WO2010135861A1

WO2010135861A1 - Mail system, junk mail processor and method for marking junk mails

Info

Publication number: WO2010135861A1
Application number: PCT/CN2009/071969
Authority: WO
Inventors: 赵克锋; 石光
Original assignee: Chiao Hakfung; Shi Guang
Priority date: 2009-05-25
Filing date: 2009-05-25
Publication date: 2010-12-02
Also published as: CN102067534A; CN102067534B

Abstract

A mail system, a junk mail processor and a method for marking junk mails are provided. The mail system includes: a mail server, a mail client and a junk mail processor. The mail server functions as a mail process system and processes mails received and sent by the mail client. The mail client is used for sending and receiving mails, interacting with the junk mail processor and providing the junk mail processor with information on whether a mail is a junk mail. The junk mail processor interacts with the mail client, and marks the mail confirmed by the mail client as a junk mail according to the information received from the mail client.

Description

Mail system, spam processor, and method of marking spam

The present invention relates to a mail system and a method for processing mail, and more particularly to a mail system, a spam processor, and a method for marking spam. Background technique

At present, the application of e-mail has become very common. At the same time, e-mail has brought convenience and speed to people, and there have been corresponding problems: People will receive many e-mails that are useless to themselves, such as e-mails for selling products. Moreover, most spam messages are difficult to determine on the subject whether they are the ones the user wants to receive. When a user opens a mailbox, the full mailbox is such a message, and those messages that are considered important are drowned in these useless emails, usually open one by one to determine whether the email is useful to the user, so it takes time and effort to find it. . Therefore, spam has caused great troubles for people's normal life and work.

To this end, people take various measures to deal with spam. For example, the patent document with the publication number CN1744113A discloses a method of filtering spam using the "CC" function. This method requires the creation of a separate mail server (such as www.some.com) that provides the same user with two email addresses of the same domain name, such as: a@ some.com and b @ some.com. The recipient needs to tell the reliable address to both addresses at the same time. The contact fills in the "send to" column with &@some.com when sending the email, and fills in the "cc" column b @ some .com. In this way, the message will be sent to the recipient's "important mail box". For mail senders who generally don't know 1) @ some.com, their mail will be sent to the recipient. But it only appears in the "General Inbox".

However, it is very inconvenient to use two email addresses with the same domain name, and it does not filter spam. The recipient will still receive spam.

In addition, the following patents identify spam by statistically calculating the keyword or the keyword of the content, so that the statistical result may not be applicable to each user. For example, a patent document with the publication number CN101136874 and the application number 200710029369.5 A method and system for preventing spam mis-filtering based on comprehensive decision-making is disclosed, in which a scoring system is disclosed, by which spam is defined.

The patent document with the publication number CN101227435 and the application number of 200810059602.9 discloses a Chinese spam filtering method based on Logistic regression, which extracts the message header, the subject of the email and the related information of the attachment, and classifies the content, and counts the terms in the email. The word frequency, and then the logistic model is established to determine the probability value of spam.

The patent document with the publication number CN1889108 and the application number of 200510081627.5 discloses a method for identifying spam, and also by segmenting the content of the mail and the subject of the mail, and by learning a large number of sample mails, using the Bayesian formula to calculate a new one. The probability that a message is spam.

A method and system for determining spam is disclosed in the patent document with the publication number CN101166159 and the application number 200610152807.2. By helping the user to determine the spam keyword and the keyword score reasonably, and determining whether the information is spam based on the keyword and the corresponding score. The patent can be widely used in a variety of different applications and systems, such as feedback systems, message systems, forums, spam processing and more.

Each of the above methods has advantages and disadvantages, and the common disadvantage is that the system is proactive in filtering mail. For example, most anti-spam methods and systems identify an email as spam without confirmation by the recipient, and whether a email is spam is generally different for different users. For example, for those who want to collect information, the mail that promotes the product is not spam, but for those who do not have the information needs, it is spam, so, a confirmation is made without confirmation from the recipient. Mail is spam, not very reasonable, and direct filtering will increase the possibility of missing information. Summary of the invention

In view of the above deficiencies of the prior art, the present invention provides a mail system and a method for marking spam, which are used to accurately determine whether a mail is spam, and mark the spam for easy viewing by people.

To solve the above technical problem, the present invention provides a mail system, including a mail server, a mail client, and a spam processor; A mail system, comprising a mail server and a mail client, characterized in that it further comprises a spam processor;

The mail server processes the mail sent and received by the mail client as a mail processing system;

The mail client is configured to send and receive mail, and interact with the spam processor to provide the spam processor with information about whether the mail is spam;

The spam processor interacts with the mail client and tags spam confirmed by the mail client based on information obtained from the mail client.

In addition, the present invention also provides a spam processor, including a receiving unit, an information extracting unit, an analyzing processing unit, and a transmitting unit; the information extracting unit extracts information from the mail received by the receiving unit, and extracts the extracted information. The information is sent to the analysis processing unit; the analysis processing unit analyzes the extracted information, confirms whether the mail is spam, and sends the mark information representing the spam to the mail client through the sending unit, so that the mail client inbox The message in the message carries the tag information described.

Furthermore, the present invention also provides a method for marking spam based on the above mail system, comprising the following steps:

Step 1. When the sender sends the mail through the mail server through the mail client, the sender sends the mail to the spam processor;

Step 2: After receiving the email, the spam processor sends an inspection email to the sender's mail client according to the sender's address;

Step 3: The spam processor checks whether the information returned from the sender's mail client has a garbage mark, and if the information has a garbage mark, sends a mark to the recipient's mail client. Spam message, go to step 4; if the message does not contain junk mark or, within the specified time does not receive the information returned from the sender's mail client, go to step 5;

Step ₄ : The recipient's mail client marks the corresponding mail in the inbox as spam according to the information of the marked spam, and proceeds to step 8;

Step 5: The spam processor opens the review permission to the recipient's mail client;

Step 6, the recipient's mail client confirms whether the mail is spam, And sending confirmation information to the spam processor;

Step 7, after receiving the information, the spam processor sends a message marking the spam to the recipient's mail client, and returns to step 4;

Step 8. The review ends.

The advantage of the present invention is that, firstly, the spam is marked only without being deleted or filtered, so that the recipient's right to process the mail is fully guaranteed; secondly, when marking the spam, it is based on The information given by the client to mark the spam greatly improves the accuracy of the mark. Thirdly, according to the system and method of the present invention, when the recipient judges the spam, it can refer not only to whether the message is marked. You can also refer to the factors that introduce "time", such as the time of review, system review time, etc. Recipients can also make different levels of spam judgment based on the length of these times. DRAWINGS

1 is a structural block diagram of a mail system according to the present invention;

2 is a schematic diagram of an interface for setting attributes by a user in an embodiment;

Figure 3 is a write mail interface in an embodiment;

Figure 4 is an inbox interface in an embodiment;

FIG. 5 is a schematic diagram of an interface for opening a received email in an embodiment; FIG.

Figure 6 is an inspection box interface in an embodiment;

Figure 7 is a block diagram of a spam processor;

8 is a flow chart of a method for marking spam in an embodiment of the present invention; and FIG. 9 is another flow chart of a method for marking spam in an embodiment of the present invention. detailed description

As shown in FIG. 1, the present invention provides a mail system, including a mail server 1, a mail client 21, 22, and a spam processor 3;

The mail server 1 processes the mails sent and received by the mail client 21, 22 as a mail processing system;

The mail client 21, 22 is configured to send and receive mail, and interact with the spam processor 3 to provide the spam processor 3 with information on whether the mail is spam; The spam processor 3 interacts with the mail clients 21, 22 and tags the spam confirmed by the mail client 21, 22 based on the information obtained from the mail clients 21, 22.

About the mail client, including it includes writing mail, outbox, inbox, and check box. And includes attributes set by the user, as shown in FIG. 2, the attributes include:

1. Whether the sent mail passes the inspection of the spam processor (for example, the name of the spam processor is buffer);

2. Review the mail sent by the mail client to determine if it is the time of review of the spam

3. The waiting time for the mail client to review the received mail.

More specifically, regarding writing a mail, as shown in FIG. 3, the written mail attributes include: "sender", "recipient", "whether using Buffer", "accepting review time", "title", and "text" " Wait.

Regarding "whether or not to use Buffer", it is used to determine whether the mail has passed the spam processor 3 check. If checked, the sender will be sent to the spam processor 3 for verification when sending the message.

Also included in this option is the "Accept for review time" setting, which can be the value set by the user in Figure 2, or it can be reset by the sender for the specific situation, when the sender has not reset The value here defaults to the value set by the user.

The other items are the same as those written in the existing mail, and will not be described again.

Regarding the inbox, as shown in Figure 4, the attributes of each message include: "Sender", "Recipient" "Whether Buffer", "Junk Mark", "System Review Time", "Voting", "Title" and "Receive time".

Among them, the "garbage mark" is marked in two cases, one is determined by the spam processor 3 based on the confirmation information sent by the mail only; the other case is determined by the spam processor 3 according to " The vote "marked after reaching a certain threshold.

"Voting" means that the message is the ratio of the number of people who mark spam to the total number of recipients. This is a dynamic display, that is, once someone marks the message as spam, the value on the numerator changes.

The system review time (ie the buffer time) is the review of the spam processor Time, in general, this time can be set by the administrator at the spam processor. The system review time is calculated starting from the first time the spam processor receives the mail sent by the mail client. If the time is up, or the time when the mail is subject to review or the time for review is up, that is, if any of the three times arrives, the mail can no longer be reviewed, that is, the garbage mark function of the mail will no longer be Use, that is, the review ends.

About the Outbox, the properties of each message in the Outbox include: "From", "Recipient",

"Whether to use Buffer", "Accept for review time", "Title", "Body". Among them, it should be noted that "whether using Buffer" and "accepting review time" mark this message using the mode checked by the spam processor 3.

About the check box, as shown in Figure 6. The properties of each message in the checkbox include: "From", "Recipient", "Garbage Tag", "Title", "Body", "Send Time", "Time to Arrival (Return Time)" . Among them, when the user checks that a message is very suspicious, the "junk mark" in the message can be set as a garbage mark.

Regarding the spam processor 3, as shown in Fig. 7, a receiving unit, an information extracting unit, an analyzing processing unit, and a transmitting unit are included.

The receiving unit is configured to receive an email sent by the sender;

The information extracting unit extracts information from the mail received by the receiving unit, and sends the extracted information to the analysis processing unit; the information includes:

The recipient list of the message, the system review time for the message, feedback information with the sender, information obtained from the recipient, and so on.

The analysis processing unit analyzes the extracted information, confirms whether the mail is spam, and sends the mark information representing the spam to the mail client through the sending unit, so that the mail in the mail client inbox carries the mail Tag information.

The present invention provides a method for marking spam based on the above mail system. As shown in FIG. 8 and FIG. 9, the method includes the following steps:

Step 1: When the sender sends the mail through the mail server through the mail client, the mail is sent to the spam processor; the implementation manner can be set in the user setting of the mail client, select the buffer check, and set the accept The review time can also be set by checking the "Use Buffer" and time setting options in the write mail properties. In It should be noted that, usually, after the user sets, the attribute of each letter complies with this attribute. Once the letter does not want to comply with the attribute, it can be reset in the mail of the mail. The above two attributes in the written mail are generally the contents of the default user settings, but if reset here, the mail complies with the attribute contents set here;

Step 2: After receiving the email, the spam processor sends an inspection email to the sender's email client according to the sender's address, to verify whether the sender's address is a forged address, and whether it is true. ;

Step 3: The spam processor checks whether the information returned from the sender's mail client includes a garbage mark, and if the information includes a garbage mark, sends the mark garbage to the recipient's mail client. Mail information; if the spam processor does not receive the information returned from the sender's mail client within the specified time or there is no garbage mark in the message, then go to step 5;

Step _4, according to the recipient's mail client information of the tag spam, messages in the Inbox corresponding marked as spam, go to step 8;

Step 6: At this time, the recipient's mail client can confirm whether the email is spam, and send the confirmation information to the spam processor. The specific implementation manner is as shown in FIG. 5, after the email is opened. In the lower right corner, there is a "Report Spam". This item is optional when the spam processor opens the review permission to this mail client. If the sender reports spam, that is, when the message has been marked, This item is not optional. By selecting this item, the confirmation message is sent to the garbage processor;

Step 7, after receiving the information, the spam processor sends a message marking the spam to the email client of the recipient if there is only one recipient, and returns to step 4;

Step 8. The review ends.

If the recipient is multiple, including step 7 also includes:

Step 71: The spam processor receives information confirmed as spam from the mail client as the recipient, and sends the message to all the mail clients as recipients. send Message;

Step 72: The number of "number of people marking the mail as spam" of the mail in the mail client as the recipient is dynamically increased according to the information received from the spam processor;

Step 73: The spam processor determines whether a ratio of the number of people marking the email as spam to the total number of recipients reaches a predetermined threshold. If yes, sending a spam to each recipient's mail client. Information, then return to step 4.

In step 73, if the mail client where the recipient is located sets a threshold for the ratio of the number of people who mark the mail as spam to the total number of recipients, the spam processor determines that the mail is marked as spam. Whether the ratio of the number of people to the total number of recipients reaches the threshold set by the mail client, such as 50%. If it is reached, only the information marking the spam is sent to the mail client, and then returns to step 4.

If the spam processor does not receive the confirmation message from the recipient's mail client, the recipient is deemed to have confirmed that the message is not spam.

In step 1, users can send emails through dedicated mail delivery software, or they can use web pages. The way to send it can be done by selecting "Use Buffer" in the write message properties, or by setting the settings uniformly.

Of course, you can also enter the address of the spam processor directly on the CC address, treat the spam processor as a recipient without having to select "Use Buffer" in the message properties.

In step 2, when the spam processor sends a check mail to the sender's mail client, the check mail is placed in the check box by the mail client. If the user sets a junk tag on the "junk mark" of the message, the mail client automatically sends the message with the junk tag information to the spam processor.

In step 3, the information returned from the sender's mail client may be a message, which may be an untranslated machine code message, or a translated text message, if it is a machine code message, You can query the bit of a byte in the middle of the message. For example, when the bit is set to 0, it is considered that the sender considers the message to be spam. If the bit is set to 1, it means the sender thinks This message is not spam.

In step 3, the spam processor checks the mailer from the sender When the information returned by the client indicates whether the current time has exceeded the time for review or the system review time set in the email, if one has been exceeded, go to step 8.

In step 6, when the recipient's mail client confirms whether the mail is spam, whether the current time has exceeded the waiting time for checking the mail, the time when the mail is subject to review, or the system review time, if any one is exceeded , then do not send any information to the spam processor, go to step 8.

The user can confirm whether the email is spam based on the content of the email, or determine the probability of spam based on whether the sender uses the buffer, the time of the review set by the sender, and the system review time.

This is because the sender of spam generally does not want his mail to be censored, so he generally does not choose to use the buffer for review. For the recipient, the sender is divided into two categories, one is used. Buffer, — class is not using buffer. The chances of using a buffered message for spam will be very small, and the chance of using a buffered message for spam will be very large.

In addition, if the sender of the spam does not want to be reviewed even if the buffer is used, the implementation method that does not want to be reviewed can set the time for review to be short, and when the recipient receives the time is short. In the case of a mail, it can also be considered that the chance of the mail being spam will be large, which makes the recipient more vigilant.

In addition, the spam tag may be selected, such as by ticking in a small box or by color.

The invention does not directly delete or filter spam, but only marks it for people to confirm, and avoids the negative impact of other spam processing methods on different users due to direct deletion or filtering.

Claims

Rights request:

1 . A mail system, comprising a mail server and a mail client, further comprising a spam processor; wherein

The mail server processes the mail sent and received by the mail client as part of the mail processing system;

The mail system according to claim 1, wherein the mail client includes an attribute set by a user, and the attribute includes: whether the sent mail passes the check of the spam processor.

3. The mail system according to claim 2, wherein the attribute further comprises: setting a time for accepting the review of the sent mail.

4. The mail system of claim 2, wherein the attribute further comprises: setting a time to wait for the review of the received mail.

5. The mail system according to claim 1, wherein the mail client attribute of the mail client includes an option of whether or not the spam processor has checked.

6. The mail system according to claim 1, wherein the mail client attribute of the mail client includes a time for which the review of the mail is subject to review.

7. The mail system according to claim 1, wherein the mail client comprises an outbox, an inbox, and a check box, and an attribute of each mail in the outbox and the inbox includes whether After the spam processor's inspection and garbage mark.

8. The mail system according to claim 7, wherein the attributes of each mail in the outbox and the inbox include when the mail is subject to review.

9. The mail system according to claim 7 or 8, wherein the attribute of each mail in the inbox includes a time at which the mail is waiting for review by the mail client where the recipient is located.

10. The mail system according to claim 7 or 8, wherein the attribute of each mail in the inbox includes a ratio of the number of people marking the mail as spam to the total number of recipients.

11. The mail system according to claim 7, wherein the attribute of each mail in the check box includes an option to set a garbage mark.

12. The mail system according to claim 7 or 11, wherein the attribute of each mail in the check box includes whether to automatically reply.

The mail system according to any one of claims 1 to 8, wherein the spam processor comprises a receiving unit, an information extracting unit, an analyzing processing unit, and a transmitting unit, and the information extracting unit receives the receiving The information received by the unit extracts information, and the extracted information is sent to the analysis processing unit, and the analysis processing unit analyzes the extracted information, confirms whether the mail is spam, and sends the mark information representing the spam through the sending unit. Go to the mail client and have the message in the mail client's inbox with the tag information.

14. The mail system according to claim 13, wherein the analysis processing unit in the spam processor detects, when analyzing the extracted information, whether the review time in the mail exceeds the The system review time set in the spam processor, if the system review time is exceeded, the spam is not sent to the mail client. Tag information.

A spam processor, comprising: a receiving unit, an information extracting unit, an analyzing processing unit, and a transmitting unit; the information extracting unit extracts information from the mail received by the receiving unit, and extracts the extracted information Sending to the analysis processing unit; the analysis processing unit analyzes the extracted information, confirms whether the mail is spam, and sends the mark information representing the spam to the mail client through the sending unit, so that the mail client is in the inbox Mail with the tag information described.

The spam processor according to claim 15, wherein the information extracting unit extracts an address of the departure person, a address of the recipient, and a time when the mail is subject to review from the mail received by the receiving unit. And the time the message is waiting for review at the recipient's mail client.

The spam processor according to claim 15 or 16, wherein the information extracting unit receives feedback from a sender and/or a recipient sent by the receiving unit

I from,

The spam processor according to claim 17, wherein the analysis processing unit receives the garbage of the mail in the feedback information of the sender or the recipient sent by the information extracting unit The flag confirms that the message is spam.

19. A method of marking spam based on a mail system according to any of claims 1-14, comprising the steps of:

Step 3: The spam processor checks whether the information returned from the sender's mail client has a garbage mark, and if the information has a garbage mark, The person's mail client sends the message marking the spam, go to step 4; if the message does not contain the garbage mark, or does not receive the information returned from the sender's mail client within the specified time, turn Go to step 6;

Step 6, the recipient's mail client confirms whether the mail is spam, and sends confirmation information to the spam processor;

Step 8. The review ends.

The method for marking spam according to claim 19, wherein if the number of recipients is plural, step 7 further includes:

Step 71: The spam processor sends information to all mail clients that are recipients by receiving information confirmed as spam from the mail client as the recipient;

The method for marking spam according to claim 20, wherein in step 73, if the mail client where the recipient is located sets the number of people who mark the mail as spam and the total number of recipients The threshold of the ratio, the spam processor determines whether the ratio of the number of people marking the mail as spam to the total number of recipients reaches a threshold set by the mail client, and if so, sends only to the mail client Mark spam Information, then return to step 4.

22. The method of marking spam according to claim 21, wherein in step 7, if the spam processor does not receive the confirmation message sent by the recipient's mail client, The recipient confirms that the message is not spam.

23. The method of marking spam according to claim 20, wherein said prescribed time is set at a spam processor.

24. The method of marking spam according to claim 21, wherein said threshold value is set in the spam processor and in the mail client, respectively, and is prioritized by the setting of the mail client.

25. The method of marking spam according to claim 19, wherein in step 3, the spam processor checks the current time when checking information returned from the sender's mail client. Whether the time for review or the system review time set in the message has been exceeded, and if one has been exceeded, go to step 8.

26. The method of marking spam according to claim 20, wherein, in step 6, the recipient's mail client confirms whether the mail is spam, and checks whether the current time has exceeded the mail waiting The time of the review, the time the review was accepted for review, or the time of the system review. If any of the messages is exceeded, no information is sent to the spam processor. Go to step 8.