JP2001331422A - Mail evaluation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically evaluate the priority of a received mail or the like by automatically learning the keyword of a mail text regarding a mail evaluation device for evaluating the priority of the mail or the like by automatic learning and a recording medium. SOLUTION: This device is provided with a means for calculating reference time per unit character based on the total reference time of the received mail and the number of characters, comparing it with statistic data regarding the reference time of the plural pieces of the mail received in the past and imparting a rank to the mail, a means for recording points for the respective ranks for the various kinds of the keywords, a means for updating the points for the respective ranks of the keyword in a keyword information table corresponding to the rank added to the mail for an extracted keyword in the received mail and a means for extracting the plural keywords from the received mail and evaluating the contents of the received mail based on the result of totaling the points for the respective ranks of the keywords in the keyword information table means corresponding to the keywords for the respective ranks.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mail evaluation device for automatically evaluating the priority of received mail and the like and a recording medium storing a program for realizing the mail evaluation function.

[0002] In recent years, with the development of the Internet, E-Mail (e-mail) has been used in many companies and ordinary households, and various kinds of information have been transmitted.

Under such circumstances, news mail distribution such as communication between companies and internal company information, information distribution, or E-Mail (hereinafter simply referred to as e-mail) from a mobile terminal such as a mobile phone. Due to widespread use, it is not uncommon for the number of emails received per day to exceed 100. As a result, there is also a problem of convenience, that is, being busy with the processing of the received mail.

[0004] For this reason, a mechanism is needed to evaluate which of the received large amounts of mail is the most important one to be read first, and to automatically prioritize the mail. .

[0005]

2. Description of the Related Art E-mail reception by conventional e-mail software is as follows.
All received mails were stored in a receiving folder called an inbox. For example, as shown in FIG.
All emails received from the company, company B, company general affairs department, etc. to their own terminals are stored in the inbox and the corresponding file icons displayed on the screen (for example, company A, company B, etc.)
By clicking, a list of the received mails of the file is displayed on the screen.

Conventionally, however, the only function for sorting these received mails is to sort the received mail into a corresponding folder only when there is a keyword registered in advance from the sender name (From) or the title (Subject). In order to know whether an email is important, the user must determine the importance of the email by displaying (referencing) the text of the email on the screen and reading it. Information was insufficient, and it was not possible to accurately identify important emails. In particular, when there are a large number of received mails as described above, the user himself / herself must refer to and check all the mails for a great deal of labor.

[0007]

As described above, in the conventional sorting function based on the sender's name and the title, the user has to refer to the body of the mail to determine the importance of the mail. was there.

The present invention has been made in view of such a problem, and has as its object to automatically evaluate the priority of a received mail by automatically learning a keyword of the mail body.

[0009]

In order to solve the above-mentioned problems, a mail evaluation device according to the present invention calculates a reference time per unit character based on a total reference time and the number of characters of a received mail. A rank assigning means for assigning a rank to a plurality of mails received in the past, based on statistical data relating to the reference time; a keyword information table means for recording rank-specific points for various keywords; and a received mail. For keyword extracted from the keyword information, keyword information updating means for updating the rank-based points of the corresponding keyword in the keyword information table means according to the rank attached to the mail, and extracting a plurality of keywords from the received mail, Rank of the corresponding keyword in the corresponding keyword information table means Based point of the result of the sum for each rank, and a evaluation means for evaluating the contents of the mail and the received.

[0010]

In the e-mail evaluation device of the present invention, upon receiving a mail reception request from a user, keywords are extracted from the body of the received e-mail, and for all extracted keywords,
The points by rank (for example, high rank points / medium rank points / low rank points) taken out from the keyword information table means are totaled for each rank, and the rank of the received mail is evaluated based on the highest rank point. Then, if necessary, the corresponding mail is stored in a rank-specific folder corresponding to the rank.

[0011] In the stored mail, a keyword is extracted again when the mail is referred to, and for the extracted keyword, statistical data (for example, a deviation value) obtained based on the total reference time and the number of characters indicates the rank of the mail. Points are added to rank-specific points in the keyword information table.

Therefore, the keyword information table is constantly updated by the automatic learning, so that the received mail is automatically sorted according to rank (for example, high priority / medium priority / low priority) according to, for example, a matter indicating the user's interest. Can be performed.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a functional block diagram of a mail evaluation device as one embodiment according to the present invention. As shown in the figure, the apparatus of this embodiment includes a mail transmitting / receiving unit 1, a priority sorting unit 2, a display unit 3, a keyword extracting unit 4, an initial setting information table 5, an initial mail folder 6, a priority mail folder 7, a keyword information It comprises a table 8, a deviation value information table 9, and the like. The priority mail folder 7 includes a high priority mail folder 7H, a medium priority mail folder 7M, and a low priority mail folder 7
There are three types of L.

The mail transmitting / receiving unit 1 is a functional block for transmitting / receiving mail via a network. When the system is initially installed, a predetermined number (for example, 50) of received mails are stored in the initial mail folder 6, and thereafter, the received mails are transferred. It has a function to pass to the priority distribution unit 2.

The priority distribution unit 2 analyzes the received mail data in the initial mail folder 6 and the keywords in the received mail from the keyword information table 8 and assigns a priority to the received mail based on the keywords. It has a function of storing received mail in the corresponding priority mail folder 7.

The display unit 3 has a function of displaying received mail stored in the initial mail folder 6 and the priority mail folder 7 on a screen, and recording the total reference time, the number of characters in the mail, and the like.

The keyword extracting section 4 has a function of extracting a keyword from a received mail, and adding / updating a record in the keyword information table 8 based on the keyword and the contents of the deviation value information table 9.

FIG. 10 shows the data structure of the initial setting information table 5. The initial setting information table 5 is a table in which various data necessary for the priority determination calculation of the received mail are initially set. The meaning of each item in the table is as follows. -Number of deviation value calculation starts The number of mail receptions from the time of initial introduction until the calculation of deviation value (for example, initial value: 50). -High priority deviation threshold This is a higher average value (for example, initial value: 55) in the deviation value judgment. -Low priority deviation threshold This is the lower average value (eg, initial value: 45) in the deviation value judgment.・ Printing time This is the time added to the total reference time when printing the mail while referring to the mail. -Copy time This is the time added to the total reference time when copying to the clipboard while referring to an email. Reference upper limit The upper limit of the total reference time (for example, the initial value: 1800 seconds). The reference upper limit is set in the initial setting information table based on the total reference time in consideration of the case where the received mail is left on the screen. The maximum value of the reference time is defined, and if there is a total reference time that is longer than that, the total reference time is processed as this reference upper limit (maximum value). -Number of received mails This is the number of received mails (initial value: 0).

FIG. 11 is a diagram showing a data structure of the keyword information table 8. The keyword information table 8 is
Temporary priority "high point / medium point / low point" based on the deviation value calculated from the received mail
And the content is updated according to the keyword extracted from the mail. The meaning of each item in the table is as follows.・ Keywords These are keywords extracted from emails and keywords registered in advance at the beginning.・ High points The number of appearances of the keyword with high priority (from the time of introduction)
[Initial value: 0].・ Medium point Number of occurrences of the keyword with medium priority (from the time of introduction)
[Initial value: 0].・ Low points The number of occurrences of the keyword at low priority (from the time of introduction)
[Initial value: 0]. -Magnification Magnification for adding points after referring to mail [Initial value:
1].

FIG. 12 is a diagram showing a data structure of the initial mail folder 6. The initial mail folder 6 is a folder for temporarily storing e-mails received at the time of initial introduction of the system until the number of deviation value calculation starts specified in the initial setting information table 5 is reached. The meaning of each item in the table is as follows.・ Send date and time It is the date and time when the email was sent. -Receiving date and time It is the date and time when the mail was received.・ Sender The sender who sent the e-mail (From).・ Title This is the title of the email (Subject). Read flag This flag is a flag [0: unread / 1: read] indicating whether the corresponding mail has already been read. -Stored file name This is the name of the file that stores the email text. -Number of characters This is the number of characters in the mail body. -Total reference time This is the time required when the corresponding mail was first referenced (displayed on the screen and read).

FIG. 13 is a diagram showing a data structure of the deviation value information table 9. The deviation value information table 9 is a table for managing and updating data necessary for calculating the deviation value of the received mail. The meaning of each item in the table is as follows. -Number of data This is the number of mails referred to, and is calculated as follows: Number of data (n) = Number of data (n) + 1. -Total reference time This is the total reference time per character, and is obtained by the following expression: total reference time (A) = total reference time (A) + reference time.・ Average of reference time This is the average of the reference time per character, and is obtained by the average of the reference time (B) = A / n. -Sum of reference time squares This is the sum of the squares of the reference time per character, and is calculated by the sum of the squares of the reference time (C) = the sum of the squares of the reference time (C) + the square of the reference time. - sample variance is calculated using to sample variance of the standard variation, sample variance (D) = determined with [C- (A ²⁾ / n] / n. -Standard deviation This is the standard deviation used for calculating the deviation value of the reference time for each mail, and is calculated by the standard deviation (E) = square root of (D).

FIG. 14 shows the data structure of the high-priority mail folder 7H, and FIG.
FIG. 16 is a diagram showing the data structure of M, and FIG. 16 is a diagram showing the data structure of the low-priority mail folder 7L. These priority mail folders 7 are folders for storing received mails according to their assigned priorities. The meaning of each item in each table is as follows.・ Send date and time It is the date and time when the email was sent. -Receiving date and time It is the date and time when the mail was received.・ Sender The sender who sent the e-mail (From).・ Title This is the title of the email (Subject). Read flag This flag is a flag [0: unread / 1: read] indicating whether the corresponding mail has already been read. -Stored file name This is the name of the file that stores the email text.

The operation of this embodiment will be described below with reference to the flowchart. FIGS. 3 to 9 are flowcharts showing one embodiment of the processing procedure in this embodiment apparatus. First, a schematic operation of the present system will be described.

Extract keywords from the received mail,
Keyword information table 8 for those keywords
, The points are calculated for each priority (high point / medium point / low point), and the corresponding mail is stored in the priority folder corresponding to the highest priority point.

Then, when referring to the received mail,
The reference time in units of one character is calculated from the total reference time and the number of text characters. A deviation value is calculated based on this reference time, and a keyword point corresponding to the deviation value is added to the keyword information table 8 and used as determination information for the next reception.

The system of the embodiment will be described below in the order of processing of the mail transmitting / receiving section 1 / priority sorting section 2, processing of the display section 3 / keyword extracting section 4, initial mail folder expansion processing, and keyword registration.

[Mail transmission / reception unit 1 / priority distribution unit 2
3 and 4 are flowcharts showing an embodiment of a processing procedure in the mail transmitting / receiving section 1 and the priority allocating section 2. Hereinafter, this processing procedure will be described. In response to a mail reception request from the user when the user clicks the reception button, mail reception is performed, and the received mail is stored in the initial mail folder 6 (step S1). Regarding the received mail, the “number of received messages” in the initial setting information table 5 is counted up by one. Next, “the number of received data” and “the number of started deviation value calculation” in the initial setting information table 5 are compared (step S3).

As a result of the comparison, if the number of receptions is smaller, the processing is terminated. On the other hand, if the "number of deviation value calculation starts" is larger, the body of the corresponding mail is searched, and when two or more consecutive Chinese characters or katakana appear, the character string is extracted as a keyword (step S4).

The points are calculated after all keywords are extracted from the mail (step S).
5). This point calculation is performed as follows. The keyword information table 8 is searched for all the extracted keywords to check whether the corresponding keyword exists in the keyword information table 8, and if so, each point value of “high point / medium point / low point” for the keyword Take out. If the corresponding keyword is not in the keyword information table 8, "high points /
Each point value of "medium point / low point" is all "0"
And After performing this process for all the keywords extracted from the email, the total points of adding "high points", the total points of adding "medium points", and the points of adding "low points" for all the keywords Calculate the total. In this way,
The point sum value for each priority (high point / medium point / low point) is obtained.

Next, a priority determination is performed based on the point total value of "high point / medium point / low point" for each priority obtained by the above calculation (step S6). In this priority determination, the point having the largest value (the highest value of the total points) among the point sum values of the “high point / medium point / low point” for each priority is determined.

If the highest value is the point total value of "low points", the received mail is determined to be "low priority" and is moved from the initial mail folder 6 to the low priority mail folder 7L and stored therein ( Step S7). If the highest value is the total value of the "medium points", the received mail is determined to be "medium priority" and is moved from the initial mail folder 6 to the medium priority mail folder 7M and stored (step S8). . If the highest value is the total value of the "high points", the received mail is determined to be "high priority" and is moved from the initial mail folder 6 to the high priority mail folder 7H and stored (step S9).

[Processing of Display Unit 3 / Keyword Extraction Unit 4]
5 to 7 are flowcharts showing one embodiment of the processing procedure in the display unit 3 / keyword extraction unit 4. Less than,
This processing procedure will be described.

On the screen, the user clicks a reference button of the mail to be read (here, as shown in FIG. 2, the folder icons of the high-priority mail folder 7H, the medium-priority mail folder 7M, and the low-priority mail folder 7L are displayed. In response to a mail reference request from the user by clicking on the corresponding mail icon), the mail body is extracted from the priority mail folder 7 corresponding to the clicked reference button and displayed on the screen (step S11).

The number of characters of the displayed mail is obtained (step S12), and the time during which the mail is displayed on the screen is obtained as the total reference time (step S1).
3-S17). As a procedure for obtaining the total reference time, first, a reference start time is obtained (step S13), and it is monitored whether reference (screen display) of the corresponding mail is completed (step S14). ), And obtains the reference end time (step S16). And
A total reference time is calculated from the obtained start time and end time (step S17).

On the other hand, there is a case where a new mail is referred to while the corresponding mail is being referred to. Therefore, it is monitored whether a reference button for referring to the new mail is clicked (step S15). Assuming that the reference to the relevant mail has been completed at that time, the reference end time of the relevant mail is acquired as described above (step S16), and the total reference time is calculated (step S1).
7).

Next, it is determined whether or not the mail has been copied or printed while referring to the mail (step S1).
8). When copying or printing is performed, “copy time” or “print time” is obtained from the initial setting information table 5 and added to the calculated total reference time. This is because if an e-mail is copied or printed during reference, it can be assumed that the e-mail was copied or printed because it is an important e-mail. It is added to the time.

It is to be noted that, in consideration of the case where the mail text is left displayed while being displayed, the “reference upper limit” in the initial setting information table 5 is set.
Defines the maximum value of the total reference time, and if the calculated total reference time is equal to or greater than the maximum value, the total reference time is processed as the maximum value.

Next, the total reference time obtained for the mail text is divided by the number of characters to calculate a reference time per character (step S20). Here, it is expected that the user can read an important mail over time, and that the total reference time changes depending on the number of characters in the mail body. Therefore, it can be estimated that a mail having a longer reference time per character is a more important mail.

In the present invention, by using this principle, a provisional priority is assigned to a mail based on a deviation value calculated from the reference time per character, so that a keyword existing in the mail is assigned a priority (high). (Point / medium point / low point) and updates the keyword information table 8 using the priority.
As described in the flowchart of FIG. 4, the final mail priority determination is performed using the updated keyword information table 8.

After calculating the reference time per character (step S20), the “number of received data” in the initial setting information table 5 updated by the mail is replaced with the “number of deviation calculation started data” in the same table. (Step S2)
1). If the “number of received messages” is smaller than the “number of calculated deviation values”, it means that the minimum number of emails required to calculate the deviation value has not yet been received. It is stored in the mail folder 6 (step S25), and this processing flow ends.

On the other hand, if the “number of received messages” is larger than the “number of calculated deviation values” in the initial setting information table 5,
The content of the deviation value information table 9 is read out, the deviation value of the corresponding mail is calculated, and the content of the deviation value information table 9 is updated based on the calculation result (step S22). The formula for calculating this deviation value is shown below.

Reference time deviation value = 50 + [(reference time−average of reference time) / standard deviation of reference time] × 10 The calculation of each parameter in the above formula is performed as follows. -Number of data (n) = Number of data (n) + 1-Total of reference time (A) = Total of reference time (A) + reference time-Average of reference time (B) = A / n-Square value of reference time (C) = total of reference time squared value (C) + reference time squared value Sample variance (D) = [C− (A ² ) / n] / n Standard deviation (E) = (D ) Square root

The above parameters are necessary for calculating the deviation value. The reference time average B, the sample variance D, and the standard deviation E
Is the number of data n, reference time total A, reference time square total C
The parameter data to be stored include the number of data n, the accumulated reference time A, and the accumulated reference time square C
May be three.

Next, the keyword is extracted again from the mail (step S23), and it is checked whether or not the extracted keyword exists in the keyword information table 8 (step S24). If not, the keyword is newly added to the keyword information table 8 as a record (step S26).

Next, the rank (temporary priority, hereinafter simply referred to as priority) is determined for the deviation value obtained in step S22 (step S27). This deviation value determination is performed based on the “high priority deviation value threshold” and the “low priority deviation value threshold” in the initial setting information table 5.
The priority of the calculated deviation value is determined. If the calculated deviation value is smaller than the “low priority deviation value threshold value”, the “low point” of the record of the corresponding keyword in the keyword information table 8 is determined. 1 point (magnification: ×
1) Perform (Step S29).

Similarly, if the calculated deviation value is larger than the “low priority deviation value threshold value” and smaller than the “high priority deviation value threshold value”, the record of the corresponding keyword in the keyword information table 8 is stored. One point is added to the “middle point” (magnification: × 1) (step S30).

Similarly, if the calculated deviation value is larger than the “high priority deviation value threshold value”, one point is added to the “high point” of the record of the corresponding keyword in the keyword information table 8 (magnification: × 1) (Step S2)
9).

In this embodiment, one point is added per mail to the corresponding keyword record in the keyword information table 8 even when one mail includes a plurality of the same keywords. However, in addition to this, it is also possible to add points for the same number of keywords in one mail to the corresponding keyword record in the keyword information table 8.

The above-mentioned magnification may be able to estimate the importance of the mail even from the meaning of the character of the extracted keyword itself. For example, keywords such as “important”, “urgent”, and “confidential” are different from other keywords. Since it can be said that the mail is an important mail, the weight is increased for such a keyword. For example, the magnification is (× 3) for such a keyword, and the number of points to be added is increased. To make it bigger.

Next, it is determined whether the keyword extraction process has been performed up to the last sentence of the mail (step S31). If not, the process returns to the keyword extraction process of step S23. Thus, the keyword extraction and the updating of the keyword information table 8 are performed until the last sentence of the received mail body is reached, and when the last sentence is reached, it is determined whether or not the reference button of the new mail is clicked (step S32). If it has been clicked, the process returns to step S11 to perform the above-described display section / keyword extraction processing for the new mail. If not clicked, this processing flow ends.

[Expansion processing of the initial mail folder 6] As described above, when the system is started (at the time of introduction), since the number of received mails is small, a deviation value is calculated from the beginning based on those small mails. Then, an accurate average value cannot be obtained. For this reason, the initially received mail is stored in the initial mail folder 6 and stored in the initial mail folder 6 when the number of received mails becomes “the number of deviation calculation start cases” in the initial setting information table 5. All the e-mails are read out, the deviation value is calculated, and the contents of the deviation value information table 9 are formed / updated based on the result.

FIG. 8 and FIG. 9 are flowcharts showing an embodiment of the processing procedure of the expansion processing of the initial mail folder 6. Hereinafter, this processing procedure will be described.

It is determined whether the “number of received messages” of the initial mail folder 6 has become “the number of started deviation value calculations” (step S33). Ends. On the other hand, when “number of received messages” = “number of calculated deviation values”, one mail stored in the initial mail folder 6 is taken out (step S34), and the total reference time and the number of characters of the mail (display unit / keyword) From the data already recorded in the initial mail folder 6 in step 25 of the extraction unit processing flow, 1
The reference time per character is calculated (step S35), and based on the calculation result, the number of data n, the reference time total A, the reference time average B, and the reference time square total C are updated in the deviation value information table 9 (step S36). ). It is checked whether there is still a mail in the initial mail folder 6 (step S34),
If there is an e-mail, the e-mail reading step S34
And the same process is repeated until all the mails in the initial mail folder 6 are performed.

When the updating process of the deviation value information table 9 is completed for all the mails in the initial mail folder 6, the sample variance D and the standard deviation E are calculated based on the updated data of the deviation value information table 9, and the deviation is calculated. The corresponding item in the value information table 9 is updated (step S38).

Thereafter, in order to sort the priority of the received mail stored in the initial mail folder 6 at the initial time, one mail is read from the initial mail folder 6 again (step S39), and the total of the mail body is read. The reference time per character is calculated from the reference time and the number of characters (step S40). Based on the calculated reference time per character and the data of the deviation value information table 9, the deviation value is calculated again according to the above-described formula, and the deviation value information table 9 is updated based on the calculation result (step S4).
1).

Next, the calculated deviation value is determined in the same manner as described above. The mail is stored in the folder 7M (step S42). If the value is smaller than the "low priority deviation value threshold value", the corresponding mail is stored in the low priority mail folder 7L (step S4).
3) If it is larger than the "high priority deviation value threshold value",
The corresponding mail is stored in the high priority mail folder 7H.

Thereafter, steps S23 to S31 in the display part / keyword extraction part processing flow (FIGS. 5 to 7) (the processing within the dotted line in the processing flow) are performed (step S4).
5) A keyword is extracted from the received mail (mail received at the initial stage of system introduction) stored in the initial mail folder 6, and the keyword information table 8 is formed and updated. The mail from which the priority distribution and the keyword extraction have been completed is deleted from the initial mail folder 6 (step S47). As a result of the above processing, it is determined whether or not there is any mail stored in the initial mail folder 6 (step S47). If there is a corresponding mail in the initial mail folder 6, the processing after step S39 is performed again. The remaining mail is read from the initial mail folder 6 repeatedly. If there is no corresponding mail in the initial mail folder 6, this processing flow ends.

[Registration of Keyword] As described above, the keyword that determines the priority of the mail can be performed by the automatic learning as described above, but can be customized by user registration. This can be reflected in the priority determination from the next reception by setting a keyword and a magnification in the keyword information table 8 from the screen as described above.

In the embodiment described above, the mail evaluation apparatus according to the present invention is described as being realized on a computer (such as a personal computer) on which software is installed. However, the present invention is not limited to this.
For example, a program for realizing each function described in the embodiment is stored in a recording medium, and an arbitrary computer is installed from the recording medium, whereby the e-mail evaluation function can be realized in the computer. That is, the recording medium stores programs for causing a computer to function as each of the above-described function realizing means, and these programs are of course distributed and installed on a remote computer via a communication line. It may be.

(Supplementary Note 1) The reference time per unit character is calculated based on the total reference time and the number of characters of the received mail, and the calculated reference time is compared with statistical data relating to the reference times of a plurality of mails received in the past. A rank assigning means for assigning a rank to the mail,
A keyword information table means for recording rank-specific points for each keyword for each keyword; and a keyword for updating keywords according to the rank assigned to the corresponding keyword in the keyword information table means for keywords extracted from received mail. The information updating unit and a plurality of keywords are extracted from the received mail, and the received mail is determined based on the result obtained by summing the rank-based points of the corresponding keyword in the keyword information table corresponding to the keywords for each rank. An e-mail evaluation device comprising: an evaluation unit that evaluates the content of a mail. (Supplementary Note 2) The statistical data calculation start number is provided in advance, and the received mail is referred to until the number of received mails reaches the statistical data calculation start number. The e-mail evaluation device according to claim 1, wherein the e-mail evaluation device is configured to accumulate data together with the data and, when the number of received e-mails reaches the number of start of statistical data calculation, calculate the statistical data based on the received e-mails accumulated so far. (Supplementary note 3) Regarding the received mail, when referring to the received mail, the reference time per character is calculated from the total reference time and the number of characters of the mail body that were referred by the mail recipient for each mail, and the result and the past mail 2. A deviation value for each referenced mail is calculated based on the standard deviation calculated based on the reference time per character, and the priority of each mail is estimated based on the calculated deviation value for each mail. Email evaluation device. (Supplementary Note 4) In the received mail, a plurality of keywords for each mail are extracted, and a total value of the rank points for each mail is calculated based on the calculated rank points for each keyword. 4. The e-mail evaluation device according to any one of Appendices 1 to 3, which has a function of storing the corresponding e-mail in a receiving folder. (Supplementary note 5) The keyword information table means according to any one of Supplementary notes 1 to 4, wherein the keyword information table means is configured to allow a user to register an arbitrary keyword in advance and configured to allow arbitrary weighting of the recorded keyword. E-mail evaluation device. (Supplementary Note 6) A recording medium that stores a program for causing a computer to function as each function realizing unit of the mail evaluation device according to any one of Supplementary Notes 1 to 5.

[0061]

As described above, according to the present invention, the priority of received mail can be displayed at a glance, so that important mail can be distinguished without being confused even for a large amount of mail every day. As a result, it is possible to efficiently check and process mail in a short time. In addition, since it does not depend on the sender's mail software, it can be expanded to automatically learn not only the text, but also the sender's name and title, etc. It greatly contributes to efficiency improvement.

[Brief description of the drawings]

FIG. 1 is a diagram showing a mail priority determination device as an embodiment of the present invention in the form of a functional block configuration.

FIG. 2 is a diagram illustrating an example of a screen display in which mail priority is determined for a mail user by the apparatus according to the embodiment.

FIG. 3 is a flowchart (1/2) illustrating a processing procedure of a mail transmission / reception unit / priority assignment unit;

FIG. 4 is a flowchart illustrating a processing procedure (1/2) of a mail transmission / reception unit / priority assignment unit in the apparatus according to the embodiment.

FIG. 5 is a flowchart illustrating a processing procedure (2/2) of a mail transmission / reception unit / priority assignment unit in the apparatus according to the embodiment.

FIG. 6 is a flowchart illustrating a processing procedure (1/3) of a display unit / keyword extraction unit in the embodiment device.

FIG. 7 is a flowchart illustrating a processing procedure (2/3) of a display unit / keyword extraction unit in the embodiment device.

FIG. 8 is a flowchart illustrating a processing procedure (3/3) of a display unit / keyword extraction unit in the embodiment device.

FIG. 9 is a flowchart illustrating a processing procedure (1/2) of initial mail folder expansion in the apparatus according to the embodiment.

FIG. 10 is a flowchart illustrating a processing procedure (2/2) of initial mail folder expansion in the apparatus according to the embodiment.

FIG. 11 is a diagram illustrating an example of a data structure of an initial setting information table in the embodiment device.

FIG. 12 is a diagram illustrating an example of a data structure of a keyword information table in the example device.

FIG. 13 is a diagram illustrating an example of a data structure of an initial mail folder in the embodiment device.

FIG. 14 is a diagram illustrating an example of a data structure of a deviation value information table in the apparatus according to the embodiment.

FIG. 15 is a diagram illustrating an example of a data structure of a high-priority mail folder in the embodiment device.

FIG. 16 is a diagram illustrating an example of a data structure of a medium priority mail folder in the embodiment device.

FIG. 17 is a diagram illustrating an example of a data structure of a low-priority mail folder in the embodiment device.

FIG. 18 is a diagram showing an example of screen display to a mail user by a conventional method.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Mail transmission / reception part 2 Priority distribution part 3 Display part 4 Keyword extraction part 5 Initial setting information table 6 Initial mail folder 7 Priority mail folder 8 Keyword information table 9 Deviation value information table

────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] July 4, 2000 (200.7.4)

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Brief description of drawings

[Correction method] Change

[Correction contents]

[Brief description of the drawings]

FIG. 1 is a diagram showing a mail priority determination device as an embodiment of the present invention in the form of a functional block configuration.

FIG. 2 is a diagram illustrating an example of a screen display in which mail priority is determined for a mail user by the apparatus according to the embodiment.

FIG. 3 is a flowchart illustrating a processing procedure (1/2) of a mail transmission / reception unit / priority assignment unit in the apparatus according to the embodiment.

FIG. 4 is a flowchart illustrating a processing procedure (2/2) of a mail transmission / reception unit / priority assignment unit in the apparatus according to the embodiment.

FIG. 5 is a flowchart illustrating a processing procedure (１ ／) of a display unit / keyword extraction unit in the apparatus according to the embodiment.

FIG. 6 is a flowchart illustrating a processing procedure (2/3) of a display unit / keyword extraction unit in the embodiment device.

FIG. 7 is a flowchart illustrating a processing procedure (3/3) of a display unit / keyword extraction unit in the embodiment device.

FIG. 8 is a flowchart illustrating a processing procedure (1/2) of initial mail folder expansion in the apparatus according to the embodiment.

FIG. 9 is a flowchart illustrating a processing procedure (2/2) of initial mail folder expansion in the apparatus according to the embodiment.

FIG. 10 is a diagram illustrating an example of a data structure of an initial setting information table in the embodiment device.

FIG. 11 is a diagram illustrating an example of a data structure of a keyword information table in the embodiment device.

FIG. 12 is a diagram illustrating an example of a data structure of an initial mail folder in the embodiment device.

FIG. 13 is a diagram illustrating an example of a data structure of a deviation value information table in the example device.

FIG. 14 is a diagram illustrating an example of a data structure of a high-priority mail folder in the embodiment device.

FIG. 15 is a diagram illustrating an example of a data structure of a medium-priority mail folder in the embodiment device.

FIG. 16 is a diagram illustrating an example of a data structure of a low-priority mail folder in the embodiment device.

FIG. 17 is a diagram showing an example of screen display to a mail user by a conventional method.

[Description of Signs] 1 Mail transmitting / receiving unit 2 Priority sorting unit 3 Display unit 4 Keyword extraction unit 5 Initial setting information table 6 Initial mail folder 7 Priority mail folder 8 Keyword information table 9 Deviation value information table

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Jin Ikeda 3-9-118 Shin-Yokohama, Kohoku-ku, Yokohama, Kanagawa Prefecture Inside Fujitsu Communication Systems Limited (72) Inventor Hidefumi Shiozawa 3 Shin-Yokohama, Kohoku-ku, Yokohama, Kanagawa No. 9-18, Fujitsu Communication Systems Limited

Claims (3)

[Claims] 1. A method of calculating a reference time per unit character based on a total reference time and a number of characters of a received mail, comparing the calculated reference time with statistical data related to reference times of a plurality of previously received mails, and ranking the mail. Means for assigning ranks, keyword information table means for recording rank-specific points for various keywords, and, for keywords extracted from received mail, the rank-specific points of the corresponding keywords in the keyword information table means for the mail. Keyword information updating means for updating according to the rank attached to the result of extracting a plurality of keywords from the received e-mail, and summing up the points by rank of the corresponding keyword in the keyword information table means corresponding to the keywords for each rank Based on the content of the received mail Email evaluation device provided with an evaluation means for evaluating. 2. The method according to claim 1, wherein the number of the start of the statistical data calculation is set in advance, and the received mail is referred to until the number of the received mail reaches the number of the start of the statistical data calculation. 2. The e-mail evaluation device according to claim 1, wherein the e-mail evaluation device is configured to accumulate together with related data, and when the number of received e-mails reaches the number of start of statistical data calculation, calculate the statistical data based on the received e-mails accumulated up to that time. 3. When the received mail is referred to, when the received mail is referred to, the reference time per character is calculated from the total reference time and the number of characters of the mail body which are referred to by the mail receiver for each mail. 3. The system according to claim 2, wherein a deviation value for each of the referred mails is calculated based on a standard deviation calculated based on a reference time per character of the mail, and a priority for each mail is estimated based on the calculated deviation value for each mail. The described e-mail evaluation device.