JP2006245813A - Filtering system, filter type creation engine, method of filtering and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a filtering system, a filter type creation engine, a method of filtering and a program which can create a filter type rule which is adapted for each as opposed to a new troublesome mail coming out with changes of time or a distinct troublesome mail according to an individual taste. <P>SOLUTION: The filter type creation engine 106 classifies the troublesome mail of an individual mail acceptor 100 into a receiving rejection mail BOX103, analyzes about the feature of unwanted mail for the user from the content of the mail of the BOX, and sets a filter type rule 107. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a filtering system, a filter creation engine, a filtering method, and a program based on characteristic analysis of spam mails rejected by a user.

  General mail software provides a mail filtering function based on keywords.

  An example of a filtering system that reduces the management burden by sharing filter rules between groups is described in Patent Document 1. This filtering system automatically applies the filter rule to all groups managed as a directory when the same filter rule is registered exceeding a certain threshold.

Here, each firewall device performs packet filtering in accordance with a predetermined filtering rule, and further monitors a packet that can pass through the current filtering rule, and if it determines that the packet should be rejected, the header of the packet The transmission source IP address is extracted from the unit, the filtering rule is updated based on the extracted information, and the transmission source IP address is recorded and periodically transmitted to the maintenance center server. Information sent from each device is periodically statistically processed, and rejected IP addresses, etc. that are judged to be shared by a plurality of devices are sent to each device. Each device receives information sent from the server. Based on the above, update the own filtering rules, etc. Technology that can provide a Rutaringu type firewall systems have been proposed (e.g., see Patent Document 2).
JP 2004-021623 A JP 2003-273936 A

  However, the conventional example described above has the following problems.

  This is achieved by the user setting the sender or keyword as a filter target, but avoiding the filter by spoofing the sender or deliberately changing the keyword (eg, mall is m @ ll). Once filtered, filtering did not work effectively.

  In addition, the filtering function can be used effectively because the user himself / herself has to manually set the filter action rule, and it is necessary to set the filter action rule every time a new junk mail is generated. Was not.

  In addition, junk mail may be filtered by Internet service providers and telecommunications carriers. However, when blocking mail from senders that send a large amount of mail or mail containing specific keywords, the filtering operation is performed with the intention of the above-mentioned vendor. Since it is set, there is a possibility that the mail desired by the user is blocked. In addition, because of concern about this, a contractor may not be able to set an effective filter operation.

  Therefore, the present invention can create a filter rule adapted to each new junk e-mail that appears as the times change or junk e-mail that varies according to individual preferences. It is an object to provide a filtering system, a filter creation engine, a filtering method, and a program.

  The invention described in claim 1 is a filtering system including a personal mail receiving terminal and a filter creation engine. The operation means classifies the junk mail into a reception refusal mail BOX, and the contents of the mail of the reception refusal mail BOX. The filtering system is characterized in that a filter creation engine analyzes characteristics of mail unnecessary for a user and sets a filter rule.

  According to a second aspect of the present invention, in the filtering system according to the first aspect, the filter creation engine is newly added to the mail BOX and means for periodically referring to the update of the reception refusal mail BOX and / or the reception essential mail BOX. Means for decomposing the text of the received email and the sender's lexical by parsing, means for calculating the number and percentage of emails that contain the lexical in the email BOX, a preset threshold and the number of emails, And a means for comparing the ratios and adding a lexical phrase satisfying the condition as a new rule to the filter rule.

  The invention described in claim 3 further includes a wide area filter creation device, analyzes the characteristics of the reception refusal mail BOX of each personal mail receiving terminal, creates a wide area filter rule, and uses this as a filter rule for each personal mail receiving terminal. The filtering system according to claim 1, wherein the filtering system is added.

  According to a fourth aspect of the present invention, there is provided a filter creation engine characterized in that a mail rule unnecessary for a user is analyzed from a mail content of a reception refusal mail BOX of a personal mail receiving terminal, and a filter rule is set.

  According to a fifth aspect of the present invention, in the filter creation engine according to the fourth aspect, means for periodically referring to the update of the reception refusal mail BOX and / or the mandatory reception mail BOX of the personal mail receiving terminal, and a new mail BOX Means for disassembling the added mail text and sender's lexical by parsing, means for calculating the number and ratio of mails that contain the lexical in the mail BOX, preset threshold and number of mails And a means for comparing the ratios and adding a lexical phrase satisfying the condition as a new rule to the filter rule.

  The invention described in claim 6 is a filtering method using a personal mail receiving terminal and a filter creation engine. The step of classifying junk mail into a reception refusal mail BOX, and the content of the mail of the reception refusal mail BOX The filtering method is characterized in that the filter creation engine analyzes the characteristics of the mail unnecessary for the user and sets a filter rule.

  According to a seventh aspect of the present invention, in the filtering method according to the sixth aspect, the filter creation engine periodically refers to the update of the reception refusal mail BOX and / or the reception essential mail BOX, and is newly added to the mail BOX. Disassembling the text of the received email and the sender's lexical by parsing, calculating the number and percentage of emails that contain the lexical in the email BOX, a preset threshold and the number of emails, And comparing the ratios and adding a lexical phrase satisfying the condition to the filter rule as a new rule.

  The invention according to claim 8 is the filtering method according to claim 6 or 7, wherein the wide area filter creation device analyzes the characteristics of the reception rejection mail BOX of each personal mail receiving terminal and creates a wide area filter rule. Is added to the filter rule of each personal mail receiving terminal.

  According to the ninth aspect of the present invention, the personal mail receiving terminal classifies the junk mail into the reception refusal mail BOX, and the filter creation engine provides the mail characteristics unnecessary for the user from the contents of the mail of the reception refusal mail BOX. It is a program characterized by causing a step of analyzing and setting a filter rule to be executed.

  The invention described in claim 10 is the program according to claim 9, which is newly added to the mail BOX, a step of periodically referring to the filter creation engine for the update of the reception refusal mail BOX and / or the reception essential mail BOX. Disassembling the body of the mail and the sender's lexical by syntactic analysis, calculating the number and ratio of mails that contain the lexical in the mail BOX, a preset threshold and the number and ratio of the mail And a step of adding a lexical phrase satisfying the condition to the filter rule as a new rule is executed.

  The invention according to claim 11 is the program according to claim 9 or 10, further causing the wide area filter creation device to analyze the characteristics of the reception refusal mail BOX of each personal mail receiving terminal and create a wide area filter rule. The step of adding to the filter rule of each personal mail receiving terminal is executed.

  According to the filtering system, the filter creation engine, the filtering method, and the program of the present invention, a new junk e-mail comes out with the change of the times, or a junk e-mail that varies according to individual preference, Filter rules adapted to each can be created.

  Next, an embodiment of the present invention will be described in detail with reference to the drawings.

  Referring to FIG. 1, the filtering system according to the present embodiment includes a personal mail receiving terminal (computer, mobile phone, etc.) 100 for each user, and a wide area filter creation device 110 by an Internet service provider, a communication company, or the like. .

  A personal mail receiving terminal (computer, mobile phone, etc.) 100 includes a mail receiving I / F unit 101, a received mail BOX 102, a reception refusal mail BOX 103, a reception pending mail BOX 104, a reception mandatory mail BOX 105, a filter creation engine 106, a filter rule 107, A storage folder 108 and a trash box 109 are included.

  The wide area filter creation apparatus 110 includes a wide area filter creation engine 111 and a received mail sample 112.

  Each of these means generally has the following functions.

  The mail reception I / F unit 101 receives a mail from the Internet service provider, and classifies the mail into a received mail BOX 102 and a reception pending mail BOX 104 with reference to the filter rule 107.

  The received mail BOX 102 stores the received mail, copies the mail selected by the user to the storage folder 108, and moves the mail classified as spam by the user in the GUI shown in FIG. 4 to the reception refusal mail BOX.

  The reception refusal mail BOX 103 stores the junk mail moved from the reception mail BOX 102.

  The reception hold mail BOX 104 stores mail classified as spam mail by the mail reception I / F 101 and the filter rule 102.

  The reception-required mail BOX 105 stores a mail that has been determined to be received by the user, out of the mails of the reception pending mail BOX.

  The filter creation engine 106 creates a filter rule 107 based on the received mail BOX 102, the received rejection mail BOX 103, and the mandatory reception mail BOX 105. A specific functional block diagram is shown in FIG.

  The wide area filter creation engine 111 creates a filter rule with reference to the reception refusal mail BOX 103 and the received mail BOX 102 of each user's personal mail receiving terminal 100 and adds it to the filter rule 107 of each user.

  The range of the target personal mail receiving terminal 100 is set according to the situation such as all mail recipients of the Internet service provider and a list of trusted parties.

  The received mail sample 102 randomly extracts and stores some mails from the personal mail receiving terminal 100 of each user.

  Next, the operation of the filtering process in the present embodiment will be described in detail with reference to the flowcharts of FIGS.

  First, the mail reception I / F unit 101 receives a mail (step A1 in FIG. 2).

  Next, referring to the filter rule 107 (step A2), the mail is classified into a received mail BOX 102 (step A3) and a reception pending mail BOX 104 (step A4).

  Separately from this, the filter creation engine 106 periodically refers to the update of the reception refusal mail BOX 103 and the reception mandatory mail BOX 105, and creates a rule to be newly added to the filter rule 107.

  First, the body of the mail newly added to the reception refusal mail BOX 103 and the sender's lexical are disassembled by syntax analysis (step B1), and the number of mails including the lexical in each of the BOXs 102, 103, and 105, And the ratio is calculated (step B2).

  Next, the preset threshold value is compared with the number of mails and the ratio (step B3), and if there is a lexical word satisfying the condition, it is added to the filter rule 107 as a new rule.

  The above flow is repeated for all the words (step B4).

  Examples of threshold values include: (1) included in X mails of the rejected mail box 103, (2) not included in Y% or more of the mails of the received mail box 102, and (3) Z of the mandatory mail box 105 Set it as not included in%.

  The threshold value may be appropriately changed according to the condition of the filter rule or the designation of each user.

  Similarly, the wide-area filter creation engine 111 periodically refers to each user's reception refusal mail BOX 103 (step C1), searches the received mail sample 112 for the included lexical phrase (step C2), and compares it with a threshold (step C3). ).

  The words that satisfy the condition are added as new rules to the filter rule 107 of each user (step C4).

  In parallel, a sample mail is randomly extracted from the received mail BOX 102 of each user, and the received mail sample 112 is updated.

  The user sees the mail in the received mail BOX 102, copies the necessary mail to the storage folder 108, and moves the mail to be classified as spam mail to the reception refusal mail BOX 103 by specifying it in the menu shown in FIG. 4, for example. Let

  Further, if necessary, the contents of the reception pending mail BOX 104 are confirmed, and if there is a mail to be received, it is moved to the reception essential mail BOX 105 by designating it.

  According to the above embodiment, the filter rule can be easily created. The reason is that a filter rule is automatically created by a minimum user operation equivalent to deleting a mail, which designates the mail as a spam mail.

  Moreover, it can respond also to the camouflage of a transmission origin. The reason is that since the filtering rule is set based on the contents of the spam mail, the spam mail can be filtered without specifying the sender.

  In addition, it is possible to provide an environment that can quickly prevent the spread of junk mails with the cooperation of each user. The reason is that the common points of the nuisance rules classified by each user are analyzed and added to the filter rule, thereby preventing the subsequent nuisance rules from spreading.

  In addition, since it is configured to compare the lexical text contained in the mail classified as spam mail (reception refusal mail BOX 103) and normal mail (received mail 102), the lexical text often included in spam mail is automatically Can be added to filter rules.

  Moreover, since it is comprised so that the common point of the mail which each user classify | categorized as the junk mail may be analyzed, the mail generally regarded as junk can be automatically filtered.

  Also, depending on the user, there is a case where a mail that is not desired to be classified as a junk mail is stored in the reception pending mail BOX 104. By preparing means for moving the mail to the reception essential mail BOX 105, the mail is picked up.

  Furthermore, since the mail is explicitly determined to be necessary by the user's operation, the filter rule can be changed so that similar mail will not be classified as spam mail in the future.

  Next, the configuration of the second exemplary embodiment of the present invention will be described.

  For voice calls, it has a mechanism for recording calls, and provides an operation means that allows users to clearly indicate that they are nuisance calls. An engine that analyzes voice characteristics (voice print, voice analysis, etc.) instead of lexical analysis of emails. By preparing it, it becomes possible to automatically cope with a rule that warns in advance or does not receive the same annoying call or reports.

  Each of the above-described embodiments is a preferred embodiment of the present invention, and various modifications can be made without departing from the scope of the present invention. For example, the program for realizing the functions of the user terminal 100, the bulletin board 150, the information acquisition apparatus 200, the settlement apparatus 300, and the information server 400 in the above-described embodiment is read and executed by each apparatus or the like. You may perform the process which implement | achieves a function. Further, the program is transmitted to another computer system by a transmission wave via a CD-ROM or a magneto-optical disk that is a computer-readable recording medium, or via the Internet or a telephone line that is a transmission medium. Also good.

  In each of the above-described embodiments, the system configuration in which the user terminal 100, the bulletin board 150, the information acquisition apparatus 200, the settlement apparatus 300, and the information server 400 are separately connected has been described. However, each function is realized as one computer system. Of course, the present invention can also be applied to a configuration in which a plurality of server devices and the like are added for each configuration and function.

  According to the present invention, it can be applied to junk mail filtering and prevention of diffusion. Moreover, if applied to voice calls such as telephones, it can also be applied to uses such as detecting nuisance calls such as “transfer fraud”, calling attention, and reporting.

It is a schematic block diagram of the filtering system in embodiment of this invention. It is a flowchart which shows the operation | movement of the filtering process in embodiment of this invention. It is a functional block diagram of a filter creation engine. It is a figure which shows an example of GUI.

Explanation of symbols

100 Personal Mail Receiving Terminal 101 Mail Receiving I / F Unit 102 Received Mail BOX
103 Receiving refusal mail BOX
104 Receive hold mail BOX
105 Required mail BOX
106 filter creation engine 107 filter rule 108 storage folder 109 trash can 110 wide area filter creation device 111 wide area filter creation engine 112 received mail sample

Claims (11)

A filtering system including a personal mail receiving terminal and a filter creation engine,
An operation means for classifying the junk mail into a reception refusal mail BOX,
A filtering system, wherein the filter creation engine analyzes a mail characteristic unnecessary for a user from the mail contents of the reception refusal mail BOX, and sets a filter rule. The filter creation engine periodically refers to the update of the reception refusal mail BOX and / or the reception mandatory mail BOX;
A means of disassembling the body of the mail newly added to the mail BOX and the lexical of the sender by parsing;
A means of calculating the number and percentage of emails that contain the lexical in the email BOX,
A means for comparing a preset threshold value with the number of mails and the ratio, and adding a lexical condition that satisfies the condition to the filter rule as a new rule;
The filtering system according to claim 1, further comprising:   Further, a wide-area filter creation device is included, and the characteristics of the reception refusal mail BOX of each personal mail receiving terminal are analyzed to create a wide-area filter rule, which is added to the filter rule of each personal mail receiving terminal. The filtering system according to claim 1 or 2.   A filter creation engine that analyzes the characteristics of mail unnecessary for the user from the contents of the mail of the rejection mail BOX of the personal mail receiving terminal and sets a filter rule. Means for periodically referring to the update of the reception refusal mail BOX and / or the reception mandatory mail BOX of the personal mail receiving terminal;
A means of disassembling the body of the mail newly added to the mail BOX and the lexical of the sender by parsing;
A means of calculating the number and percentage of emails that contain the lexical in the email BOX,
A means for comparing a preset threshold value with the number of mails and the ratio, and adding a lexical condition that satisfies the condition to the filter rule as a new rule;
The filter creation engine according to claim 4, further comprising: A filtering method using a personal mail receiving terminal and a filter creation engine,
A step of classifying junk mail as a refusal mail BOX;
A filtering method, comprising: a step of analyzing a characteristic of a mail unnecessary for a user from a mail content of the reception refusal mail BOX, and setting a filter rule. The filter creation engine periodically refers to the update of the rejected email BOX and / or the mandatory email BOX;
Disassembling the mail text and sender's lexical newly added to the mail box by parsing;
Calculating the number and percentage of emails that contain the lexical in the email BOX,
Comparing the threshold value set in advance with the number of mails, the ratio, and adding a lexical condition that satisfies the condition to the filter rule as a new rule;
The filtering method according to claim 6, further comprising:   Further, the wide-area filter creation device has a step of analyzing the characteristics of the reception refusal mail BOX of each personal mail receiving terminal to create a wide-area filter rule and adding it to the filter rule of each personal mail receiving terminal. The filtering method according to claim 6 or 7. Classifying the junk mail as a refusal mail BOX on the personal mail receiving terminal;
A program for causing a filter creation engine to analyze a characteristic of mail unnecessary for a user from a mail content of the reception refusal mail BOX and set a filter rule. Referring to the filter creation engine periodically for the update of the rejected email BOX and / or the mandatory email BOX;
Disassembling the mail text and sender's lexical newly added to the mail box by parsing;
Calculating the number and percentage of emails that contain the lexical in the email BOX,
Comparing the threshold value set in advance with the number of mails, the ratio, and adding a lexical condition that satisfies the condition to the filter rule as a new rule;
10. The program according to claim 9, wherein the program is executed.   Further, the step of causing the wide-area filter creation device to analyze the characteristics of the reception refusal mail BOX of each personal mail receiving terminal to create a wide-area filter rule and add this to the filter rule of each personal mail receiving terminal is executed. The program according to claim 9 or 10, characterized in that

Images