes possible to appropriately manage the e-mail generated by the company, as compared with the case where the commercial flow depth is mechanically set.

Here, in order to determine whether or not the received e-mail is sent from the company, for example, the domain of the e-mail address described in “From” included in the header information of the received e-mail (sender) Domain) can be used. Specifically, the user of the information providing system 1 is requested to notify the domain (user domain) of his/her own email address in advance, and the user domain is stored in the storage unit 122. Then, when the source domain matches any of the user domains stored in the storage unit 122, it can be determined that the e-mail corresponding to the receiving domain is sent from the company.

The matching element extraction unit 117 extracts a matching element from each of the case information and the personnel information distributed by the mail distribution unit 114. In the present embodiment, the matching element extraction unit 117 extracts, from among the words specified by the mail analysis unit 112, a predetermined word as a matching element used for matching the case information and the personnel information. As the matching element, for example, the following elements can be used.

(1) Items that can be acquired mechanically from the body information of emails, email subject, email sender, email date and time, etc.

(2) Items that can be acquired with regular expressions/commercial flow (regular employees belonging to the company, BP1 company freelance, etc.)
・Remuneration, age, number of interviews, nationality, nearest station, technical details (Java (registered trademark), detailed design, etc.)

(3) Items that can be statistically acquired ・Important points (This can be achieved by distinguishing "important lines related to matching" from the words contained in each line and putting a break from unimportant parts such as signatures and greetings)

In this embodiment, the matching element is extracted from the analysis result of the mail analysis unit 112 stored in the storage unit 122. For example, if the name of the matching element is “place”, it is associated with “place”. A word that has been read, for example, "Akasaka Mitsuke" is extracted as the value of the matching element. The name of the matching element extracted by the matching element extraction unit 117 may be determined by the learning unit 121 described later. For example, even if the place and place name are written in different places in the email due to an error in the email author, etc., by using machine learning, it will be written in the inappropriate place in the email. The place name that exists can be determined as a matching element.

The matching processing unit 118 calculates the matching score between the case information and the personnel information distributed by the mail distribution unit 114, based on the matching element extracted by the matching element extraction unit 117. In addition, the matching processing unit 118 may calculate a matching score for each combination of each item information and each personnel information based on the matching element extracted by the matching element extraction unit 117. Further, the matching score may be calculated in a brute force manner for each job type sorted by the mail sorting unit 114. For example, if there are 1,000 e-mails distributed to the item information by the mail distribution unit 114 and 1,000 e-mails distributed to the personnel information, 1,000,000 (1,000 x A matching score of 1,000) is calculated. For such calculation, for example, a neural network can be used. The matching score is calculated by approximating each matching element name, for example, the value of the matching element associated with each of the reward amount, age, number of interviews, nationality, place (nearest station), technical content, etc. It can be performed by giving a predetermined score according to the degree and adding them. Here, the determination of the degree of approximation between matching elements is preferably performed using the learning result of the learning unit 121.

The provided information generation unit 119 generates information listing the electronic mails sorted by the mail sorting unit 114 using at least a part of the words specified by the mail analysis unit 112 as items. Here, the “item” is used when inputting a condition for searching (limiting) the item information. In the present embodiment, the provision information generation unit 119 generates information for displaying a screen provided by the web server 13 in response to a request from a terminal (for example, the terminal 32). For example, when the provision information generating unit 119 receives a request to display a list of case information from the web server 13, the provision information generating unit 119 generates information for displaying the case information as a list. In the case information list, for each case information, in addition to the title of the e-mail containing the case information, the date and time of reception, the sender, etc., the commercial flow, the amount of compensation, and the age used by the matching processing unit 118 as matching elements. The number of interviews, nationality, place, technical content, etc. can be included as items.

Further, when the list includes the item information or the personnel information for which the retransmission flag is set by the retransmission determination unit 115, the provision information generation unit 119 can confirm that the item information or the personnel information is a retransmission. When the information to be displayed is generated and the case information or the personnel information for which the commercial depth is set by the commercial depth determination unit 116 is included in the list, only the case information or the personnel information with a shallow commercial depth is displayed. It is preferable to generate such information.

In addition, when the request from the web server 13 includes the search condition, for example, when the search condition includes that the number of interviews is “1”, the provided information generating unit 119 determines the number of interviews. The information for displaying only the item information of 1 is generated. When the request from the web server 13 includes the sort instruction, for example, when the ascending order of the reward amount is included as the sort instruction, the provision information generating unit 119 sets the item information in the ascending order of the reward amount. Information for rearranging so as to be displayed as a list may be generated.

Further, when the request from the web server 13 includes the information indicating that the specific case information is selected, the provision information generating unit 119 determines that the matching information for the selected case information is high. Information for displaying the list may be generated.

Similarly, when the provision information generation unit 119 receives a request for displaying a list of personnel information from the web server 13, the provision information generation unit 119 generates information for displaying the personnel information, and the personnel corresponding to the search condition or the sort designation. Information for displaying information is also generated. Further, when the request from the web server 13 includes information indicating that the specific human resource information is selected, the provision information generating unit 119 determines that the item information having a high matching score with respect to the selected human resource information. Information for displaying the list may be generated.

The offer management unit 120 corresponds to a case where a set of case information and human resource information is designated for the information displayed by the web server 13 on the terminal 32 or the like based on the information generated by the provision information generation unit 119. A set of item information and personnel information is subject to offer management. Further, the offer management unit 120 provides the offer management screen 208 (see FIG. 11) via the web server 13 based on the case information and the human resource information that are subject to offer management. In the offer management screen 208, it is possible to instruct the ordering source or the intermediary of the relevant item and the automatic creation or assistance of creation of an offer email to be sent to the source of the personnel or the intermediary, and also the result of the offer (approval). Or, you can enter the result of the interview, etc.

In addition, the offer management unit 120 stores a combination of the case information and human resource information targeted for offer management, the result of the offer input on the offer management screen 208, the result of the interview, and the like in the storage unit 122, and stores them in a log. It is provided as learning data to the learning unit 121. These results can be performed via the input screen for pressing the result input button 215b (see FIG. 11) and inputting each result.

The learning unit 121 executes machine learning and determines a word to be extracted as a matching element based on the result of machine learning. In the present embodiment, the learning unit 121 performs machine learning, for example, learning by a three-layer neural network. Here, the hierarchy of the neural network is not limited to this, and any hierarchy can be used. The learning is performed by using the log generated by the offer management unit 120, that is, the result of whether the offer is approved or the result of the interview, as teacher data, and the result obtained by the matching processing unit 118. It will be used for matching processing.

The result of learning by the learning unit 121 can be used for the processing by the mail distribution unit 114. In this case, it is determined from the characteristics of the words included in the email whether the email includes the case information or the personnel information, and the job type is determined, and the provision information generation unit 119 lists the case information. When the information for displaying the list of personnel information is provided, an error in the provided information can be accepted via the web server 13 and can be used as teacher data. For example, if "Embroidery freedom" is mentioned in the e-mail, and if it is statistically significant that the recruitment for "a smartphone app development project" is found, the e-mail may be sorted using such knowledge. it can.

Here, as a result of the learning by the learning unit 121, the mail analysis unit 112, the mail division unit 113, the mail distribution unit 114, the resend determination unit 115, the commercial flow depth setting unit 116, the matching element extraction unit 117, the matching processing unit 118. Can be used for the processing executed by.

Each of these components may be realized by software or hardware. When realized by software, various functions can be realized by the CPU executing programs. The program may be stored in a built-in storage unit or a computer-readable non-transitory recording medium. Alternatively, a program stored in an external storage unit may be read and realized by so-called cloud computing. When it is realized by hardware, it can be realized by various circuits such as ASIC, FPGA, or DRP. In this embodiment, various kinds of information and a concept including the same are dealt with, but these are represented by a high or low of a signal value as a binary bit aggregate formed of 0 or 1, and the software or hardware described above is used. The communication and the calculation can be executed depending on the mode.

3. Process Flow of Information Providing Device 11 Next, a process flow of the information providing device 11 will be described.

FIG. 3 is a flowchart showing the flow of processing of the information providing device 11.

(S101)
In the information providing device 11, first, the mail acquisition unit 111 acquires an electronic mail of a predetermined destination (addressed to the information providing system 1) from the mail server 12, and stores the acquired electronic mail in the storage unit 122.

(S102)
Then, the mail analysis unit 112 analyzes each of the electronic mails acquired by the mail acquisition unit 111 and stored in the storage unit 122, and stores the words obtained by the analysis in the storage unit 122 in association with the corresponding electronic mails. Let

(S103)
Next, the mail dividing unit 113 determines whether each of the emails acquired by the mail acquisition unit 111 and stored in the storage unit 122 includes a plurality of item information or a plurality of personnel information. The result of analysis by the mail analysis unit 112 is used for the determination.

(S104)
As a result of the determination in S103, when the mail dividing unit 113 determines that a plurality of item information or personnel information is included (YES in S103), the email dividing unit divides the email for each item information or personnel information and divides the email. The stored case information or personnel information is stored in the storage unit 122.

(S105)
Next, the mail distribution unit 114 determines that both the electronic mail divided by the mail dividing unit 113 (YES in S103) and the electronic mail not requiring division (NO in S103) are included in the electronic mail including the item information. E-mail containing personnel information. The mail distribution unit 114 uses the analysis result of the mail analysis unit 112 for distribution of electronic mail.

(S106)
Subsequently, the mail distribution unit 114 further distributes each of the electronic mails distributed in S105, by job type. Here, the job type means development, infrastructure, and others (Web designer, etc.), and may include categories other than these.

(S107)
Next, the resend determination unit 115 determines whether or not the electronic mail sorted by the mail sorting unit 114 has been resent.

(S108)
The resend determination unit 115 sets a resend flag for the e-mail determined to be resent (YES in S107). On the other hand, if it is determined that the sorted electronic mail has not been retransmitted, that is, it is a new electronic mail (NO in S107), S108 is omitted and the process proceeds to S109.

(S109)
Next, the commercial flow depth determination unit 116 determines the commercial flow depth of the electronic mail distributed by the mail distribution unit 114, and sets the determined depth to the corresponding electronic mail.

(S110)
Then, the matching element extraction unit 117 extracts a matching element from each of the electronic mails sorted by the mail distribution unit 114 into those containing the case information and those containing the personnel information.

(S111)
When the matching element extraction unit 117 extracts the matching element, the matching processing unit 118 performs the matching process of the case information and the personnel information based on the matching element.

(S112)
The result of the matching processing performed by the matching processing unit 118 is used when the provision information generation unit 119 generates information to be provided via the web server 13.

The processing of S101 to S111 described here is performed every arbitrary time or every predetermined time, and the processing of S112 is performed when the provision information generation unit 119 is requested to provide information via the web server 13. May be done.

4. Information Generated by Provision Information Generating Unit 119 Next, referring to FIG. 4 to FIG. 11, refer to an example of a screen provided by the web server 13 for information generated by the provision information generating unit 119 based on the information. explain.

When the web server 13 finishes the authentication process of the terminal 32 and requests the information of the screen to be displayed on the terminal 32 from the provision information generation unit 119, the provision information generation unit 119 first displays the screen 201 as shown in FIG. Generate information for display.

<Item information>
On the screen 201, a tab 211a for displaying a list of item information, a tab 211b for displaying a list of human resource information, a tab 212a for displaying a combination of automatically matched items and human resources, an offer management screen 208 212b for displaying the item information, an input field 212 for inputting a search (limitation) condition for the item information, an item name of each item included in the item information, and an instruction for instructing to sort the item information by item name An item display field 213 for displaying a button, a case information list display field 214 for displaying a list of case information, and a detail display button 215 for displaying details of the case information are included. Here, the detail display button 215 is not essential, and the details of the case information may be displayed by selecting a row representing the case, for example. The same applies to the detail display button 215 in other figures.

When the tab 211a is selected, the screen 201 that displays a list of item information is displayed. When the tab 211a is selected, the screen 201 is displayed.

The tab 211b is for displaying a screen 202 that displays a list of personnel information when selected. When the tab 211b is selected, the provision information generation unit 119 generates information for displaying the screen 204 shown in FIG.

The tab 212a, when selected, displays a screen 207 (described later with reference to FIG. 10) that displays a combination of the case information and the personnel information generated by the automatic matching function of the matching processing unit 118.

The tab 212b is for displaying the offer management screen 208 (described later with reference to FIG. 11).

The input column 212 is a column for inputting conditions for searching (limiting) the item information to be displayed, and each item of the item information, such as commercial distribution, remuneration amount, age, number of interviews, nationality, place, technical content, etc. The item information to be displayed can be searched (limited) by inputting a numerical value or a character string for each.

The item display column 213 is a column where the item name of each item of the item information to be displayed and an instruction button for instructing sorting of the item information by item name are displayed. The instruction button for instructing sorting uses, for example, symbols such as arrows and triangles as symbols.

The case information list display field 214 is a field in which the case information candidates are displayed according to the conditions input in the input field 212 in the case information list.

Here, for each case or personnel (see FIG. 7), a focus flag may be attached to an item of interest, and only the items with the focus flag may be displayed by the sort function.

The detail display button 215 is an instruction button for displaying the details of the case information, and displays details when any case information is selected from the list of case information displayed in the case information list display field 214. When the button 215 is pressed, the provision information generation unit 119 generates information for displaying the screen 202 shown in FIG.

The screen 202 shown in FIG. 5 is a screen for displaying details of item information. On this screen 202, tabs 211a to 212b, a case information display column 221 for displaying details of case information, a personnel information list display column 222 for displaying a list of personnel information, and an input for inputting search conditions for personnel information. A column 223 and a detail display button 215 for displaying details of human resource information are included.

The case information display column 221 is a column for displaying detailed information of the case information selected when the detail display button 215 shown in FIG. 4 is pressed.

The personnel information list display column 222 is a column for displaying a list of personnel information. The human resource information displayed in the human resource information list display field 222 has a predetermined matching score with the case information whose details are displayed in the case information display field 221 as a result of the matching process in the matching processing unit 118 of the information providing apparatus 11. It is preferably higher than the value of. Here, when displaying the candidates, consider whether the remuneration and age match between the project and the human resources, and whether the nearest station of the project and the nearest station of the human resources (at home) are close. It is preferable not to display candidates that meet the conditions. The reward can also take into consideration the gross margin desired by the person in charge. For example, for a person in charge who wants a gross profit of at least 100,000 yen by mediating between a project and human resources, the remuneration of the project (eg 1 million yen) ＞= the remuneration of human resources (eg 900,000 yen) + It is possible to display only the information that results in a gross profit (eg, 100,000 yen).

In addition, in the present embodiment, if the resend determination unit 115 determines that the email acquired by the email acquisition unit 111 includes the same information as the information already included in the received email, the matching is performed. You may perform the process which reduces a score. As a result, the matching between the case and the human resources does not go well, the priority of the electronic mail that is retransmitted many times can be lowered, and efficient matching can be realized.

The input field 223 is a field for inputting conditions for searching (limiting) the human resource information to be displayed in the human resource information list display field 222, and each item of the human resource information, for example, commercial distribution, remuneration amount, age, number of interviews, nationality. It is possible to search (limit) the personnel information to be displayed by inputting a numerical value or a character string for each of the location, the technical content, and the like.

The detail display button 215 is an instruction button for displaying the details of the personnel information, and displays the details when any one of the personnel information is selected from the list of personnel information displayed in the personnel information list display field 222. When the button 215 is pressed, the provision information generation unit 119 generates information for displaying the screen 203 as shown in FIG.

The screen 203 shown in FIG. 6 is a screen for displaying details of item information and human resource information. This screen 203 includes a case information display column 221, a personnel information display column 224a, and an offer management instruction button 225 for instructing offer management.

The personnel information display column 224a is a column for displaying the details of the personnel information selected when the detail display button 215 shown in FIG. 5 is pressed.

The offer management instruction button 225 indicates that the item information whose details are displayed in the item information display field 221 and the human resource information whose details are displayed in the human resource information display field 224a are subject to offer management. It is a button for. The offer management will be described later.

<Human Resource Information>
The screen 204 shown in FIG. 7 is a screen for displaying a list of personnel information, and is a screen in a state where the tab 211b is selected.

On this screen 204, tabs 211a to 212b, an input field 216 for inputting search conditions for human resources information, item names of each item included in the human resources information, and an instruction for sorting the human resources information by item name. An instruction button, an item display column 217 for displaying, a personnel information list display column 218 for displaying a list of personnel information, and a detail display button 215 for displaying details of personnel information are included.

The input column 216 is a column for inputting conditions for searching (limiting) the human resource information to be displayed, and each item of the human resource information, such as commercial distribution, reward amount, age, number of interviews, nationality, place, technical content, etc. It is possible to search (limit) the human resource information to be displayed by inputting numerical values or character strings.

The item display field 217 is a field in which an item name of each item of the personnel information to be displayed and an instruction button for instructing sorting of the personnel information by the item name are displayed. The instruction button for instructing sorting uses, for example, symbols such as arrows and triangles as symbols.

The human resource information list display field 218 is a field in which a list of human resource information is displayed in the order instructed in the item display field 217 according to the conditions input in the input field 216.

The detail display button 215 is an instruction button for displaying the details of the personnel information, and displays details when any one of the personnel information is selected from the list of personnel information displayed in the personnel information list display field 218. When the button 215 is pressed, the provision information generation unit 119 generates information for displaying the screen 205 shown in FIG.

The screen 205 shown in FIG. 8 is a screen for displaying details of personnel information. On this screen 205, tabs 211a to 212b, a personnel information display field 231 for displaying details of personnel information, a case information list display field 232 for displaying a list of case information, and an input for inputting search conditions for the case information. A column 233 and a detail display button 215 for displaying the details of the item information are included.

The personnel information display column 231 is a column for displaying the details of the personnel information selected when the detail display button 215 shown in FIG. 7 is pressed.

The case information list display field 232 is a field for displaying a list of case information. The case information displayed in the case information list display field 232 has a predetermined matching score with the personnel information whose details are displayed in the personnel information display field 231 as a result of the matching processing in the matching processing unit 118 of the information providing apparatus 11. It is preferably higher than the value of. Here, when displaying candidates, consider whether the remuneration and age match between the project and the human resources, and whether the nearest station of the project and the nearest station of the human resources (at home) are close. Therefore, it is preferable not to display the candidates whose conditions are not satisfied. The reward can also take into consideration the gross margin desired by the person in charge. For example, for a person in charge who wants a gross profit of at least 100,000 yen by mediating between a project and human resources, the remuneration of the project (eg 1 million yen) ＞= the remuneration of human resources (eg 900,000 yen) + It is possible to display only the information that results in a gross profit (eg, 100,000 yen).

The input field 233 is a field for inputting conditions for searching (limiting) the human resource information to be displayed in the case information list display field 232, and each item of the human resource information, for example, commercial distribution, remuneration amount, age, number of interviews, nationality. It is possible to search (limit) the personnel information to be displayed by inputting a numerical value or a character string for each of the location, the technical content, and the like.

The detail display button 215 is an instruction button for displaying the details of the item information, and displays details when any item information is selected from the list of item information displayed in the item information list display field 232. When the button 215 is pressed, the provision information generation unit 119 generates information for displaying the screen 206 as shown in FIG.

The screen 206 shown in FIG. 9 is a screen for displaying details of personnel information and case information. This screen 206 includes a personnel information display column 231, a case information display column 224b, and an offer management instruction button 225 for instructing offer management.

The case information display field 224b is a field for displaying details of the case information selected when the detail display button 215 shown in FIG. 8 is pressed.

The offer management instruction button 225 indicates that the personnel information whose details are displayed in the personnel information display field 231 and the matter information whose details are displayed in the matter information display field 224b are subject to offer management. It is a button for.

<Automatic matching>
As shown in FIG. 10, the screen 207 is displayed by selecting the tab 212a. The automatic matching function is a function of the matching processing unit 118 to propose a combination of a case and a human resource when there is a combination of the case information and the human resource information whose matching score is higher than a predetermined value. In the example of FIG. 10, of the case information whose outline is displayed in the case information display field 241 and the human resource information whose outline is displayed in the human resource information display field 242, the case and the human resource displayed in the same line Are matched. Then, by pressing the offer management button instruction 225, the item and the human resources can be sent to the offer management screen 208. This makes it easy for the person in charge to determine whether or not the combination proposed by the matching processing unit 118 is to be subject to offer management. Here, the proposal may be pushed to the terminal 32. The push notification is realized by any communication means (eg, mail, chat). Further, when it is desired to check the details of the case information displayed in the case information display column 241 or the personnel information displayed in the personnel information display column 242, the detail display button 215 is pressed with the case information or the personnel information selected. By doing so, details of the case information or the personnel information may be displayed.

<Offer management>
As shown in FIG. 11, the offer management screen 208 is displayed by selecting the tab 212b. In this embodiment, for the purpose of matching the personnel information whose details are displayed in the personnel information display column 251 with the item information whose details are displayed in the item information display column 252, offer management such as interviews is performed. Can be the target of.

Further, the offer management screen 208 preferably displays a detail display button 215 and a result input button 215b. By pressing the detail display button 215, it is possible to display further details of the personnel information whose details are displayed in the personnel information display field 251 and the case information whose details are displayed in the case information display field 252. By pressing the result input button 215b, a button for accepting an offer instruction, an input field for accepting the input of the offer result, an input field for accepting the input of the interview result, etc. may be displayed.

In this way, unlike the general mailer, the screen 201 to the offer management screen 208 generated based on the information generated by the provision information generation unit 119 sorts or searches (restricted) by reward, commercial flow, nearest station, etc. ) Is made possible, greatly improving the convenience for the person in charge.

5. Operation of Offer Management Unit 120 Next, offer management by the offer management unit 120 of the information providing apparatus 11 will be described. 12 and 13 are flowcharts showing the flow of operations of the offer management unit 120.

(S201, S202)
When the offer management unit 120 receives the request from the web server 13 (YES in S201), if the request is the addition of the offer management target (YES in S202), the offer management unit 120 adds the offer management target. The offer management target is added by the web server 13 when the offer management instruction button 225 is pressed on the screen 204 (FIG. 6), the screen 206 (FIG. 9), or the screen 207 (FIG. 10). The web server 13 requests the offer management unit 120 based on the received operation.

(S203)
When the offer management unit 120 receives a request to add an object of offer management, the offer management unit 120 adds a set of relevant case information and personnel information to the object of offer management.

(S204)
When the offer management unit 120 adds the target of the offer management, the offer management unit 120 stores the fact in the storage unit 122 as a log.

(S205)
Further, the offer management unit 120 notifies the learning unit 121 as teacher data of information that can identify the case information and the personnel information added as the target of the offer management.

(S206, S207)
Further, when the request from the web server 13 is a display request for the offer management screen 208 (YES in S206), the offer management unit 120 provides the web server 13 with information for displaying the offer management screen 208. To do.

The operation of the offer management unit 120 when there is a display request for the offer management screen 208 from the web server 13 will be described below with reference to FIG. 13 in S207.

(S301)
As shown in FIG. 13, when the offer management unit 120 receives a request to display the offer management screen 208 from the web server 13, the offer management screen 208 includes a list indicating a set of item information and human resource information targeted for offer management. Is provided to the web server 13. This offer management screen 208 includes a button that receives an offer instruction, an input field that receives an input of an offer result, an input that receives an input of an interview result, in addition to a list showing a set of item information and human resource information targeted for offer management. Columns and the like may be included. This input field can be input from the result input screen displayed by pressing the result input button 215b. At this time, it is preferable to record as a log what has been sent to the offer management and what has not been sent to the offer management. For example, as shown in FIG. 10, it is recorded as a log that the screen 207 representing the matching candidates proposed by the automatic matching function is displayed, and then the case or the personnel sent to the offer management and the offer management are sent to the offer management. It is preferable to record a log for the projects or human resources that could not be obtained. As a result, whether or not it is sent to the offer management is used as teacher data for machine learning by the learning unit 121, which leads to improvement in accuracy of matching proposal.

(S302)
When the offer management unit 120 receives the offer instruction via the offer management screen 208 displayed on the terminal 32 or the like by the web server 13, the offer management unit 120 sends an offer email to each contact information included in the designated case information and personnel information. Processing for doing so (YES in S302).

(S303)
In the processing for transmitting the offer mail, for example, a mail transmission form is displayed via the web server 13 or an instruction for activating a mailer such as the terminal 32 is transmitted. As a result, for example, it is preferable that an empty mail is launched and the destination and the subject are automatically entered. Further, it is also possible to configure the body of the electronic mail so that the contents of the mail to be sent to the other party are copied to the clipboard and the person in charge only has to paste the copied contents. In addition, the approval link and the refusal link are described in the e-mail newly sent by the person in charge so that the person who receives the e-mail (information provider etc.) can indicate the intention simply by clicking those links. Good.

(S304)
Further, when the offer result or the interview result is input to the offer management screen 208 displayed by the web server 13 on the terminal 32 or the like, the offer management unit 120 accepts the input (YES in S304).

(S305)
The offer result is input whether the case information side and the personnel information side have accepted or declined the offer, and the interview result is the success or failure thereof. Specifically, when the information provider clicks the link of the sent mail, it is transferred to a predetermined Web page regardless of approval or refusal, and the information providing system 1 is still used for the web page. A link to an LP (landing page) may be provided for users who do not have it. Then, whether the information provider approves or declines is recorded in a log and can be used as teacher data for machine learning by the learning unit 121.

In the case of refusal, a questionnaire "Please tell me the reason for refusal" is displayed on the web page, and "The engineer's technical content was insufficient" "Proposals from other companies had already decided. You may prepare options such as "ta". The important point here is to reflect in learning by separating "whether the content of the proposal was not good" and "whether the content of the proposal was not a problem, but it was not established due to other external factors". Is. For example, if the reason for refusal is an external factor, it can be learned that the content of the proposal was good. After that, when both the person in charge on the case side and the person in charge on the personnel side consent, the status of the matching candidate changes to “interviewing”. On the other hand, when either the person in charge on the case side or the person in charge on the personnel side declines, the status of the matching candidate becomes “not established”.

(S306)
When the offer management unit 120 receives the offer result (eg, “accept”, “decline”, “under consideration”) or the interview result, the offer management unit 120 stores the content in the storage unit 122 as a log. Specifically, as a result of the interview, if the employment is decided on a clear day, the status of the relevant item in the offer management is changed to “established”. Such a result is also recorded in a log and can be used as teacher data for machine learning by the learning unit 121. As described above, by using the part other than the communication by e-mail for machine learning, it becomes possible to further improve the accuracy of the proposal.

(S307)
Further, the offer management unit 120 notifies the learning unit 121 of the received offer result and interview result as teacher data.

(S308)
Then, if there is an end instruction (YES in S307), the process ends. On the other hand, if there is no end instruction (NO in S307), the process returns to S302 again to continue the process. Here, the termination instruction is not limited to an explicit input from the user, and may be, for example, page transition, browser closing, program termination, or the like.

By using the service provided by the information providing system 1 by the processing described above, the following effects can be obtained.
-Reduce the work load without changing existing business processes that use e-mail.
-A plurality of items required for matching can be comprehensively determined, and the matching accuracy of machine learning by the learning unit 121 can be expected to improve.
・Since an e-mail arrives, a matching candidate can be proposed immediately, so a matching opportunity can be provided promptly.

<Other>
The present invention can also be implemented in the following modes.

A system including a mail server, a web server, and an information providing device,
The mail server receives an email of a predetermined destination via a network,
The web server provides the information generated by the information providing device via a network,
The information providing device includes a mail acquisition unit, a mail analysis unit, a mail distribution unit, and a provided information generation unit,
The mail acquisition unit acquires an email of a predetermined destination from the mail server,
The mail analysis unit analyzes the electronic mail acquired by the mail acquisition unit to identify words included in the electronic mail,
The mail distribution unit distributes the electronic mail acquired by the mail acquisition unit based on the word specified by the mail analysis unit for each predetermined condition,
The provision information generation unit generates information providing a list of the electronic mails sorted by the mail distribution unit with at least a part of the words specified by the mail analysis unit as an item, and provides the information to the web server. ,
system.

A program that causes a computer to function as an information providing device that aggregates and provides information included in a plurality of electronic mails,
The information providing device includes a mail acquisition unit, a mail analysis unit, a mail distribution unit, and a provided information generation unit,
The mail acquisition unit acquires an electronic mail of a predetermined destination,
The mail analysis unit analyzes the electronic mail acquired by the mail acquisition unit to identify words included in the electronic mail,
The mail distribution unit distributes the electronic mail acquired by the mail acquisition unit based on the word specified by the mail analysis unit for each predetermined condition,
The provided information generation unit generates information listing the emails sorted by the mail sorting unit, using at least a part of the words identified by the mail analysis unit as items.
program.

Further, it may be provided as a computer-readable non-transitory recording medium that stores the above program.

Furthermore, the present invention can be implemented in the following modes.
・Provide a function to download the skill sheet or work history attached to the e-mail of the original human resource at any place such as the human resource list or the matching candidate screen.
-Provide a function to download the original email on any screen, such as the list screen and matching candidate screen.
Two or three of the information providing device 11, the mail server 12, and the web server 13 may be configured by the same hardware (eg, a single server).

1: Information providing system 2: Network 3: Dispatching company 4: Dispatching company 5: Dispatching company 6: Dispatching company 11: Information providing device 12: Mail server 13: Web server 31: Mail servers 32-34: Terminal 111: Mail acquisition Part 112: E-mail analysis part 113: E-mail division part 114: E-mail distribution part 115: Re-transmission determination part 116: Distribution depth determination part 117: Matching element extraction part 118: Matching processing part 119: Provision information generation part 120: Offer management Part 121: Learning part 122: Storage parts 201 to 207: Screen 208: Offer management screen 211a: Tab 211b: Tab 212a: Tab 212b: Tab 212: Input field 213: Item display field 214: Item information list display field 215: Details Display button 215b: Result input button 216: Input field 217: Item display field 218: Human resource information list display field 221: Project information display field 222: Human resource information list display field 223: Input field 224a: Human resource information display field 224b: Project information Display column 225: Offer management instruction button 231: Human resources information display column 232: Project information list display column 233: Input column 241: Project information display column 242: Human resources information display column 251: Human resources information display column 252: Project information display column

Claims (11)

An information providing device that aggregates and provides information included in a plurality of emails,
Equipped with a mail acquisition unit, a mail analysis unit, a mail distribution unit, and a provided information generation unit,
The mail acquisition unit acquires an electronic mail of a predetermined destination,
The mail analysis unit analyzes the electronic mail acquired by the mail acquisition unit to identify the words included in the electronic mail,
The mail distribution unit, based on the word specified by the mail analysis unit, sorts the electronic mail acquired by the mail acquisition unit for each predetermined condition,
The provided information generation unit generates information listing the emails sorted by the mail sorting unit, using at least a part of the words identified by the mail analysis unit as items.
Information providing device. Further comprises a retransmission determination unit,
Whether the resend determination unit, based on the word specified by the mail analysis unit, whether the email acquired by the mail acquisition unit includes the same information as the information included in the already received email. Judge,
The information providing device according to claim 1. The e-mail distribution unit distributes the e-mails acquired by the e-mail acquisition unit to those containing job information indicating a job requiring human resources and those containing human resource information showing human resources that can be introduced.
The information providing device according to claim 1. The mail distribution unit further distributes each of the e-mail including the case information and the e-mail including the human resource information based on a job type,
The information providing device according to claim 3. Further equipped with a commercial depth determination unit,
The commercial flow depth determination unit determines a commercial flow depth for a case or a human resource based on the word specified by the mail analysis unit,
The information providing device according to claim 3 or 4. Equipped with a mail division unit,
The email division unit, when the email acquired by the email acquisition unit includes a plurality of item information or a plurality of personnel information, the information included in the email for each item information or personnel information To divide,
The information providing device according to any one of claims 3 to 5. A matching element extraction unit is further provided,
The matching element extraction unit extracts a matching element from each of the case information and the personnel information distributed by the mail distribution unit,
The information providing device according to any one of claims 3 to 6. Further equipped with a learning unit,
The learning unit is
Machine learning is performed, and a word to be extracted as the matching element is determined based on the result of the machine learning.
The information providing device according to claim 7. Further provided with a matching processing unit,
The matching processing unit calculates a matching score between the case information and the personnel information based on the matching element,
The information providing device according to claim 8 or 9. A system including a mail server, a web server, and an information providing device,
The mail server receives an email of a predetermined destination via a network,
The web server provides the information generated by the information providing device via a network,
The information providing device includes a mail acquisition unit, a mail analysis unit, a mail distribution unit, and a provided information generation unit,
The mail acquisition unit acquires an email of a predetermined destination from the mail server,
The mail analysis unit analyzes the electronic mail acquired by the mail acquisition unit to identify the words included in the electronic mail,
The mail distribution unit, based on the word specified by the mail analysis unit, sorts the electronic mail acquired by the mail acquisition unit for each predetermined condition,
The provision information generation unit generates information providing a list of the electronic mails sorted by the mail distribution unit with at least a part of the words specified by the mail analysis unit as an item, and provides the information to the web server. ,
system. A program that causes a computer to function as an information providing device that aggregates and provides information included in a plurality of electronic mails,
The information providing device includes a mail acquisition unit, a mail analysis unit, a mail distribution unit, and a provided information generation unit,
The mail acquisition unit acquires an electronic mail of a predetermined destination,
The mail analysis unit analyzes the electronic mail acquired by the mail acquisition unit to identify the words included in the electronic mail,
The mail distribution unit, based on the word specified by the mail analysis unit, sorts the electronic mail acquired by the mail acquisition unit for each predetermined condition,
The provided information generation unit generates information listing the emails sorted by the mail sorting unit, using at least a part of the words identified by the mail analysis unit as items.
program.