WO2008116424A1

WO2008116424A1 - Method of user classification, and method, device and system of directional advertisement delivery

Info

Publication number: WO2008116424A1
Application number: PCT/CN2008/070606
Authority: WO
Inventors: Zhao Dai; Qi Mo; Ziguang Huang
Original assignee: Tencent Technology (Shenzhen) Company Limited
Priority date: 2007-03-28
Filing date: 2008-03-27
Publication date: 2008-10-02
Also published as: CN101071437A

Abstract

A method of the user classification, and a method, device and system of the directional advertisement delivery are provided. The user classification method includes: determining an interesting information of the user according to the choice information of the user terminal; searching the user classification most matching to the interesting information of the user from the predetermined user classification table; and adding the user into the most matching user classification. The directional advertisement delivery method includes: after the user enters by the corresponding user terminal, searching the user classification to which the user pertains; on the basis of the user classification, delivering the directional advertisement to the user according to the predetermined rule.

Description

TECHNICAL FIELD The present invention relates to the field of Internet technologies, and in particular, to a method for classifying users, a method, an apparatus, and a system for directed advertisement placement. Background of the invention

With the development of Internet technology, the personalized needs of users have been further improved. The information on the Internet, such as online advertisements, is constantly changing the way of playing in the past. The new form of advertisement "narrowing" has gradually become an important part of Internet advertising. Among them, the narrow meaning of the original meaning is "narrow and advertised" or "specially advertised", mainly refers to the user's advertising directly placed around the article on the online media related to its content, while narrowing will also be based on the viewer The preferences, usage habits, geographic location, access history and other information, targeted narrow-cut ads to the viewers who are really interested. It can be seen that the targeted advertisement for the segmented user group can be targeted to the viewers who are really interested in the targeted advertisement according to the preferences, usage habits, geographical location, and access history of the viewer. A new type of advertising technology can also be called targeted advertising technology, and targeted advertising technology can meet the needs of advertisers for advertising their potential user groups.

An existing targeted advertisement delivery technology is as follows: collecting the registration information of the logged-in user, determining the classification of the user according to the collected user information, thereby playing related advertisements to the corresponding user, wherein the registration information is the user. When you register, you can fill in according to your own situation, which can reflect the user's preferences to a certain extent. It can be seen that the existing targeted advertisement delivery technology is a targeted advertisement delivery technology based on registered user information and catering to user interest habits.

The use of the above-mentioned existing targeted advertising technology can achieve the purpose of delivering targeted advertisements to the user group. However, such targeted advertisement delivery technology relies too much on the user's registration information, and At present, some domestic factors such as the real name of the Internet environment are not high, the lag of unnamed users and user data updates, and false user data will affect the accurate segmentation of users, and thus cannot guarantee that advertisements can be delivered to truly interested browsing. In front of them. Summary of the invention

An embodiment of the present invention provides a method and apparatus for classifying users, so that users with the same or similar behaviors are placed in the same user category.

Another aspect of an embodiment of the present invention provides a directed advertisement delivery method, apparatus, and system for placing targeted advertisements in front of a viewer who is truly interested in the targeted advertisement.

The technical solution of the embodiment of the present invention is implemented as follows:

A method of classifying users, including:

Determining the user's interest information according to the selection information of the user terminal;

Finding a user classification that best matches the user's interest information from the preset user classification table;

Adding the user to the best matching user category.

A method for targeted advertising, including:

After the user logs in through the corresponding user, the user classification of the user is searched; and the targeted advertisement is delivered to the user according to the preset rule according to the user classification.

A targeted advertisement delivery system, comprising: a user classification device and a targeted advertisement delivery device; wherein

The user classification device is configured to search for a user classification to which the user belongs after the user logs in through the user terminal;

The targeted advertisement placing device is configured to deliver a corresponding targeted advertisement to the user according to a preset rule according to a user classification to which the user belongs.

A user classification device, comprising: an interest information determination unit, and a user classification matching unit And saving unit;

The interest information determining unit is configured to determine, according to the selection information of the user end, the interest information of the user;

The user classification matching unit is configured to search, from a preset user classification table, a user classification that most closely matches the user's interest information;

The saving unit is configured to join the user to the best-matched user category, and save the user information corresponding to the user.

A targeted advertisement delivery device includes: a user classification determining unit and a targeted advertisement delivery unit; wherein

The user classification determining unit is configured to search for a user classification to which the user belongs after the user logs in through the user end;

The targeted advertisement delivery unit is configured to deliver targeted advertisements to the user according to preset rules according to the user classification.

It can be seen from the above technical solution provided by the embodiment of the present invention that the user is classified by the user terminal selection information, so that users with the same or similar behaviors are put into the same user classification, so that only the user classification policy needs to be set to avoid The policy is set for each user, saving processing resources.

Further, because the embodiment of the present invention delivers targeted advertisements to users for user classification, the preferences of users in the same user classification are the same or similar, and when the users are from the same user classification, the user's preferences are systems. As is known, when a user logs in, the targeted advertisement can be placed on the user according to the user classification to which the user belongs, and the content of the targeted advertisement is pre-set for the user classification, thereby ensuring that the targeted advertisement is delivered to the real In front of the viewers who are interested in the targeted advertisement, the hit rate of targeted advertising is improved. BRIEF DESCRIPTION OF THE DRAWINGS

1 is a flowchart of a first embodiment of a method for classifying a user according to an embodiment of the present invention; FIG. 2 is a flowchart of a third embodiment of a method for classifying a user according to an embodiment of the present invention; FIG. 4 is a flowchart of a second embodiment of a method for placing a targeted advertisement according to an embodiment of the present invention; FIG. 5 is a flowchart of a third embodiment of a method for placing a targeted advertisement according to an embodiment of the present invention; A structural diagram of a targeted advertisement placement system according to an embodiment of the present invention;

7 is a structural diagram of a user classification apparatus according to an embodiment of the present invention;

FIG. 8 is a structural diagram of a directed advertisement placing apparatus according to an embodiment of the present invention. Mode for carrying out the invention

The present invention will be further described in detail below with reference to the accompanying drawings.

The embodiment of the present invention is mainly: determining, according to the selection information of the user end, the interest information of the user corresponding to the user end; searching for the user classification that most closely matches the interest information of the user from the preset user classification table; The best matching user classification. This enables users with the same or similar behavior to be placed in the same user classification, thereby implementing corresponding policies for user classification without having to formulate policies for each user, thereby saving resources and ensuring for each user. The strategies are all effective.

The selection information of the user terminal may be in various forms, such as an advertisement link content triggered by the user end, or a selection content sent by the user end.

Before determining the interest information of the user according to the selection information of the user end, the method may further include: acquiring the selection information of the user end. In this way, the embodiment of the present invention can determine the interest information of the user corresponding to the user end according to the content triggered by the user.

The method for classifying users in the embodiment of the present invention is described in detail below. Referring to FIG. 1 , FIG. 1 is a specific flowchart of a first embodiment of a method for classifying a user according to an embodiment of the present invention. In this embodiment, an advertisement link content triggered by a user terminal is taken as an example. Specifically, the selection information of the client is the link content of the advertisement that the user clicks sent through the client. As shown in FIG. 1 , the process may include the following steps: Step 101: Acquire content of a link that a user clicks, and determine interest information of the user by the content of the link.

Different users are interested in different information, so different users pay different attention to the content on the Internet, so the user's behavior can determine the user's preference to a certain extent, so that after the user clicks on a link, the user can The link content that the user clicks determines the user's interest information, including the user's behavior information and hobby information, etc., wherein the link content clicked by the user may be the link content of the advertisement clicked by the user. For example, if a user clicks on an advertisement for a digital class, the user may be interested in the digital information to a certain extent; thus, the user may be provided with some digital links, such as links to digital advertisements, thereby improving the advertisement. Click rate, improve the hit rate of the advertisement; of course, you can also provide users with some links to digital information, so that users can get the latest information and improve the click rate of the website.

Step 102: Search, from a preset user classification table, a user classification that best matches the difference information of the user classification with the interest information of the user.

In this embodiment, the user classification is usually set in advance, and the user classification table is obtained, and each user classification in the user classification table is guaranteed to have some specific difference information corresponding to the user's interest information. Then, after performing step 101 to determine the user's interest information, a user classification corresponding to the above determined interest information may be found from the preset user classification table. Often, the behavior and love of users in the same user category are the same or similar.

Step 103: Add the user to the best matching user category. It can be seen that, in this embodiment, the user is added to the most-matched user classification, so that users with the same or similar behaviors can be put into the same user classification, and the corresponding policy can be formulated for the user classification, without Each user develops a strategy that saves resources and ensures that each user's policies are effective.

Generally, when the user is in a user classification for a long time without updating, the user's interest information may change. In the embodiment of the present invention, when the user spends too long in a user classification without updating. The user's interest information may change to provide a second embodiment of the method for classifying the user. Compared with the first embodiment of the method for classifying the user, the embodiment further includes the following steps:

If the time of the user in the most matching user category reaches a preset time, the interest information of the user is updated according to a preset rule, and the execution slave preset is returned according to the updated user interest information. The step of finding the user classification that the difference information of the user classification matches the interest information of the user in the user classification table.

It can be seen that when the user is in a user classification for a long time without updating, the user's interest information may change. In this case, the user's interest information may be updated through a preset rule, thereby ensuring the user to which the user belongs. Classification can reflect the user's interest information as much as possible.

Another type of embodiment of the method for classifying users provided by the embodiments of the present invention is to classify users by simulating a bee colony, and to treat a specific user as a bee; and some information inherent to the bee itself, such as gender, Age as an odor; user's behavioral information, hobby information, etc. as honey, usually, honey can also be called nectar; the user is classified as a hive. Therefore, the third embodiment of the method for classifying users according to the embodiment of the present invention is as shown in FIG. 2, which mainly includes:

Step 201, returning nectar.

After returning the nectar, after determining the user information of the user, the user information is passed to the user. The process of the device for processing the user information; wherein the user information mainly includes the behavior information and the hobby information of the user.

Step 202: Put the nectar into a processing queue.

In practical applications, there may be many bees whose nectar needs to be processed, so that a processing queue is needed to buffer the nectar. Because each nectar has no priority, it is processed in a first-in, first-out strategy in practical applications. The nectar in the queue is processed.

Step 203: Obtain a nectar of a bee from the processing queue.

The nectar that first entered the processing queue is obtained from the processing queue according to a first in first out strategy. Of course, the nectar that has been processed will not exist in the processing queue.

Step 204: Select a cell for the nectar, and store the nectar in the hive. Step 205, brewing honey.

In order to reduce the nectar that has been stored for too long, it has a bad influence on the user's current nectar. The nectar with long storage time is processed, so that the nectar that is stored for too long can also reflect the user's current nectar. This process is called honey.

The present invention also provides a more preferred embodiment in which the treatment of the odor is added during the processing of the user classification method described in FIG. 2, wherein the scent, also called breath, is used to indicate the inherent Some information, such as gender, age, etc. Although the odor is not necessarily accurate or timely, it can still reflect the user's preferences to some extent.

FIG. 3 is a first embodiment of a method for placing a targeted advertisement according to an embodiment of the present invention, which mainly includes:

Step 301: After the user logs in through the user end, the user classification of the user belongs to the user classification database.

After the user logs in through the user, the user's identity can be found in the user's classified database according to the user's identity. The user's identity can be the user name, user number, etc. Uniquely identifies the user's information. Step 302: Deliver a targeted advertisement to the user according to the preset rule according to the user classification. Since the preferences of the users in each user category are different, the targeted advertisements are delivered to the users according to the preset rules according to the user classification to which the user belongs. For example, if a user belongs to a user classification of a digital class, when the user logs in through the client, the user is given a digital advertisement, because the user is interested in the digital, so the probability of the targeted advertisement being clicked is relatively large, thereby achieving The purpose of targeted advertising is to achieve the purpose of placing targeted advertising to the viewers who are really interested in the targeted advertising.

Because the user's preferences may change over time, or the user needs to be placed in another, more precise category, such as the user's original classification in the digital category, further discovering that the user is more DV in the digital. Interested, so you can further put the user into the user category of the DV class, so that targeted advertising can be targeted to the user. Therefore, the embodiment of the present invention provides a second embodiment of a targeted advertisement delivery method. After the targeted advertisement is delivered to the user, the embodiment further determines the advertisement that the user really needs currently according to the selection information of the user end corresponding to the user. As shown in Figure 4, the process mainly includes the following steps:

Step 401: Determine a user classification in which the user is currently located.

Step 402: Deliver a targeted advertisement to the user according to the preset rule according to the user classification. Step 403: Determine, according to the selection information of the user end, the interest information of the user.

In this embodiment, the selection information of the client is the advertisement link content of the user clicked by the client.

After the step 402, the user may click on the targeted advertisement delivered to the user, or click on other targeted advertisements or non-targeted advertisements, and the click behavior of the user reflects the interest information of the user, so that the user may click on the orientation. The content of the advertisement or the non-targeted advertisement is classified to correspond to the user's interest information, and the content classification of the targeted advertisement or the non-targeted advertisement is manually processed. If the user clicks on a targeted ad served to that user, then The user classification of the user is still correct. Of course, the user classification can be further refined based on the current user classification. If the user clicks on other targeted advertisements or non-directional advertisements, the user's interest information may be changed. At this time, it may be necessary to change the user classification in which the user is located, or put the user in the user classification of the upper level, thereby ensuring the correctness of the user classification of the user, and ensuring that the targeted advertisement to be served is of interest to the user.

Step 404: Search, from the preset user classification table, a user classification that best matches the difference information of the user classification with the interest information of the user.

Step 405: Add the user to the best matching user category.

In this way, after the user's preference is changed, the user can be placed into the corresponding user category by the targeted advertisement clicked by the user; or the user can be placed into a more refined user category by the targeted advertisement clicked by the user, thereby Targeted ads can be delivered to users more specifically.

Another embodiment of the method for targeted advertisement delivery provided by the present invention is also implemented by simulating a bee colony. As shown in FIG. 5, the method mainly includes:

Step 501: Load the hive information into the shared memory.

The hive is the nest of the bees, and each hive is in the hive.

Step 502: Select a cell corresponding to the bee from the shared memory.

Each bee has its corresponding hive, and there is information about all hive in the shared memory, so you can select the hive corresponding to the bee from the shared memory.

Step 503: Deliver a targeted advertisement according to the obtained hive.

Once the hive is identified, nectar can be obtained so that the targeted advertisement can be targeted to the bee, thereby directing the targeted ad to the viewer who is actually interested in the targeted ad.

FIG. 6 is a structural diagram of a targeted advertisement delivery system according to an embodiment of the present invention, which mainly includes: a user classification device 601 and a targeted advertisement delivery device 602. The user classification device 601 is configured to search for a user classification to which the user belongs after the user logs in through the user.

The targeted advertisement placement device 602 is configured to deliver the corresponding targeted advertisement to the user according to a preset rule according to the user classification to which the user belongs.

The structure diagram of the user classification device 601 can be as shown in FIG. FIG. 7 is a structural diagram of a user classification apparatus according to an embodiment of the present invention. The apparatus mainly includes: an interest information determination unit 701, a user classification matching unit 702, and a storage unit 703.

The interest information determining unit 701 is configured to determine the user's interest information according to the selection information of the user terminal.

In this embodiment, the selection information of the user end is the advertisement link content of the user clicked by the user terminal, and the interest information determining unit 701 determines the interest information of the user according to the content of the advertisement link of the click of the user. Since the content of the link is known, the content of the user can be determined by the content of the link. For example, if the user clicks on the link of the sports class, the user can be interested in the information of the sports class to a certain extent.

The user classification matching unit 702 is configured to search for a user classification that best matches the user's interest information from the preset user classification table.

The user classification is preset. Each user classification has different information. When the user's interest information matches the difference information of a user classification, the user can be considered as belonging to the user classification.

The saving unit 703 is configured to add the user to the best-matched user category, and save the user information corresponding to the user sent by the client.

The apparatus can further include: an acquisition unit 704.

The obtaining unit 704 is configured to acquire selection information of the UE. In this embodiment, the selection information of the client is the content of the advertisement link that the user clicks through the user terminal, and the obtaining unit 704 is configured to acquire the link content of the advertisement that the user clicks. The content of each link is known So, you can add a database or other way to record the content of each link. When the user clicks on a link, you can immediately know the content of the clicked link.

It can be seen that, by using the foregoing embodiment, the user classification of the user can be easily obtained at any time after the user's user classification is saved, so that the user can be served according to the policy formulated for the user classification.

However, when the user has a certain time in a user classification for a certain period of time without updating, the user's interest information needs to be updated according to a preset rule, and the apparatus may further include: the interest information updating unit 705.

The interest information update unit 705 is configured to determine whether the presence time of the user in the user classification reaches a preset time, and if yes, update the interest information of the user according to a preset rule, according to the updated user's The interest information, trigger user classification matching unit 702 searches for an operation of the user classification that the difference information of the user classification and the interest information of the user most closely match from the preset user classification table.

The interest information update unit 705 enables the user's interest information to be realistically reflected in the user's current situation, thereby enabling the user to be in the user category that the user should currently be in.

The structure of the targeted advertisement placement device 602 can be as shown in FIG. 8. FIG. 8 is a structural diagram of a targeted advertisement placement device according to an embodiment of the present invention. As shown in FIG. 8, the device mainly includes: a user classification determining unit 801 and a targeted advertisement placing unit 802.

The user classification determining unit 801 is configured to search for a user classification to which the user belongs after the user logs in through the user.

After the user logs in through the client, the user classification determining unit 801 directly reads the user classification data from the shared memory, or loads the shared memory image of the user classification device into its own shared memory, thereby determining the user classification to which the user belongs. When the user classification determining unit 801 and the user classification device share the shared memory, that is, the user classification determining unit 801 and the user classification device use the same server, the user classification data read by the user classification determining unit 801 It's all in real time. When the user classification determining unit 801 needs to load the shared memory image of the user classification device into its own shared memory, the user classification determining unit and the user classification device may use different servers, and the user classification device periodically captures the shared memory image for storage. The user classification determining unit periodically acquires the shared memory image from the user classification device, and loads the acquired shared memory image into its own shared memory. With this scheme, one user classification device can serve the plurality of user classification determination units. , which can reduce operational investment.

The targeted ad serving unit 802 is configured to deliver targeted advertisements to the user according to preset rules according to the user classification.

Because users belong to different user categories, users in different user categories should have the same or similar places regardless of the classification method. Therefore, targeting advertisements to users according to different user categories can reduce the need for targeted advertisements. The processing load of the delivery device can also deliver targeted advertisements to the viewers who are truly interested in the targeted advertisements.

In the second embodiment of the targeted advertisement delivery method provided by the present invention, since the user's preference may change over time, the user's click behavior indicates the user's current preference, so after the user clicks on the targeted advertisement, according to the user The click behavior updates the user classification described by the user, so that the user classification of the user is accurate in real time. Therefore, in this embodiment, after the notification, the content triggered by the user end is obtained, and the interest information of the user corresponding to the user end is determined according to the content triggered by the user end, and the difference of the user classification is searched from the preset user classification table. The user classification that the information most closely matches the determined user's interest information adds the user to the best matching user classification.

In the embodiment of the present invention, the specific operation process of each unit in the user classification device and the targeted advertisement placement device may be consistent with the operation described in the corresponding method flow, and each unit It can be a physical functional unit, or a software functional unit, and each unit can be subdivided or merged. When the specific implementation is implemented, one of ordinary skill in the art can perform processing according to the actual situation, and is not enumerated here.

It can be seen that the method for classifying users according to the embodiment of the present invention classifies the users by the behavior of the user, so that users with the same or similar behaviors are put into the same user classification, so that only the policy for the user classification is needed to avoid The policy is set for each user, saving processing resources. Moreover, when the time in the user classification is too long and the update is not performed, the user's interest information may be updated by a preset rule, thereby ensuring that the user classification to which the user belongs can reflect the user's interest as accurately as possible. Information, so that the strategy for classifying users can meet the needs of users.

Further, the method, device and system for targeting advertisements provided by the embodiments of the present invention, because the content of the targeted advertisement is pre-set for the user classification, thereby ensuring that the targeted advertisement is served to the browser that is really interested in the targeted advertisement. In front of the user, the hit rate of targeted advertising is improved. Moreover, the embodiment of the present invention can update the user classification of the user in real time according to the behavior of the user, thereby ensuring the accuracy of the user classification to which the user belongs, and further improving the hit rate of the targeted advertisement delivery.

The method for classifying users, the method, device and system for locating advertisements provided by the embodiments of the present invention are described in detail above. The description of the above embodiments is only for helping to understand the method and the idea of the present invention; The present invention is not limited by the scope of the present invention, and the scope of the present invention is not limited by the scope of the present invention.

Claims

Claim

A method for classifying users, comprising:

Adding the user to the best matching user category.

The method according to claim 1, wherein before the determining the user's interest information according to the selection information of the user end, the method further includes:

Get the selection information of the client.

The method according to claim 1 or 2, wherein the selection information of the user end is an advertisement link content triggered by the user end;

The determining the user's interest information according to the selection information of the user terminal includes:

The user's interest information is determined according to the content of the advertisement link triggered by the user.

The method according to claim 1, wherein after the user is added to the best-matched user classification, the method further includes:

Determining whether the existence time of the user in the user classification reaches a preset time, and if yes, updating the interest information of the user according to a preset rule, and returning execution execution according to the updated user interest information The step of finding the user classification that best matches the interest information of the user in the set user classification table.

The method according to claim 1 or 4, wherein the preset user classification table includes difference information of user classifications corresponding to the user's interest information;

The searching for the user category that best matches the interest information of the user from the preset user classification table includes:

Finding difference information of the user classification from the preset user classification table and the user The interest information best matches the user classification.

A method for advertising a targeted advertisement, comprising:

After the user logs in through the user, the user classification of the user is searched; and the targeted advertisement is delivered to the user according to the preset rule according to the user classification.

The method according to claim 6, wherein after the advertised advertisement is delivered to the user according to the preset rule according to the user classification, the method further includes:

Obtaining selection information of the client;

Determining the interest information of the user according to the selection information of the user terminal;

Finding a user classification that the difference information of the user classification and the interest information of the user most closely match from the preset user classification table;

Adding the user to the best-matched user classification, and returning to perform an operation of searching for the user classification to which the user belongs after the user logs in through the user.

8. A targeted advertisement delivery system, the system comprising: a user classification device and a targeted advertisement placement device; wherein

A user classification device, comprising: an interest information determination unit, a user classification matching unit, and a saving unit; wherein

The saving unit is configured to join the user to the best matching user category, and save the And storing the user information corresponding to the user sent by the client.

The device according to claim 9, wherein the device further comprises: an acquiring unit;

The obtaining unit is configured to acquire selection information of the user end.

The device according to claim 9, wherein the device further comprises: an interest information updating unit;

The interest information update unit is configured to determine whether the time of existence of the user in the user classification reaches a preset time, and if yes, update the interest information of the user according to a preset rule, according to the updated The user's interest information triggers the user classification matching unit to search for a user classification that most closely matches the user's interest information from the preset user classification table.

12. A targeted advertisement placement device, the device comprising: a user classification determining unit and a targeted advertisement delivery unit; wherein

The user classification determining unit is configured to search for a user classification to which the user belongs after the user logs in through the corresponding user terminal;

The device according to claim 12, wherein the user classifies the selection information, and determines the user's interest information according to the selection information of the user terminal, and searches for the user from the preset user classification table. The interest information best matches the user classification, and the user is added to the best matching user classification.