KR20200138069A - Mail analysis server, mail analysis method and program - Google Patents

Abstract

The present invention relates to a mail interpretation server, a mail interpretation method and a program which can improve interpretation precision of a received mail, and reduce a processing burden of a user for the received mail. The mail interpretation server comprises: a determination unit determining presence or absence of a reply to an accumulated received email; a learning unit interpreting at least one side of characters included in the received mail determined to be replied and characters included in the received email determined not to be replied and constructing a machine learning model by machine learning using a neural network; and an estimation unit estimating necessity of the reply to the received mail using the machine learning model.

Description

Mail analysis server, mail analysis method and program

One embodiment of the present invention relates to a mail analysis server, a mail analysis method, and a program.

In recent years, e-mail is one of the important tools frequently used in business and academic circles. The user receives a large amount of e-mails every day, and determines one by one of the received e-mails that an immediate reply is required and which does not, and performs processing. Users spend a lot of time dealing with large amounts of e-mail.

In Patent Document 1, the e-mail already received by the user is used as teacher data, and for the e-mail newly received by the user, the degree of action required for each user (action priority Fig.) is described for an e-mail system that can evaluate and present the results.

In Patent Document 1, machine learning is performed using both an e-mail in which an operation such as reply or transmission has been made and an e-mail in which nothing has been operated as teacher data. In addition, by assigning weights to keywords extracted from the e-mail in which the action has been performed, the action required for the newly received e-mail is evaluated as a score.

Patent Document 1: International Publication No. 2016/117132

In Patent Document 1, machine learning is performed using received mail stored in a receiving folder of a user's mail application as teacher data. Therefore, even if the accounts are the same, if the terminal devices receiving the received mail are different, the analysis result may be different for each terminal device.

In addition, in Patent Document 1, keywords included in both the received mails that have been returned and received mails that have not been returned are extracted and evaluated by score, but the relevance of the keywords is not considered. For this reason, when the same keyword is included in both the returned received mail and the non-replied received mail, there is a fear that the analysis precision of the received mail may be lowered.

Accordingly, an object of an embodiment of the present invention is to improve the analysis precision of received mails and to reduce the user's processing burden on received mails.

The mail analysis server according to an embodiment of the present invention includes a determination unit for determining the presence or absence of a reply to the accumulated received mail, and a character included in the received mail determined to have been returned and included in the received mail determined not to be answered. A learning unit that analyzes at least one of the characters and constructs a machine learning model by machine learning using a neural network, and the need to reply to the received mail using a machine learning model for the characters included in the received mail. It has an estimating unit that estimates.

In the above configuration, the determination unit compares the accumulated received mail and the accumulated transmitted mail, determines whether the body of the transmission mail contains a part of the body of the received mail, and If so, it is determined that the received mail has been replied.

In the above configuration, the determination unit compares the accumulated received mail and the accumulated transmitted mail, determines whether or not the transmitted mail includes incidental information of the received mail, and the incidental information of the transmitted mail includes the incidental information of the received mail. If so, it is determined that the received mail has been replied.

In the above configuration, the determination unit further includes a time calculation unit that determines the presence or absence of a reply to the accumulated received mail, and then calculates a time required to reply to the received mail determined to be reply.

In the above configuration, the learning unit analyzes the text included in the received mail and the time required to reply, and constructs a machine learning model by machine learning using a neural network.

In the above configuration, the learning model is generated based on the received mail determined to be returned without distinguishing a plurality of accounts.

In the above configuration, the machine learning model is updated based on the received mail determined to be returned for each account.

In the above configuration, the neural network is a character-level convolutional neural network.

In the mail analysis method according to an embodiment of the present invention, it is determined whether there is a reply to the accumulated received mail, and the text included in the received mail determined to be replyed and the character included in the received mail determined not to be returned. Mail analysis that analyzes at least one side, constructs a machine learning model by machine learning using a neural network, and estimates the need for a reply to the received mail using a machine learning model for characters included in the newly received mail. That's the way.

In the above method, the presence or absence of a reply to the accumulated received mail is determined whether or not a part of the body of the received mail is included in the transmitted mail compared to the accumulated received mail and the accumulated transmitted mail, and the transmitted mail is received. When a part of the body of the mail is included, it is determined that the received mail is returned.

In the above method, the presence or absence of a reply to the accumulated received mail compares the accumulated received mail and the accumulated transmitted mail, and determines whether or not additional information of the received mail is included in the transmitted mail, and the transmitted mail is If the incidental information of is included, it is determined that the corresponding received e-mail has been returned.

In the above method, after determining the presence or absence of a reply to the accumulated received e-mail, the time required to reply to the received e-mail determined to have been answered is calculated.

In the above method, the text included in the received mail and the time required to reply are analyzed, and a machine learning model is constructed by machine learning using a neural network.

In the above method, the neural network is a character level convolutional neural network.

In the above method, the machine learning model is generated based on the received mail determined to be returned without distinguishing a plurality of accounts.

In the above method, the machine learning model is updated based on the received mail determined to be returned for each account.

A program according to an embodiment of the present invention is a program for causing a computer to execute the mail analysis method described above.

According to an embodiment of the present invention, it is possible to reduce the processing burden on the user for the received mail by improving the analysis precision of the received mail.

1 is a diagram showing an outline of a mail analysis server according to an embodiment of the present invention.
Fig. 2 is a diagram showing a display screen when a received e-mail is received together with a probability of a main part of a reply in a mail application.
FIG. 3 is a diagram illustrating a display screen when a received mail is selected in the mail application shown in FIG. 2.
4 is a block diagram illustrating the hardware configuration of the mail analysis server according to an embodiment of the present invention.
5 is a functional block diagram of a mail analysis server according to an embodiment of the present invention.
6 is a diagram for explaining a process of generating a machine learning model in the mail analysis server.
Fig. 7 is a diagram for explaining a process in which the mail analysis server analyzes a newly received mail and estimates a probability of a main part of a reply.
8 is a diagram showing an outline of a mail analysis server according to an embodiment of the present invention.
9 is a functional block diagram of a mail analysis server according to an embodiment of the present invention.
10 is a diagram showing an outline of a mail analysis server according to an embodiment of the present invention.
11 is a functional block diagram of a mail analysis server according to an embodiment of the present invention.
Fig. 12 is a diagram showing a display screen when a received mail is received together with a probability of a main part of a reply in a mail application.
Fig. 13 is a diagram showing a display screen when a received mail is selected in the mail application.
14 is a functional block diagram of a mail analysis server according to an embodiment of the present invention.
15 is a functional block diagram of a mail analysis server according to an embodiment of the present invention.

(First embodiment)

In the present embodiment, a process in which the mail analysis server 10 according to an embodiment of the present invention outputs a main part of a reply to a newly received received mail will be described with reference to FIGS. 1 to 7.

In this specification and the like, an electronic mail is one having information including one or more characters in the body of the mail. The electronic mail includes not only the body of the mail, but also attached files such as side information, document files, table calculation files, and presentation files. Here, the incidental information refers to information incidental to the document data of the mail. For example, information on the recipient name, subject name and document data, operation information such as transmission and reply, operation execution date, creation date and document data Information such as the author of may be used.

In addition, in this specification and the like, the transmission mail refers to an email created by a user using the mail analysis server 10 and transmitted to an account (mail address) of an arbitrary user. In addition, the received mail refers to a mail created by an arbitrary sender and received from an account (mail address) of a user using the mail analysis server 10. In addition, in the case of simply describing the e-mail, the received e-mail, the transmitted e-mail, the forwarded e-mail and the deleted e-mail are not distinguished.

[Overview of Mail Analysis Server]

1 is a diagram illustrating a mail analysis server 10 according to an embodiment of the present invention. In FIG. 1, the mail analysis server 10, the mail servers 20_1 and 20_2, and the terminal devices 30_1 and 30_2 are shown. The mail servers 20_1 and 20_2 transmit and receive electronic mail through the network 40.

In addition, the terminal device 30_1 is a terminal device of a user who uses the mail analysis server 10, and the terminal device 30_2 is a terminal device used by an arbitrary user. In addition, the mail server 20_1 is a server that manages an account of a user who uses the mail analysis server 10, and the mail server 20_2 is a server that manages an account of an arbitrary user. In the following description, when the mail servers 20_1 and 20_2 are not distinguished from each other, the mail server 20 will be described. In addition, when the terminal devices 30_1 and 30_2 are not distinguished from each other, the terminal devices 30 are described.

The mail server 20 has a function of transmitting and receiving an e-mail of the registered account. Further, the mail server 20 has a function of maintaining mail data of an electronic mail for each account. Here, the e-mail data includes an address book, body and incidental information of the e-mail, and attached files. Here, the mail data of the electronic mail for each account is held in the recording medium of the mail server 20. The mail server 20_1 transmits, to the mail analysis server 10, a received mail received from the outside from an account of a user to a user's account. Further, the mail server 20_1 transmits, to the mail server 20_2, a transmission mail transmitted from the user's account to a user's account.

The mail analysis server 10 analyzes the characters of the transmitted/received mail received from the mail server 20_1, and estimates the necessity of a reply to the received mail. Here, the e-mail text may include at least the body of the e-mail, and may include additional information and a citation portion of other received e-mails. Additional information includes the IP address of the sender mail server (Received field), sender information (From field), mail internal ID (Message-ID field), mail title (Subject field), and version information of the file format of the e-mail. (MIME-Version field), e-mail file format information (Content-Type field), destination information (To field), e-mail copy destination information (Cc field), e-mail creation date and time information (Date field), and reply The destination mail information (References field, In-Reply-To field) is included. The mail analysis server 10 estimates a probability of 0% to 100%, for example, taking 100% of the need for a reply. The mail analysis server 10 has a function of transmitting the received mail to the terminal device 30 together with a probability regarding the necessity of a reply.

The terminal device 30 connects to the mail analysis server 10 and executes processing related to electronic mail. E-mail processing is, for example, a function of inputting mail data for transmission transmitted from a user's account, a function of displaying mail data to be viewed, and the like. For example, when the terminal device 30 performs an electronic mail process, it can be executed by using a mail application.

As shown in FIG. 1, the terminal device 30_2 transmits an e-mail to the sender's mail server 20_2, and the sender's mail server 20 sends an electronic mail to the sender's mail server 20_1 via the network 40. Send mail. The mail server 20_1 of the transmission destination transmits the received electronic mail to the mail analysis server 10. The mail analysis server 10 estimates the probability of the need for a reply to the account of the sender as 0% to 100% by analyzing the received electronic mail. The mail analysis server 10 can transmit an e-mail to the terminal device 30 together with a probability of a need for a reply, and the terminal device 30 can receive an e-mail with a probability of a need for a reply.

FIG. 2 is a diagram showing a display screen when a received mail is received together with a probability regarding the essential part of a reply in the mail application 400 of the terminal device 30. The terminal device 30 displays an account 401, an operation display area 402, and a received mail display area 404 by the mail application 400. In the account 401, an account of a user who uses the mail application 400 is displayed. In the operation display area 402, operations such as new mail, reply, reply to the power supply, and transmission are displayed. The user selects either action and composes an e-mail. In the received mail display area 404, in each of the received mails 405_1 to 405_5, the probabilities 406_1 to 406_5 related to the need for a reply are displayed.

A predetermined limit value may be set in advance by the user as to the probability regarding the necessity of the reply. The terminal device 30 may change the display of the received e-mail when the probability regarding the necessity of the reply is equal to or greater than a predetermined threshold. The terminal device 30 may highlight, for example, a probability regarding the essential part of a reply described in the received mail. The terminal device 30, for example, as shown in Fig. 2, the essential part of a reply when the probability related to the essential part of the reply is 80% or more, 60% to less than 80%, and 0% to 60% or less. The highlighting of the probabilities for may be different.

FIG. 3 is a diagram showing a display screen when a received mail 405_1 is selected in the mail application 400 shown in FIG. 2. As shown in Fig. 3, in the received mail 405_1, incidental information of the received mail 405_1 and the mail body are displayed along with the probability of the essential part of the reply. In addition, an icon such as a reply may be displayed along with a probability regarding the necessity of the reply, so that the user may be prompted to reply.

In the terminal device 30, the user can easily determine whether the corresponding received mail has a high or low need for a reply by checking the probability of the need for a reply. Since the user needs to process the received e-mail with a high probability of the necessity of a reply, the user's processing burden for the received e-mail can be reduced.

[Hardware configuration of mail interpretation server]

4 is a block diagram illustrating a hardware configuration of the mail analysis server 10 according to an embodiment of the present invention. As shown in Fig. 4, the mail analysis server 10 includes a control unit 101, a ROM 102 (Read Only Memory), a RAM 103 (Random Access Memory), a recording medium 104, and a communication unit 105. Include.

The control unit 101 is, for example, a CPU (Central Processing Unit) or the like, and controls the operation of a functional block to be described later. The control unit 101 reads, for example, an operation program recorded in the ROM 102 and the recording medium 104, expands it in the RAM 103, and executes the operation program, so that the mail analysis server ( 10) to realize the processing function.

The recording medium 104 is a recording device capable of permanent information retention and information change such as a hard disk drive (HDD), for example. In the recording medium 104, mail data of an electronic mail transmitted and received for each account acquired from the mail server 20 is stored. Further, in the recording medium 104, teacher data based on the reply-completed received mail to be described in detail later is stored. Further, in the recording medium 104, a machine learning model generated by machine learning described in detail later is stored. Further, e-mail mail data, teacher data, and machine learning models for each account may be stored in different recording media.

The ROM 102 is a recording device capable of permanently holding information such as a nonvolatile memory. In the ROM 102, for example, a program for executing mail analysis processing is stored.

The RAM 103 is a changeable memory device used for temporary information storage such as a volatile memory, for example. The RAM 103 is used as a storage area for storing a development area such as a program of a functional block recorded in the ROM 102 and an intermediate data output from the operation of the functional block.

The communication unit 105 is a communication interface. The mail analysis server 10 connects to the terminal device 30 and the mail server 20 through the communication unit 105, and transmits and receives mail between the terminal device 30 and the mail server 20. can do.

In addition, in the present embodiment, in order to simplify the description, the hardware configuration of the mail analysis server 10 has been described as having various blocks, but the embodiment of the present invention is not limited thereto. For example, at least one of the various blocks may be configured as a separate device from the mail analysis server 10.

[Functional block diagram of mail analysis server]

5 is a functional block diagram of the mail analysis server 10 according to an embodiment of the present invention. The mail analysis server 10 includes an e-mail acquisition unit 111, an e-mail storage unit 112, a determination unit 113, teacher data 114, a learning unit 115, a machine learning model 116, and an estimation unit. (117) and an output unit (118). The e-mail acquisition unit 111, the determination unit 113, the learning unit 115, the estimation unit 117, and the output unit 118 are processed by the control unit 101. Further, the electronic mail storage unit 112, the machine learning model 116, and the teacher data 114 are stored in the recording medium 104.

First, functional blocks used by the mail analysis server 10 when performing machine learning will be described.

The electronic mail acquisition unit 111 acquires the transmitted and received electronic mail accumulated in the mail server 20_1 every predetermined period. The acquired e-mail is used when constructing the machine learning model 116 by the learning unit 115 described later. In addition, the e-mail acquisition unit 111 acquires the received mail newly received from the mail server 20_1. The acquired newly received received mail is used when estimating the probability of the presence or absence of a reply by the estimating unit 117 described later.

The electronic mail storage unit 112 stores the transmitted and received electronic mail acquired by the electronic mail acquisition unit 111. The acquired transmitted/received electronic mail is stored in the electronic mail storage unit 112 for each account.

The determination unit 113 reads out the received mail and the transmission mail accumulated in the electronic mail storage unit 112 every predetermined period, and determines the presence or absence of a reply for each received mail. The determination unit 113 compares the accumulated received mail and the accumulated transmitted mail, and determines whether or not a part of the body of the received mail is included in the body of the transmitted mail. When the sending mail contains a part of the body of the received mail, it is determined that the received mail has been returned. Further, the determination unit 113 may compare the accumulated received mail and the accumulated transmitted mail, and determine that the received mail has been returned when the incidental information of the transmission mail includes the incidental information of the received mail. Here, the incidental information of the received e-mail is Message-ID, In-Reply-To, and References, for example. In In-Reply-To, only one Message-ID of the reply target is recorded. In References, multiple Message-IDs of the thread's mail are recorded. For example, even if one of the outgoing mails has been deleted, the status of the reply can be restored from References of subsequent mails. The received mail determined to have been answered and the received mail determined not to be answered are stored in the teacher data 114.

The learning unit 115 analyzes characters included in the received mail with respect to the received mail accumulated in the teacher data 114, and constructs a machine learning model by machine learning using a neural network. The built machine learning model is stored in the machine learning model 116. Here, the neural network uses a character-level convolutional neural network. The character-level convolutional network is machine-learned in character units, not word units. Therefore, it is characterized by being strong in unknown words such as proper nouns, industry terms, and industry-specific abbreviations, errors, etc.

In addition, the neural network updates the machine learning model 116 by re-learning every predetermined period. The neural network updates the machine learning model 116 by relearning the received mail accumulated in the teacher data 114 as new teacher data.

In addition, the construction of the machine learning model 116 may be performed without distinguishing a plurality of accounts, or may be performed for each account. For example, the machine learning model 116 is constructed for a received mail determined to be returned without distinguishing a plurality of accounts. Further, the machine learning model 116 is updated on the basis of the received mail determined to be returned for each account. By updating the machine learning model 116 for each account, it can be desired to improve the accuracy of the analysis result of the received mail for each account.

When the e-mail acquisition unit 111 acquires the newly received received mail, the estimating unit 117 analyzes the character included in the newly received received mail using the machine learning model 116, and Estimate the probability of the need for a reply.

The output unit 118 outputs the received mail to the terminal device 30 together with the probability of the essential part of the reply estimated by the estimation unit 117.

In addition, the received mail newly received by the electronic mail acquisition unit 111 may be stored in the electronic mail storage unit 112 and may be transmitted to the estimation unit 117. Alternatively, the received mail may not be stored in the electronic mail storage unit 112 but may be transmitted to the estimation unit 117. Even if the e-mail acquisition unit 111 is not stored in the e-mail storage unit 112 at the timing when the newly received received mail is acquired, the e-mail storage unit 112 transmits and receives the e-mail transmitted and received every predetermined period to the mail server 20_1. You can get it from

[Data processing method]

Next, a mail analysis method according to an embodiment of the present invention will be described with reference to FIGS. 6 and 7. The control unit 101 causes the mail analysis server 10 to execute each process.

First, a process of constructing a machine learning model in the mail analysis server 10 will be described with reference to FIG. 6. The e-mail acquisition unit 111 acquires the e-mail transmitted and received from the mail server 20_2 every predetermined period (step S301). The e-mail acquisition unit 111 stores the acquired e-mail in the e-mail storage unit 112 (step S302). Next, the determination unit 113 reads out the transmission mail and the reception mail from the electronic mail storage unit 112, and determines the presence or absence of a reply to the received electronic mail (step S303). Next, the determination unit 113 stores the received mail determined to have been answered and the received mail determined not to be answered in the teacher data (step S304). Next, the learning unit 115 analyzes the received mail determined to be returned as teacher data, and constructs a machine learning model by machine learning using a neural network (step S305). Next, the learning unit 115 stores the constructed machine learning model in the machine learning model 116 (step S306).

Next, with reference to FIG. 7, a description will be given of a process in which the mail analysis server 10 analyzes the newly received received mail and estimates the probability of the essential part of the reply. The electronic mail acquisition unit 111 acquires the newly received received mail from the mail server 20_2 (step S311). Next, the estimation unit 117 analyzes the characters included in the received mail by the neural network (step S312). Next, the estimating unit 117 estimates the probability of the necessity of a reply to the received mail using the machine learning model (step S313). Next, the output unit 118 outputs the received mail to the terminal device 30 together with the probability of the essential part of the reply for returning the received mail (step S314).

In the case of transmitting a transmission mail from the terminal device 30_1 to the terminal device 30_2, it is not necessary to pass through the mail analysis server 10. When transmitting the transmission mail from the terminal device 30_1 to the terminal device 30_2, the transmission mail is transmitted from the terminal device 30_1 to the mail server 20_1. The mail server 20_1 transmits to the terminal device 30_2 through the mail server 20_2. When the mail server 20_1 acquires the transmission mail from the terminal device 30_1, the mail server 20_1 stores the transmission mail in the recording medium of the mail server 20_1. When the mail analysis server 10 acquires the transmission mail, it may be acquired from the mail server 20_1 every predetermined period, as described in step S301 of FIG. 6.

The mail analysis server 10 according to an embodiment of the present invention uses a machine learning model to estimate the need for a reply by estimating the probability of the need for a reply to the received mail from a character included in the newly received mail. A received mail can be transmitted to the terminal device 30 together with the probability of. The user can easily determine whether the corresponding received mail has a high or low need for a reply by checking the probability of the need for a reply in the terminal device 30. Accordingly, it is possible to reduce the user's processing burden for the received mail.

Conventionally, since the number of occurrences of keywords included in the received e-mail is only counted, the relationship or relevance before and after the word cannot be considered. For example, in the past, "Please renew the SSL certificate. A reply is unnecessary.” and “It is not necessary to renew the SSL certificate. Please reply.” Since the mail server 20 according to an embodiment of the present invention uses a character-level convolutional neural network, it is possible to analyze the e-mail in consideration of the relations and relevance before and after words.

In addition, since it was conventionally interpreted at the word level, if sentences containing words that are not in the dictionary, such as proper nouns, industry terms, and industry-specific abbreviations, could not be accurately divided into columns of words, the accuracy of the interpretation may decrease. There was. Since the mail server 20 according to the embodiment of the present invention interprets it at the character level, there is no fear of deteriorating the precision even if words not in the dictionary are included.

In addition, electronic mail judged as spam mail has high commonality with respect to characters and words used regardless of account. Therefore, regardless of the account, it is meaningful to morphemely analyze the e-mail determined to be spam.

On the other hand, for e-mails requiring a reply, the tendency to need a reply varies greatly depending on the user. In addition, in e-mails requiring a reply, there are many unknown words such as proper nouns, industry terms, and industry-specific abbreviations. Therefore, by analyzing the text of the e-mail at the character level rather than morpheme analysis, the accuracy of the probability regarding the essential part of the reply can be further improved.

Further, in the mail analysis server 10 according to the embodiment of the present invention, characters included in the received electronic mail are analyzed. Therefore, if the accounts are the same, the probability of the essential part of the reply output to the plurality of terminal devices 30 becomes the same in each of the plurality of terminal devices 30. Accordingly, the user can receive the e-mail with the probability of the need for a reply without depending on the received terminal device 30.

(2nd embodiment)

In the first embodiment, an example in which the mail analysis server 10 and the mail server 20 are configured as separate devices has been described, but in this embodiment, the mail analysis server 10 and the mail server 20 are integrated and the same An example of a device configuration will be described with reference to FIGS. 8 and 9.

[Overview of Mail Analysis Server]

8 is a diagram for explaining a mail analysis server 10A according to an embodiment of the present invention. 8 shows the mail analysis server 10A, the mail server 20_2, and the terminal devices 30_1 and 30_2. In this embodiment, the mail analysis server 10 has the function of sending/receiving electronic mail from the mail server 20_1 described in the first embodiment. In addition, in the present embodiment, the mail analysis server 10 has a function of transmitting and receiving an electronic mail of a registered account. In addition, the mail analysis server 10 has a function of maintaining an electronic mail for each account.

[Functional block diagram of mail analysis server]

9 is a functional block diagram of the mail analysis server 10A according to an embodiment of the present invention. The mail analysis server 10A shown in FIG. 9 is different from the mail analysis server 10 shown in FIG. 5 in that it further has a function of transmitting the transmission mail transmitted from the terminal device 30_1 to the mail server 20_2. In this embodiment, functional blocks having functions different from those of the first embodiment will be described in detail.

First, processing in the case where the mail analysis server 10A newly receives a received mail from the mail server 20_2 will be described. When the e-mail acquisition unit 111A acquires the received e-mail received from the mail server 20_2, it stores it in the e-mail storage unit 112. Further, the e-mail acquisition unit 111A transmits a received e-mail to the estimation unit 117. Further, the determination unit 113, the teacher data 114, the learning unit 115, the machine learning model 116, and the estimating unit 117 are the same as those described in the first embodiment, so detailed descriptions are omitted. do. Further, the output unit 118 outputs the received mail to the terminal device 30 together with the probability of the essential part of the reply estimated by the estimation unit 117.

Next, processing in the case where the mail analysis server 10A transmits the transmission mail transmitted from the terminal device 30 to the mail server 20_2 will be described. When the e-mail acquisition unit 111A acquires the transmission e-mail transmitted from the terminal device 30, the e-mail storage unit 112 stores it. Further, the e-mail acquisition unit 111A transmits a transmission e-mail to the output unit 118A. The output unit 118A transmits a transmission mail to the mail server 20_2.

(3rd embodiment)

In the present embodiment, on the screen displayed on the browser of the terminal device 30, the user inputs or performs an operation to the mail analysis server 10B that enables the exchange of e-mails in the terminal device 30. This will be described with reference to FIGS. 10 and 11.

[Overview of Mail Analysis Server]

10 is a diagram for explaining a mail analysis server 10B according to an embodiment of the present invention. In FIG. 10, the mail analysis server 10B, the mail server 20_2, and the terminal device 30 are shown.

[Functional block diagram of mail analysis server]

11 is a functional block diagram of the mail analysis server 10B according to an embodiment of the present invention. The mail analysis server 10B shown in FIG. 11 does not output an e-mail to the terminal device 30, but displays the e-mail on a screen displayed on the browser of the terminal device 30. It is different from the analysis server 10. In this embodiment, functional blocks having functions different from those of the first embodiment will be described in detail.

When the electronic mail processing unit 119 acquires the received mail received from the mail server 20_2, it stores it in the electronic mail storage unit 112. Further, the electronic mail processing unit 119 transmits a received mail to the estimation unit 117. The estimating unit 117 estimates the probability of the presence or absence of a reply, and stores the probability of the presence or absence of a reply together with the received mail stored in the e-mail storage unit 112. The electronic mail processing unit 119 displays the received mail on a screen displayed on the browser of the terminal device 30 together with the probability of the presence or absence of a reply in response to the request to view the received mail from the terminal device 30.

When the electronic mail processing unit 119 acquires the generated transmission mail on the screen displayed on the browser of the terminal device 30, the electronic mail storage unit 112 stores it. Further, the electronic mail processing unit 119 transmits the transmission mail to the mail server 20_2.

(Modified Example 1)

In the first embodiment, an example of displaying all received mails in the reception tray of the mail application 400 has been described, but the embodiment of the present invention is not limited thereto. Apart from the receiving tray, only the received e-mails having a probability of a request for a reply being equal to or greater than a predetermined threshold may be displayed on the required reply tray. Further, received mails that are less than a predetermined threshold may be displayed on the receiving tray.

FIG. 12 is a diagram showing a display screen when a received mail is received together with a probability regarding the essential part of a reply in the mail application 400 of the terminal device 30. The difference from the display of the mail application 400 shown in Fig. 2 is that, in addition to the reception tray, a request reply tray is provided. In the reply-to-request tray, only received e-mails having a probability of a responsibility of a reply being equal to or greater than a predetermined threshold can be displayed. In Fig. 12, the received mails 405_1 to 405_4 having a probability of 80% or more regarding the essential part of the reply are displayed. In addition, since a received e-mail with a probability of a request for a reply of less than 80% is displayed on the receiving tray, the user can properly confirm it. Accordingly, since the user should first check the received mail with a high priority of the reply received from the request reply tray and reply, the user's processing efficiency for the received mail is improved.

(Modified Example 2)

The probability of the need for a reply to an incoming e-mail may vary from account to account. This is because the tendency for a reply to be required even for the same incoming e-mail differs greatly from user to user.

Fig. 13 is a diagram showing a display screen when the received mail 407_1 is selected in the mail application. As shown in Fig. 3, when "Suzuki of YYYY Co., Ltd." receives a received e-mail from "Yamada of ZZZZ Co., Ltd.", the probability of the essential part of the reply is output as 80%. On the other hand, as shown in Fig. 13, when'Tanaka of YYYY Co., Ltd.' receives a received e-mail from'Yamada of ZZZZ Co., Ltd.', the probability of the essential part of the reply may be output as 50%.

The mail analysis server 10 according to an embodiment of the present invention updates the machine learning model for each account. Therefore, when a plurality of recipients are included for one received mail, the mail analysis server 10 can estimate the probability of the need for a reply for each account.

(Modified Example 3)

In one embodiment of the present invention, an example has been described in which characters included in a received mail are targeted for machine learning, but the embodiment of the present invention is not limited thereto. The time taken to reply to the received e-mail may be used as a target for machine learning.

14 is a functional block diagram of the mail analysis server 10C according to an embodiment of the present invention. The mail analysis server 10C shown in FIG. 14 is a time calculation unit that determines the presence or absence of a reply to the received e-mail in the determination unit 113, and then calculates the time taken to reply to the received e-mail. It differs from the mail analysis server 10 shown in FIG. 5 in that it further has 121. Hereinafter, the function of the time calculation unit 121 will be described in detail.

The time calculation unit 121 calculates the time required until a reply is made to the received mail. The object of the e-mail for calculating the time required until the reply may be the reply e-mail returned to the received e-mail, or all the transmitted/received e-mails. In the case of a reply mail returned to a received mail, the time required until a reply is extracted from the reception date and time of the received mail and the transmission date and time of the transmission mail returned to the received mail. Alternatively, the time required until the reply may be extracted from the reception date and time of the received mail included in the incidental information of the reply mail and the transmission date and time of the reply mail. In addition, in the case of targeting a reply e-mail that has been replied to a received e-mail, extraction of the time required until a reply may be performed. For a received e-mail that has not been answered, the time required to reply may be ∞. By the time calculation unit 121, the extracted time until the reply is stored in the teacher data 114.

For incoming e-mails, a limit value may be provided for the time required to reply. For example, if the time required to reply is less than 24 hours, it may be determined that it is a received mail of high importance to be returned, and if it is 24 hours or longer, it may be determined that it is a received mail of low importance. The text included in the sent/received e-mail and the time required for the extracted reply are stored in the teacher data.

The learning unit 115 performs machine learning by including not only the text of the transmission/reception mail but also the time required for a reply.

As an object of machine learning, by including not only the characters of the received mail but also the time required until the reply, the analysis accuracy regarding the essential part of the reply of the received mail can be further improved.

(Modified Example 4)

In one embodiment of the present invention, an example in which the mail analysis servers 10, 10A to 10C includes at least the determination unit 113, the learning unit 115, and the estimation unit 117 has been described, but is not limited thereto. The mail analysis server 10D may have at least the estimation unit 117, and may have a separate learning device for constructing a machine learning model.

15 is a functional block diagram of the mail analysis server 10D and the learning device 50 according to an embodiment of the present invention. As shown in Fig. 15, a learning device 50 is provided separately from the mail analysis server 10D.

The learning device 50 has an electronic mail acquisition unit 111, an electronic mail storage unit 112, a determination unit 113, teacher data 114, a learning unit 115, and a machine learning model 116. The determination unit 113 determines the presence or absence of a reply to the accumulated received e-mail, and the learning unit 115 determines at least one of a character included in the received e-mail determined to have been answered and a character included in the received e-mail determined not to be returned. And construct a machine learning model by machine learning using neural networks. The detailed description of each configuration of the learning device 50 is as described in the first to third embodiments.

The mail analysis server 10D has an electronic mail acquisition unit 122, an estimation unit 117, and an output unit 118. The e-mail acquisition unit 122 acquires a received e-mail from the mail server 20_1 for which it is necessary to estimate the necessity of a reply. The estimation unit 117 estimates the necessity of a reply with respect to the character of the received mail acquired by the electronic mail acquisition unit 122 using the machine learning model of the learning device 50. The output unit 118 outputs the main part of the reply estimated by the estimation unit 117 to the terminal device 30.

As described above, the mail analysis server 10D and the learning device 50 may be different devices.

(Modified Example 5)

In one embodiment of the present invention, an example in which the learning unit 115 constructs a machine learning model using a character-level convolutional neural network has been described, but is not limited thereto. In the learning unit 115, as a neural network, a machine learning model may be constructed using BERT (Bidirectional Encoder Representations from Transformers). BERT is a method for pre-training language expressions to obtain cutting-edge results of various Natural Language Processing (NLP) tasks. BERT provides interactive pre-training without a teacher. No teacher means that the BERT has been trained using only the plain text corpus.

The learning unit 115 learns the received mail accumulated in the teacher data 114 as teacher data by using the model of completion of pre-learning of BERT, and constructs a machine learning model. When the estimating unit 117 acquires the newly received received mail by the e-mail acquisition unit 111, it analyzes the character included in the newly received received mail using the machine learning model 116 by BERT, and Estimate the probability of the need for a reply to an incoming e-mail. In the BERT, since the context of both the front and the back of the word of interest can be considered, the accuracy of the probability regarding the need for a reply to an incoming mail can be improved.

10, 10A, 10B: mail interpretation server
20: mail server
30: terminal device
40: network
101: control unit
102: ROM
103: RAM
104: recording medium
105: communication department
111, 111A: E-mail acquisition unit
112: e-mail storage
113: judgment unit
114: Teacher data
115: Learning Department
116: machine learning model
117: estimation unit
118, 118A: output
119: e-mail processing unit
400: mail application
401: account
402: motion display area
404: Incoming mail display area
405: incoming mail
406: Probability of the need for reply
407: incoming mail

Claims (13)

A determination unit that determines the presence or absence of a reply to the accumulated received e-mail;
A learning unit that analyzes at least one of the characters included in the received mail determined to be replied and the characters included in the received mail determined to have not been answered, and constructs a machine learning model by machine learning using a neural network;
With respect to the text included in the received received mail, using the machine learning model to estimate the necessity of a reply to the received mail,
The determination unit,
Comparing the accumulated received mail and accumulated transmitted mail, it is determined whether or not a part of the body of the received mail is included in the body of the sending mail,
If the sending mail contains part of the body of the received mail, it is determined that the received mail has been replied.
Mail interpretation server.
A determination unit that determines the presence or absence of a reply to the accumulated received e-mail;
A learning unit that analyzes at least one of the characters included in the received mail determined to be replied and the characters included in the received mail determined to have not been answered, and constructs a machine learning model by machine learning using a neural network;
With respect to the text included in the received received mail, using the machine learning model to estimate the necessity of a reply to the received mail,
The determination unit,
Comparing the accumulated received mail and accumulated transmitted mail, it is determined whether or not additional information of the received mail is included in the transmitted mail,
When the incidental information of the outgoing mail includes the incidental information of the received e-mail, it is determined that the received e-mail has been returned.
Mail interpretation server.
A determination unit that determines the presence or absence of a reply to the accumulated received e-mail;
A learning unit that analyzes at least one of the characters included in the received mail determined to be replied and the characters included in the received mail determined to have not been answered, and constructs a machine learning model by machine learning using a neural network;
An estimating unit for estimating the necessity of a reply to the received mail using the machine learning model with respect to the text included in the received mail;
The determination unit has a time calculation unit that determines the presence or absence of a reply to the accumulated received mail, and then calculates a time required to reply to the accumulated received mail,
The learning unit analyzes the text included in the received mail and the time required until the reply, and constructs the machine learning model by machine learning using a neural network.
Mail interpretation server.
The method according to any one of claims 1 to 3,
The machine learning model does not distinguish between a plurality of accounts, and is constructed based on the received mail determined to be returned.
Mail interpretation server.
The method according to any one of claims 1 to 3,
The machine learning model is updated based on the received mail determined to be the reply for each account
Mail interpretation server.
The method according to any one of claims 1 to 3,
The neural network is a character-level convolutional neural network.
Mail interpretation server.
Determine the presence or absence of a reply to the accumulated received e-mail,
Analyzing at least one of the characters included in the received mail determined to have been answered and the characters included in the received mail determined not to be answered, and constructing a machine learning model by machine learning using a neural network,
Including estimating the necessity of a reply to the received mail using the machine learning model for the text included in the received mail,
The presence or absence of a reply to the accumulated received mail,
Comparing the accumulated received mail and accumulated transmitted mail, it is determined whether or not a part of the body of the received mail is included in the transmitted mail,
If the sending mail contains a part of the body of the received mail, it is determined that the received mail has been returned.
How to interpret mail.
Determine the presence or absence of a reply to the accumulated received e-mail,
Analyzing at least one of the characters included in the received mail determined to have been answered and the characters included in the received mail determined not to be answered, and constructing a machine learning model by machine learning using a neural network,
Including estimating the necessity of a reply to the received mail using the machine learning model for the text included in the received mail,
The presence or absence of a reply to the accumulated received mail,
Comparing the accumulated received mail and accumulated transmitted mail, it is determined whether or not additional information of the received mail is included in the transmitted mail,
If the outgoing mail contains additional information of the received mail, it is determined that the received mail has been returned.
How to interpret mail.
Determine the presence or absence of a reply to the accumulated received e-mail,
Analyzing at least one of the characters included in the received mail determined to have been answered and the characters included in the received mail determined not to be answered, and constructing a machine learning model by machine learning using a neural network,
With respect to the text included in the received mail, the machine learning model is used to estimate the need for a reply to the received mail,
After determining the presence or absence of a reply to the accumulated received mail, calculating a time required to reply to the accumulated received mail,
Analyzing the text included in the received mail and the time required until the reply, and constructing the machine learning model by machine learning using a neural network
How to interpret mail.
The method according to any one of claims 7 to 9,
The neural network is a character-level convolutional neural network.
How to interpret mail.
The method according to any one of claims 7 to 9,
The machine learning model does not distinguish between a plurality of accounts, and is generated based on the received mail determined to be returned.
How to interpret mail.
The method according to any one of claims 7 to 9,
The machine learning model is updated based on the received mail determined to be the reply for each account
How to interpret mail.
The mail analysis method according to any one of items 7 to 9,
For running on the computer
program.

Images