JP7598005B2 - Information processing device, information processing method, and program - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and a program.

As the scope of application of the Personal Information Protection Act has been expanded following amendments, the risk of information leaks for businesses has increased, and measures to prevent personal information leaks have become increasingly important.

When handling personal information, it is necessary to take necessary measures to safely manage the personal data, but because electronic personal data is easily copied, it can easily become difficult to manage, for example when copied files are left behind when working on an individual's device, making it difficult to manage.

In addition, when sending data externally via e-mail or other means, it is necessary to check not only for erroneous sending but also for files that contain personal information being attached by mistake.

However, for an organization, it would be extremely costly to check all the files stored on devices managed by each individual, or to open and check files attached to emails each time.

In order to properly manage personal information within organizations while keeping huge operational costs down, there is a demand for technology that can identify with a high degree of accuracy which files contain personal information.

Patent Document 1 describes a technology that detects personal information forming information that may form personal information, identifies the personal information from the positional relationship of the detected personal information forming information, measures the number of characters and items of the personal information, and displays the measured value as an index value indicating the degree to which personal information is included.

JP 2008-117287 A

However, in Patent Document 1, the formula for calculating the index value indicating the degree to which personal information is included must be manually adjusted, and it takes time and effort to calculate an appropriate index value.

Therefore, the present invention aims to provide a mechanism for determining whether a document contains personal information more efficiently and accurately.

The information processing system of the present invention is characterized by comprising an acquisition means for acquiring text data, and a determination means for determining whether the text data acquired by the acquisition means contains personal information using a trained model trained on data that associates the text data with information indicating that the text data contains personal information.

The present invention makes it possible to more efficiently and accurately identify documents that contain personal information.

1 is a diagram showing an example of a system configuration of a Web mail service using a personal information inspection device in an embodiment of the present invention. 2 is a block diagram showing an example of a hardware configuration of a personal information inspection device, a Web mail service, an administrator terminal, and a client terminal in an embodiment of the present invention. 1 is a diagram illustrating an example of a functional configuration of a personal information inspection device and a Web mail service in an embodiment of the present invention. 11 is a flowchart illustrating an example of a process for learning personal information in an embodiment of the present invention. 2 is a diagram showing an example of a configuration of a judgment knowledge storage area in the embodiment of the present invention. FIG. 1 is a flowchart showing an example of a process for generating a feature vector for a document in an embodiment of the present invention. FIG. 4 is a diagram showing an example of learning data in the embodiment of the present invention. 11 is a diagram showing an example of personal information forming information acquired from a document in the embodiment of the present invention. FIG. FIG. 4 is a diagram showing an example of a feature vector in the embodiment of the present invention. FIG. 4 is a diagram showing an example of a feature amount storage table in the embodiment of the present invention. FIG. 2 is a diagram showing an example of a machine learning model acquired as a result of learning in an embodiment of the present invention. 11 is a flowchart showing an example of a personal information index calculation process in the embodiment of the present invention. FIG. 4 is a diagram showing an example of a personal information index calculation formula in the embodiment of the present invention. FIG. 11 is a diagram showing an example of a process of calculating a predicted value in the embodiment of the present invention. 10 is a flowchart showing an example of an inspection process of an e-mail in a Web mail service according to an embodiment of the present invention. FIG. 13 is a diagram showing an example of a mail editing screen in a Web mail service according to an embodiment of the present invention. FIG. 13 is a diagram showing an example of a transmission confirmation screen in a Webmail service in the embodiment of the present invention.

The following describes an embodiment of the present invention in detail with reference to the drawings.

Figure 1 shows an example of the system configuration of a webmail service using a personal information inspection device in an embodiment of the present invention.

The personal information inspection device 100, the webmail service 110, and the administrator terminal 120 are configured to be connected via a local area network 130. The webmail service 110 is also configured to be connectable to a client 150 via an external network 140.

The Webmail service 110 provides the client terminal 150 with the functions of a typical email client, i.e., the ability to edit email and send and receive email.

In addition, when a user on the client terminal 150 sends an email or explicitly instructs the inspection of personal information, the webmail service 110 requests the personal information inspection device 100 to execute a process to determine whether or not personal information exists.

If the personal information inspection device 100 determines that the email to be sent may contain personal information, the webmail service 110 displays a message to the user prompting the user to confirm whether or not the email contains personal information and to decide whether or not to send the email, and controls the sending based on the user's decision on whether or not to send the email.

In an embodiment of the present invention, the personal information inspection device 100 may be configured as a service provided to the client terminal 150 via the external network 140, or may be configured as part of the functions of the webmail service 110.

It may also be configured as a function of an email client program running on the client terminal 150.

Figure 2 is a block diagram showing an example of the hardware configuration of the personal information inspection device 100, webmail service 110, administrator terminal 120, and client terminal 150 in an embodiment of the present invention.

As shown in FIG. 2, the information processing device is connected to a CPU (Central Processing Unit) 201, a ROM (Read Only Memory) 202, a RAM (Random Access Memory) 203, a storage device 204, an input controller 205, an audio controller 206, a video controller 207, a memory controller 208, and a communication I/F controller 209 via a system bus 200.

The CPU 201 provides overall control over each device and controller connected to the system bus 200.

The ROM 202 or external memory 213 holds the BIOS (Basic Input/Output System) and OS (Operating System), which are control programs executed by the CPU 201, computer-readable and executable programs for implementing this information processing method, and various necessary data (including data tables).

RAM 203 functions as the main memory, work area, etc. of CPU 201. CPU 201 loads programs and the like required for executing processing from ROM 202 or external memory 213 into RAM 203, and executes the loaded programs to realize various operations.

The input controller 205 controls input from input devices such as a keyboard 210 and a pointing device such as a mouse (not shown). If the input device is a touch panel, the user can give various instructions by pressing (touching with a finger, etc.) icons, cursors, or buttons displayed on the touch panel.

The touch panel may also be a touch panel capable of detecting positions touched by multiple fingers, such as a multi-touch screen.

The video controller 207 controls the display on an external output device such as a display 212. The display is also intended to include the display of a notebook computer integrated into the main body. Note that the external output device is not limited to a display, and may be, for example, a projector. In addition, for devices capable of accepting the aforementioned touch operations, an input device is also provided.

The video controller 207 can control a video memory (VRAM) for display control, and can use part of the RAM 203 as a video memory area, or can provide a separate dedicated video memory.

The memory controller 208 controls access to the external memory 213. The external memory can be an external storage device (hard disk) that stores the boot program, various applications, font data, user files, edit files, and various data, a flexible disk (FD), or a compact flash (registered trademark) memory connected to a PCMCIA card slot via an adapter.

The communication I/F controller 209 connects and communicates with external devices via a network, and executes communication control processing on the network. For example, communication using TCP/IP, telephone lines such as ISDN, and 4G and 5G lines for mobile phones are possible.

The CPU 201 can display the outline font on the display 212 by, for example, executing a process of expanding (rasterizing) the outline font in a display information area in the RAM 203. The CPU 201 can also allow the user to give instructions using a mouse cursor (not shown) on the display 212.

Figure 3 shows an example of the functional configuration of a personal information inspection device and a webmail service.

The email editing processing unit 311 has the function of creating and editing emails based on operations and instructions from the user received via a screen displayed on the web browser 320 of the client terminal 150.

When the transmission control processing unit 312 receives an instruction from a user to send an e-mail, it extracts text (such as the body and attachments) from the e-mail and controls the calculation process of the personal information index by each processing unit of the personal information inspection device 100. It also has a function to control the e-mail transmission process by displaying the confirmation dialog shown in FIG. 17 based on the result of the calculation process of the personal information index, and by receiving a permission or prohibition from the user.

The transmission processing unit 313 has the function of sending emails that have been decided to be sent by the transmission control processing unit 312 to the external network 140.

The personal information learning processing unit 301 has a function to execute the personal information learning process shown in the flowchart of FIG. 4 based on the information uploaded from the administrator terminal 120.

The feature vector generation processing unit 302 has the function of generating a feature vector for the document to be processed.

The judgment knowledge storage area 303 is an area that stores pairs of feature vectors generated by the feature vector generation processing unit 302 and labels indicating personal information or non-personal information that are assigned to each document in the training data.

The personal information inspection processing unit 304 has a function to execute a process to calculate the possibility that the text content shown in FIG. 12 is personal information for the text obtained from the transmission control processing unit 312.

(Personal information learning processing)
Next, the personal information learning process executed by the personal information learning processing unit 301 in the embodiment of the present invention will be described with reference to the flowchart of FIG.
Figure 4 is a flowchart showing the process in which the CPU 201 of the personal information inspection device 100 reads and executes a specified control program, and which learns the characteristics of personal information based on documents uploaded from the administrator terminal 120 that have been pre-classified into personal information and non-personal information.

In step S401, the personal information learning processing unit 301 acquires the document uploaded from the web browser 330 in the administrator terminal 130 as learning data.

In step S402, the personal information learning processing unit 301 starts repeated processing up to step S405 for the documents in the learning data acquired in step S401.

In step S403, the personal information learning processing unit 301 generates a feature vector for the document to be processed using the feature vector generation processing unit 302. The method of generating the feature vector will be described later.

In step S404, the personal information learning processing unit 301 stores a pair of the feature vector generated in step S403 and a label indicating personal information or non-personal information assigned to each document in the learning data in the feature vector storage table 501 in the judgment knowledge storage area 303 shown in FIG. 5.

In step S405, if there are still documents to be processed, the personal information learning processing unit 301 repeats the process from step S402. If there are no documents to be processed, the process proceeds to step S406.

In step S406, the personal information learning processing unit 301 uses machine learning to learn the feature vectors and labels (personal information and non-personal information) in the feature vector storage table 501, and acquires a model (trained model) as a result of the learning.

In step S407, the personal information learning processing unit 301 stores the model acquired in step S406 in the model storage table 502 in the judgment knowledge storage area 303.

In the embodiment of the present invention, the personal information learning process is configured to be executed on the same hardware as the personal information inspection process, but it may be configured to be executed on different hardware.
Although the learning data is configured to be uploaded, it may be configured to be managed on the hardware that performs the personal information learning process.

(Feature Vector Generation Process)
Next, the feature vector generation process executed by the feature vector generation processing unit 302 in the embodiment of the present invention will be described with reference to the flowchart in FIG.

In step S601, the feature vector generation processing unit 302 uses morphological analysis and pattern matching to extract personal information formation information, together with the position of appearance and the number of characters, from the text of the document received as input, based on words and patterns previously defined as personal information items or sensitive information items.

In step S602, the feature vector generation processing unit 302 calculates statistical values based on the type and number of characters from the document and the extracted personal information formation information.

In step S603, the feature vector generation processing unit 302 compiles the statistical values obtained in step S602 into a feature vector.

(Specific example of personal information learning processing)
Next, as a specific example of the personal information learning process, a case where the process shown in FIG. 4 is performed on learning data 700 shown in FIG. 7 will be described.

In step S401, the personal information learning processing unit 301 acquires the document 700 uploaded from the web browser 330 in the administrator terminal 130 as learning data.

In step S402, the personal information learning processing unit 301 starts repeated processing up to step S405 for the document 701 in the learning data 700 acquired in step S401.

In step S403, the personal information learning processing unit 301 generates a feature vector for the document 701 to be processed using the feature vector generation processing unit 302.

In step S601, the feature vector generation processing unit 302 extracts personal information formation information 801 shown in FIG. 8 from the text of the document 701 received as input, using morphological analysis and pattern matching, based on words and patterns previously defined as personal information items and sensitive information items, along with the occurrence positions and number of characters.

In step S602, the feature vector generation processing unit 302 obtains statistical values based on the type and number of characters shown in FIG. 9 from the document and the extracted personal information formation information 801.

In step S603, the feature vector generation processing unit 302 compiles the statistical values calculated in step S602 into a feature vector 901. In the embodiment of the present invention, processes such as normalization are omitted for ease of explanation, but normalization may be performed.

In step S404, the personal information learning processing unit 301 stores the feature vector 901 generated in step S403 and a pair 1001 of the label TRUE, which indicates personal information or non-personal information, assigned to each document in the learning data, in the feature vector storage table 501 shown in FIG. 10.

In step S405, the personal information learning processing unit 301 repeats the process from step S402 because there is still a document 702 to be processed.

The same steps are then performed for all documents in the training data 700.

In step S406, the personal information learning processing unit 301 performs machine learning on the feature vector storage table 501 to acquire a model.

In this embodiment of the present invention, a linear SVM (Support Vector Machine) is used.

In step S407, the personal information learning processing unit 301 stores the model shown in FIG. 11 acquired as the learning result in the model storage table 502 in the judgment knowledge storage area 303. The acquired model is judged to be personal information if the predicted value (calculation result) is a positive value.

In the embodiment of the present invention, for ease of explanation, a linear SVM was used as the machine learning method, but any method can be used as long as it can be classified into two values, true or false, and the degree of the classification can be expressed as a numerical value.

(Personal information inspection processing)
Next, the personal information inspection process executed by the personal information inspection processing unit 304 in the embodiment of the present invention will be described with reference to the flowchart of FIG.

The flowchart in FIG. 12 shows the process in which the CPU 201 of the personal information inspection device 100 reads and executes a specific control program, and is a flowchart showing the process of calculating, for an input text, the degree to which the content of the text is likely to be personal information as a numerical value.

In step S1201, the personal information inspection processing unit 304 acquires the text of the document to be processed.

In step S1202, the personal information inspection processing unit 304 obtains a feature vector for the input document using the flowchart of FIG. 6 described above for the text obtained in step S1201.

In step S1203, the personal information inspection processing unit 304 calculates a predicted value for the feature vector acquired in step S1202 using the model stored in the model storage table 502 in the judgment knowledge storage area 303.

In step S1204, the personal information inspection processing unit 304 converts the predicted value calculated in step S1203 into a value that is easy to grasp intuitively, and outputs the converted value as a personal information index.
In the embodiment of the present invention, the personal information index calculation formula shown in FIG. 13 is used, but other calculation formulas may be used, or the predicted value may be used as is.

(Specific examples of personal information inspection processing)
Next, as a specific example of personal information inspection processing, a case where the processing shown in FIG. 12 is performed on an inspection target document 1401 shown in FIG. 14 will be described.

In step S1201, the personal information inspection processing unit 304 obtains text 1402 from the document 1401 to be processed.

In step S1202, the personal information inspection processing unit 304 obtains a feature vector 1403 for the input document using the flowchart of FIG. 6 described above for the text obtained in step S1201.

In step S1203, the personal information inspection processing unit 304 calculates a predicted value of 1460.960 for the feature vector 1403 acquired in step S1202 using the model stored in the model storage table 502 in the judgment knowledge storage area 303 (1404).

In step S1204, the personal information inspection processing unit 304 calculates the personal information index for the predicted value 1460.960 calculated in step S1203, using the personal information index calculation formula shown in Fig. 13. Here, it is assumed that Rmin = -1700 and Rmax = 2000.
As a result, the personal information index is obtained as 50 + 50 x 1460.960 ÷ max (2000, - (-1700)) = 86.524.

In this way, by simply preparing a set of documents that have been determined to contain personal information, it is possible to calculate a personal information index, which is the degree of certainty that the documents are personal information, without the need for manual adjustment.
Furthermore, since the personal information index is calculated based on statistical information, even if there is some error in the extraction of the personal information forming information, it is possible to indicate the degree of certainty that the information is personal information with high accuracy.
By using the calculated personal information index, processing according to the degree of accuracy can be realized, enabling efficient management of personal information.

Next, with reference to the flowchart shown in FIG. 15, a specific application example of the present invention will be described in which the sending of personal information is confirmed when sending an e-mail and feedback is obtained from the user.
In this embodiment, sending e-mail will be described as an application example, however, there are other application examples such as determining whether a document contains personal information when printing a document, or when uploading a document to a web service.

Figure 15 is a flow chart showing the process in which the CPU 201 of the Webmail service 110 reads and executes a specific control program, and is performed when a send instruction is given by pressing the send button 1602 on an email editing screen 1601 such as that shown in Figure 16 and displayed on the Web browser 320 of the client terminal 150.

In step S1501, the transmission control processing unit 312 receives an instruction to send an email.

In step S1502, the transmission control processing unit 312 starts repeating the process up to step S1509 for the file attached to the email instructed to be sent. Note that in addition to the attached file, the body of the email may also be targeted.

In step S1503, the transmission control processing unit 312 extracts text from the attachment to be processed. If the attachment is an image, the text is obtained by performing OCR processing on the image.

In step S1504, the transmission control processing unit 312 calculates a personal information index for the text extracted in step S1503 using the personal information inspection device 100.

In step S1505, if the personal information index calculated in step S1504 is equal to or greater than the specified value, the transmission control processing unit 312 transfers the process to step S1506. If it is less than the specified value, the process transfers to step S1509.

In step S1506, the transmission control processing unit 312 displays the confirmation dialog shown in FIG. 17, and obtains the user's decision as to whether or not to transmit the information and whether or not the information is personal information.

In step S1507, the transmission control processing unit 312 sends the user's decision as to whether or not the information acquired in step S1505 is personal information to the personal information inspection device 100, and instructs the personal information inspection device 100 to store the decision in the feature vector storage table 501 together with the feature vector being processed.

In step S1508, if the transmission permission acquired in step S1505 is "send," the transmission control processing unit 312 proceeds to step S1509. If the transmission permission is "cancel," the processing ends.

In step S1509, if there are still attachments to be processed, the transmission control processing unit 312 repeats the process from step S1502. If there are no more attachments to be processed, the process proceeds to step S1510.

In step S1510, the transmission control processing unit 312 sends the e-mail to the transmission processing unit 313 and sends the e-mail.

In this way, by collecting the user's judgment as to whether or not something is personal information, the personal information inspection device 100 can periodically perform a learning process, making it possible to calculate a personal information index that matches the user's judgment without the need for manual adjustment.

In the embodiment of the present invention, learning is performed from all stored feature vectors, but when storing feature vectors, information identifying the user or the department to which the user belongs may be added, and learning may be performed for each department or user. Then, for documents related to the user or the department to which the user belongs, the model generated by this learning may be used to determine whether or not personal information is included.

The present invention can be embodied, for example, as a system, device, method, program, or recording medium. Specifically, it may be applied to a system made up of multiple devices, or to a device made up of a single device.

The program of the present invention is a program that enables a computer to execute the processing methods of the flowcharts shown in Figures 4 and 12, and the storage medium of the present invention stores a program that enables a computer to execute the processing methods of Figures 4 and 12. The program of the present invention may be a program for each processing method of each device in Figures 4 and 12.

As described above, it goes without saying that the object of the present invention can be achieved by supplying a recording medium on which a program that realizes the functions of the above-mentioned embodiments is recorded to a system or device, and having the computer (or CPU or MPU) of that system or device read and execute the program stored on the recording medium.

In this case, the program read from the recording medium itself realizes the novel functions of the present invention, and the recording medium on which the program is recorded constitutes the present invention.

Recording media for supplying the program may include, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a DVD-ROM, a magnetic tape, a non-volatile memory card, a ROM, an EEPROM, a silicon disk, etc.

Furthermore, it goes without saying that not only are the functions of the above-mentioned embodiments realized by the computer executing a program it has read, but also that the functions of the above-mentioned embodiments can be realized by an operating system (OS) or the like running on the computer carrying out some or all of the actual processing based on the instructions of the program.

Furthermore, it goes without saying that this also includes cases where a program read from a recording medium is written into a memory provided on a function expansion board inserted into a computer or a function expansion unit connected to a computer, and then a CPU or the like provided on the function expansion board or function expansion unit performs some or all of the actual processing based on the instructions of the program code, thereby realizing the functions of the above-mentioned embodiments.

The present invention may be applied to a system made up of multiple devices, or to a device made up of a single device. Needless to say, the present invention can also be applied to cases where the effects of the present invention are achieved by supplying a program to a system or device. In this case, the effects of the present invention can be enjoyed by reading a recording medium that stores a program for achieving the present invention into the system or device.

Furthermore, by downloading and reading a program for achieving the present invention from a server, database, etc. on a network using a communication program, the system or device can enjoy the effects of the present invention. Note that the present invention also includes configurations that combine the above-mentioned embodiments and their variations.

100 Personal information inspection device 110 Web mail service 120 Administrator terminal 130 LAN
140 External network 150 Client terminal 150

Claims (14)

An acquisition means for acquiring text data from an email ;
an output means for outputting a personal information index in the text data acquired by the acquisition means, using a trained model trained on data in which the text data is associated with information indicating that the text data contains personal information;
A receiving means for receiving an instruction to send an e-mail;
a selection receiving means for receiving a selection as to whether or not personal information is included in the text data obtained from the e-mail in response to receiving an instruction to send the e-mail by the receiving means when a personal information index related to the text data obtained from the e-mail satisfies a predetermined condition;
Equipped with
An information processing system characterized in that , when the selection receiving means receives a selection indicating that personal information is included, the trained model is trained using the text data . a transmission control means for suspending transmission of the e-mail when a personal information index relating to the text data acquired from the e-mail satisfies a predetermined condition;
2. The information processing system according to claim 1, further comprising: 3. The information processing system according to claim 2, wherein said selection receiving means further receives a selection as to whether or not to permit transmission of the electronic mail the transmission of which has been suspended by said transmission control means. The information processing system according to claim 3, characterized in that the transmission control means controls the transmission of an email whose transmission has been put on hold because a personal information index related to text data acquired from the email satisfies a predetermined condition, when the selection receiving means receives a selection as to whether or not the text data acquired from the email contains personal information and a selection to allow the transmission. The information processing system according to any one of claims 2 to 4, characterized in that if the personal information index relating to the text data obtained from the e-mail does not satisfy a predetermined condition, the transmission control means sends the e-mail without accepting a selection by the selection receiving means as to whether or not personal information is included. An extraction means for extracting personal information forming information from the text data ;
The information processing system according to any one of claims 1 to 5, further comprising a generation means for generating a trained model using data correlating the personal information formation information extracted by the extraction means with information on whether or not the personal information set in the text data from which the personal information formation information was extracted is included. The information processing system according to claim 6, characterized in that the generation means generates a trained model using data correlating the number of extractions of each type of personal information formation information extracted by the extraction means with information on whether or not the personal information set in the text data from which the personal information formation information was extracted is included. storing the selection of whether or not personal information is included , which is received by the selection receiving means, in association with information on the user who made the selection or the user's affiliation;
The information processing system according to claim 6 or 7 , characterized in that the generation means generates a trained model for each user or each affiliation of the user who made the selection . 9. The information processing system according to claim 1, wherein the predetermined condition is satisfied when the personal information index output by the output means is equal to or greater than a predetermined value. 10. The information processing system according to claim 1, wherein the text data includes at least one of the text of the email and text included in a file attached to the email. A program for causing a computer to function as each of the means according to any one of claims 1 to 10 . An acquisition step in which an acquisition means of the information processing system acquires text data from an email ;
an output step in which an output means of the information processing system outputs a personal information index in the text data acquired by the acquisition means using a trained model trained using data in which text data is associated with information indicating that the text data contains personal information;
a receiving step of receiving an instruction to send an e-mail by a receiving means of the information processing system;
a selection receiving step in which, when a personal information index related to the text data acquired from the email satisfies a predetermined condition, the selection receiving means of the information processing system receives a selection as to whether or not personal information is included in the text data acquired from the email in response to receiving an instruction to send the email by the receiving step;
Equipped with
An information processing method, characterized in that , when a selection indicating that personal information is included is received in the selection receiving step, the trained model is trained using the text data . An information processing device capable of communicating with an apparatus including: an acquisition means for acquiring text data; and an output means for outputting a personal information index in the text data acquired by the acquisition means using a trained model trained using data in which the text data is associated with information indicating that the text data includes personal information;
a transmission means for transmitting text data acquired from the email instructed to be transmitted to the device;
An acquisition means for acquiring a personal information index output by an output means of the device;
a transmission control means for controlling the transmission of the electronic mail instructed to be sent when the personal information satisfies a predetermined condition based on the personal information index acquired by the acquisition means;
a selection receiving means for receiving a selection as to whether or not personal information is included in text data acquired from the email whose transmission is controlled by the transmission control means;
A control means for controlling the device to train the trained model using the text data when the selection receiving means receives a selection indicating that personal information is included;
An information processing device comprising: An information processing device capable of communicating with an apparatus including: an acquisition means for acquiring text data; and an output means for outputting a personal information index in the text data acquired by the acquisition means using a trained model trained on data in which the text data is associated with information indicating that the text data contains personal information;
a transmission means for transmitting text data acquired from the email instructed to be transmitted to the device;
An acquisition means for acquiring a personal information index output by an output means of the device;
a transmission control means for controlling the transmission of the electronic mail instructed to be sent when the personal information satisfies a predetermined condition based on the personal information index acquired by the acquisition means;
a selection receiving means for receiving a selection as to whether or not personal information is included in text data acquired from the email whose transmission is controlled by the transmission control means;
A control means for controlling the device to train the trained model using the text data when the selection receiving means receives a selection indicating that personal information is included;
A program to make it function as such.

Images