CN114078016A

CN114078016A - Anti-cheating behavior identification method and device, electronic equipment and storage medium

Info

Publication number: CN114078016A
Application number: CN202010804864.4A
Authority: CN
Inventors: 孟嵩
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-08-12
Filing date: 2020-08-12
Publication date: 2022-02-22
Anticipated expiration: 2040-08-12
Also published as: CN114078016B

Abstract

The application relates to the technical field of internet, in particular to an anti-cheating behavior identification method, an anti-cheating behavior identification device, electronic equipment and a storage medium, which are used for improving the accuracy and speed of anti-cheating, wherein the method comprises the following steps: acquiring response behavior information of a target object to the released advertisements in the display page and advertisement picture information of the released advertisements; analyzing the advertisement picture information according to an advertisement display rule in the pre-acquired target anti-cheating rules to obtain a first identification result indicating whether the delivered advertisement accords with the advertisement display rule; analyzing the response behavior information according to the behavior rules in the target anti-cheating rules to obtain a second identification result which indicates whether the response behavior of the target object meets the behavior rules; and summarizing the first recognition result and the second recognition result to obtain an anti-cheating behavior recognition result. According to the method and the device, anti-cheating analysis is performed on the client locally based on the user behaviors and the advertisement pictures, so that the accuracy and the speed of anti-cheating are improved.

Description

Anti-cheating behavior identification method and device, electronic equipment and storage medium

Technical Field

The present application relates to the field of internet technologies, and in particular, to a method and an apparatus for identifying anti-cheating behavior, an electronic device, and a storage medium.

Background

Entering the mobile internet era, online advertising has become a new form of advertising, with the widespread use of networks and computers. The potential threat of advertising cheating is increasing while the online advertising brings rich traffic and revenue.

In the anti-cheating scheme in the related art, some behavior parameters are collected on the basis of an SDK (Software Development Kit) side and are sent to a server, and the server determines whether anti-cheating behaviors exist or not through anti-cheating rules. However, part of the data relates to private information, and the control of local data uploading of the user in the future becomes increasingly strict, which has high risk. Therefore, how to identify the cheating advertising behavior is one of the very important problems in the online advertising system.

Disclosure of Invention

The embodiment of the application provides an anti-cheating behavior identification method and device, electronic equipment and a storage medium, and aims to provide a local anti-cheating method which can improve the accuracy and speed of anti-cheating while guaranteeing privacy of a user.

The first anti-cheating behavior identification method provided by the embodiment of the application comprises the following steps:

acquiring response behavior information of a target object to the released advertisement in a display page and advertisement picture information of the released advertisement;

analyzing the advertisement picture information according to an advertisement display rule in a pre-acquired target anti-cheating rule to acquire a first identification result for indicating whether the delivered advertisement accords with the advertisement display rule; and

analyzing the response behavior information according to a behavior rule in the target anti-cheating rule to obtain a second identification result for indicating whether the response behavior of the target object conforms to the behavior rule;

and summarizing the first recognition result and the second recognition result to obtain an anti-cheating behavior recognition result aiming at the target object.

Optionally, the method further includes:

receiving a rule threshold value issued by the advertisement operation server;

and if the received rule threshold is different from the rule threshold in the target anti-cheating rule, replacing the rule threshold in the target anti-cheating rule with the received rule threshold, wherein the received rule threshold is obtained by the advertisement operation server performing statistical analysis on the response behavior information, corresponding to each object, of each object after information removal according to the received anti-cheating behavior identification result of each object.

The second anti-cheating behavior identification method provided by the embodiment of the application comprises the following steps:

receiving an advertisement request sent by a client, wherein the advertisement request is sent by the client after responding to the operation of requesting the advertisement by the target object;

according to the hardware parameters in the advertisement request, searching target anti-cheating rules matched with the hardware parameters from a pre-configured anti-cheating rule set, and issuing the target anti-cheating rules to the client so that the client performs anti-cheating behavior recognition on the target object according to the target anti-cheating rules, wherein the hardware parameters are the hardware parameters of the terminal equipment where the client is located, the target anti-cheating rules comprise advertisement display rules and behavior rules, the advertisement display rules are used for analyzing advertisement picture information of the delivered advertisements, and the behavior rules are used for analyzing response behavior information of the target object aiming at the delivered advertisements.

The third anti-cheating behavior identification method provided by the embodiment of the application comprises the following steps:

receiving response behavior information which is sent by each client and corresponds to each object and is subjected to target information removal, and anti-cheating behavior identification results aiming at each object;

and performing statistical analysis on the response behavior information after the target information is removed based on the anti-cheating behavior recognition result, determining a rule threshold value in a target anti-cheating rule, and respectively issuing the rule threshold value to the client side of each object, so that when the client side of each object determines that the rule threshold value in the pre-obtained target anti-cheating rule is different from the received rule threshold value, the rule threshold value in the pre-obtained target anti-cheating rule is replaced by the received rule threshold value, the target anti-cheating rule comprises an advertisement display rule and a behavior rule, the advertisement display rule is used for analyzing the advertisement picture information of the delivered advertisement, and the behavior rule is used for analyzing the response behavior information of the object aiming at the delivered advertisement.

The first anti-cheating behavior recognition device provided by the embodiment of the application comprises:

the information acquisition unit is used for acquiring response behavior information of a target object to the released advertisements in the display page and advertisement picture information of the released advertisements;

the analysis unit is used for analyzing the advertisement picture information according to an advertisement display rule in a pre-acquired target anti-cheating rule to acquire a first identification result for indicating whether the delivered advertisement accords with the advertisement display rule; analyzing the response behavior information according to a behavior rule in the target anti-cheating rule to obtain a second identification result for indicating whether the response behavior of the target object conforms to the behavior rule;

and the summarizing unit is used for summarizing the first identification result and the second identification result to obtain an anti-cheating behavior identification result aiming at the target object.

Optionally, the analysis unit is specifically configured to:

carrying out image recognition on the advertisement picture information, and determining an advertisement area in the delivered advertisement;

and analyzing the picture proportion of the advertisement area in the released advertisement based on a target advertisement display rule corresponding to the advertisement type of the released advertisement in the advertisement display rules to obtain the first identification result.

Optionally, if the advertisement type of the delivered advertisement is a target advertisement type;

the identification unit is further configured to:

acquiring target area information contained in the delivered advertisement;

the identification unit is specifically configured to:

analyzing the target area information based on a target advertisement display rule corresponding to the advertisement type of the delivered advertisement in the advertisement display rules;

and obtaining the first identification result according to the analysis result of the target area information and the analysis result of the picture proportion of the advertisement area in the delivered advertisement.

Optionally, the response behavior information includes response information of various types of response behaviors of the target object to the delivered advertisement; the identification unit is specifically configured to:

and respectively comparing the response information corresponding to the various response behaviors with rule thresholds corresponding to the various response behaviors set in the behavior rule to obtain a second identification result for indicating whether the response behavior of the target object conforms to the behavior rule.

Optionally, the summarizing unit is specifically configured to:

if the first identification result shows that the delivered advertisement accords with the advertisement display rule, determining that the anti-cheating behavior identification result comprises that the delivered advertisement is a normal advertisement; otherwise, determining that the anti-cheating behavior recognition result comprises that the delivered advertisement is a cheating advertisement;

if the second recognition result shows that the target object accords with the behavior rule, determining that the anti-cheating behavior recognition result comprises that the target object is a normal object; otherwise, determining that the anti-cheating behavior recognition result comprises that the target object is a cheating object.

Optionally, the apparatus further comprises:

a transmission unit, configured to obtain the target anti-cheating rule based on the following manner:

responding to the operation of the target object for requesting the advertisement, and sending an advertisement request to a local anti-cheating rule file server, wherein the advertisement request comprises the hardware parameters of the terminal equipment where the client is located;

and receiving a target anti-cheating rule which is matched with the hardware parameter and returned by the local anti-cheating rule file server.

Optionally, the summarizing unit is further configured to:

removing target information in the response behavior information according to a preset information removal rule to obtain the response behavior information after the target information is removed;

and uploading the response behavior information after the target information is removed and the anti-cheating behavior recognition result of the target object to an advertisement operation server, so that the advertisement operation server performs statistical analysis on the response behavior information after the target information is removed according to the anti-cheating behavior recognition result.

Optionally, the summarizing unit is specifically configured to:

classifying the response behavior information according to preset response categories, determining original categories to which various response behaviors belong, and recording response time corresponding to the various response behaviors;

removing target information in the category to which each type of response behavior belongs according to the information removal rule to obtain abstract categories of each type of response behavior;

classifying and combining the abstract categories of the various response behaviors with a preset operation set to obtain target categories of the various response behaviors;

and binding the target types of the various response behaviors and the corresponding response time to be used as the response behavior information after the target information is removed.

Optionally, the apparatus further comprises:

the updating unit is used for receiving the rule threshold value issued by the advertisement operation server;

and if the received rule threshold is different from the rule threshold in the target anti-cheating rule, replacing the rule threshold in the target anti-cheating rule with the received rule threshold, wherein the received rule threshold is obtained by the advertisement operation server performing statistical analysis on the response behavior information, corresponding to each object, of each object after the target information is removed according to the received anti-cheating behavior identification result of each object.

The second anti-cheating behavior recognition device provided by the embodiment of the application comprises:

the system comprises a first receiving unit, a second receiving unit and a third receiving unit, wherein the first receiving unit is used for receiving an advertisement request sent by a client, and the advertisement request is sent by the client after responding to the operation of requesting the advertisement of a target object;

and the rule issuing unit is used for searching a target anti-cheating rule matched with the hardware parameter from a pre-configured anti-cheating rule set according to the hardware parameter in the advertisement request, and issuing the target anti-cheating rule to the client so that the client performs anti-cheating behavior identification on the target object according to the target anti-cheating rule, wherein the hardware parameter is the hardware parameter of the terminal equipment where the client is located, the target anti-cheating rule comprises an advertisement display rule and a behavior rule, the advertisement display rule is used for analyzing advertisement picture information of the delivered advertisement, and the behavior rule is used for analyzing response behavior information of the target object aiming at the delivered advertisement.

The third anti-cheating behavior recognition device provided by the embodiment of the application comprises:

the second receiving unit is used for receiving response behavior information which is sent by each client and corresponds to each object and is obtained after the target information is removed, and anti-cheating behavior identification results aiming at each object;

and the statistical analysis unit is used for performing statistical analysis on the response behavior information after the target information is removed based on the anti-cheating behavior identification result, determining a rule threshold value in a target anti-cheating rule, and respectively issuing the rule threshold value to the client side of each object, so that when the client side of each object determines that the rule threshold value in the pre-obtained target anti-cheating rule is different from the received rule threshold value, the rule threshold value in the pre-obtained target anti-cheating rule is replaced by the received rule threshold value, the target anti-cheating rule comprises an advertisement display rule and a behavior rule, the advertisement display rule is used for analyzing the advertisement picture information of the delivered advertisement, and the behavior rule is used for analyzing the response behavior information of the object aiming at the delivered advertisement.

An electronic device provided by an embodiment of the present application includes a processor and a memory, where the memory stores a program code, and when the program code is executed by the processor, the processor is caused to execute any one of the steps of the anti-cheating-behavior recognition method.

An embodiment of the present application provides a computer-readable storage medium, which includes program code, when the program code runs on an electronic device, the program code is configured to enable the electronic device to perform any one of the steps of the anti-cheating-behavior recognition method.

Embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to cause the computer device to perform the steps of any of the above anti-cheating behavior recognition methods.

The beneficial effect of this application is as follows:

the anti-cheating behavior identification method and device, the electronic equipment and the storage medium provided by the embodiment of the application, because the embodiment of the application migrates the center of gravity of the anti-cheating from the server side to the client side, the client side acquires the anti-cheating rules for identifying the anti-cheating behaviors in advance, further, the object behaviors, the advertisement pictures and the like collected locally are analyzed to obtain the final anti-cheating behavior recognition result, information related to privacy in the information such as the object behaviors, the advertisement pictures and the like does not need to be uploaded to a server, the method and the device have the advantages that the client side directly analyzes the information, policy risks of privacy information collection can be avoided, and in addition, compared with the related technology, the method and the device increase local anti-cheating behavior judgment based on user behavior analysis and an anti-cheating judgment method based on local advertisement picture analysis, improve the accuracy and speed of anti-cheating, and save server bandwidth resources.

Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the application. The objectives and other advantages of the application may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

fig. 1 is an alternative schematic diagram of an application scenario in an embodiment of the present application;

fig. 2 is a schematic flowchart illustrating a first anti-cheating behavior recognition method in an embodiment of the present application;

FIG. 3 is a schematic diagram of a display page in an embodiment of the present application;

fig. 4 is a schematic flow chart of desensitization processing on response behavior information in the embodiment of the present application;

fig. 5 is a schematic flowchart illustrating a second anti-cheating behavior recognition method in an embodiment of the present application;

fig. 6 is a flowchart illustrating a third anti-cheating behavior recognition method in an embodiment of the present application;

fig. 7A is a schematic view illustrating an interaction timing sequence of an anti-cheating behavior recognition method according to an embodiment of the present application;

FIG. 7B is a schematic diagram illustrating an interaction timing sequence of an anti-cheating behavior recognition method according to another embodiment of the present application;

fig. 8 is a schematic structural diagram illustrating a first anti-cheating-behavior recognizing apparatus according to an embodiment of the present application;

fig. 9 is a schematic structural diagram illustrating a second anti-cheating-behavior recognizing device according to an embodiment of the present application;

fig. 10 is a schematic structural diagram illustrating a third anti-cheating-behavior recognizing device according to an embodiment of the present application;

fig. 11 is a schematic diagram of a hardware component structure of an electronic device to which an embodiment of the present application is applied.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments, but not all embodiments, of the technical solutions of the present application. All other embodiments obtained by a person skilled in the art without any inventive step based on the embodiments described in the present application are within the scope of the protection of the present application.

Some concepts related to the embodiments of the present application are described below.

OpenCV: OpenCV is a cross-platform computer vision library issued based on BSD (Berkeley Software Distribution) licenses, which can run on a variety of operating systems. It implements many general algorithms in image processing and computer vision. In the embodiment of the present application, the advertisement screen information may be analyzed by combining with the OpenCV technology.

And (3) SDK: SDK is a collection of development tools used by some software engineers to build application software for a particular software package, software framework, hardware platform, operating system, etc. In the related art, some behavior parameters of the user are collected mainly on the basis of the SDK side and are sent to the server, and anti-cheating behavior recognition is carried out by the server.

And (3) advertisement exposure: refers to the advertisement being shown at an ad slot (e.g., an ad slot in a page visited by a user, an ad slot in an application used by a user) on the user side, and one time the advertisement is shown on the user side is called an ad exposure.

And (3) clicking the advertisement: a user accesses a page of an advertiser by clicking an advertisement on a user side device (such as a terminal device like a smart phone or a tablet computer), and the user accesses the page of the advertiser by clicking the advertisement once, which is called advertisement clicking.

Advertising cheating and cheating objects: in the links of advertisement exposure, click, effect and the like, a user has behaviors capable of improving indexes such as advertisement exposure, advertisement click rate, conversion rate and the like for some malicious purposes, and the malicious behaviors of the cheating object are called advertisement cheating. Accordingly, the user who produces the cheating action is referred to as a cheating user, or a cheating object. The cheating object may be a network person who achieves the purposes of profit or public opinion building by means of clicking advertisements, downloading applications or posting replies and the like, may also be a natural person, may also be a cheating program of a disguised user and the like.

Anti-cheating behavior recognition: and checking links such as advertisement exposure, click, effect and the like, and judging whether the advertisement exposure, the advertisement click, the advertisement effect and the like are triggered by normal access of a user side or are realized by cheating objects through an advertisement cheating means.

The client, or called user side, refers to a program corresponding to the server and providing local services to the client. Except for some application programs which only run locally, the application programs are generally installed on common clients and need to be operated together with a server. After the internet has developed, the more common clients include web browsers used on the world wide web, email clients for receiving and sending emails, and client software for instant messaging. For this kind of application, a corresponding server and a corresponding service program are required in the network to provide corresponding services, such as database services, e-mail services, etc., so that a specific communication connection needs to be established between the client and the server to ensure the normal operation of the application program.

Artificial Intelligence (AI) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.

The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.

With the research and progress of artificial intelligence technology, the artificial intelligence technology is developed and applied in a plurality of fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, automatic driving, unmanned aerial vehicles, robots, smart medical care, smart customer service, and the like.

The scheme provided by the embodiment of the application relates to the machine learning technology of artificial intelligence. Machine learning is a way to realize artificial intelligence, has certain similarity with data mining, is a field cross subject, and relates to multiple subjects such as probability theory, statistics, approximation theory, convex analysis, computational complexity theory and the like. Compared with the data mining method for finding mutual characteristics among big data, machine learning focuses on the design of an algorithm, so that a computer can automatically learn rules from the data and predict unknown data by using the rules.

In the embodiment of the application, the anti-cheating behavior of the target object can be identified based on the anti-cheating model by deploying the anti-cheating model at the client. Wherein, the anti-cheating model is obtained based on machine learning technology training. The details are described below with reference to specific embodiments.

The following briefly introduces the design concept of the embodiments of the present application:

in the era of mobile internet, advertisements bring rich traffic revenue to advertisers. Currently, online advertising has become a new form of advertising, with the widespread use of networks and computers. In online advertising systems, advertisers pay advertising publishers to deliver their advertisements through web pages, browsers, Application programs (APPs), or other online media.

However, with the rapid development of the internet, the potential threat of the grey industry chain related to advertisement cheating is increasing while the online advertisement brings rich traffic revenue. In the related art, some behavior parameters are collected on the basis of the SDK side and sent to the server, and the server judges whether anti-cheating behaviors exist or not through anti-cheating rules.

However, the related art solution needs to upload the collected behavior parameters and the like to the server, and since part of the data relates to private information, such as some websites browsed by the user or some specific interactions and behaviors, and the control of uploading the local data of the user will become increasingly strict in the future, this approach has a high risk. In addition, the related art solution only supports uploading structured data with limited data length, such as click coordinates, but cannot upload some data, such as advertisement picture information in the current screen of the user.

In view of this, the embodiment of the present application provides a method for locally determining whether a user has a cheating behavior at a client, which adds a local anti-cheating behavior determination based on user behavior analysis and an anti-cheating determination method based on local screenshot and analysis on the basis of an original online anti-cheating behavior determination, wherein the client identifies the anti-cheating behavior according to a pre-obtained target anti-cheating rule, and private information and the like do not need to be transmitted to a server, and the private information or some complex structured data and the like can be directly collected locally, so that policy risk of future private information collection is avoided, accuracy and speed of anti-cheating are improved, and bandwidth resources of the server are also saved.

The preferred embodiments of the present application will be described below with reference to the accompanying drawings of the specification, it should be understood that the preferred embodiments described herein are merely for illustrating and explaining the present application, and are not intended to limit the present application, and that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.

Fig. 1 is a schematic view of an application scenario in the embodiment of the present application. The method is an application scenario schematic diagram of the embodiment of the application. The application scenario diagram includes two terminal devices 110 and a server 130, and the advertisement interface 120 can be logged in through the terminal devices 110. The terminal device 110 and the server 130 can communicate with each other through a communication network. In fig. 1, the user corresponds to the left terminal device 110, and the user B corresponds to the right terminal device 110 as an example, so that the number of terminal devices is not limited in practice. In some cases, the terminal devices may communicate with each other through the server 130 first, direct communication may be established between the terminal devices, and the manner of direct communication between the terminal devices may be referred to as point-to-point communication, in which case, some interaction processes between the terminal devices 110 may not require the relay of the server 130.

Wherein, an online document client can be installed in each terminal device. In this embodiment of the application, the client may be social software, such as instant messaging software and short video software, and may also be an applet, a web page, and the like, which is not limited herein. The terminal device 110 is required to be installed with a client, where the client may be a software client, or a web page, an applet, or the like, and the server 130 is a server corresponding to the software client, the web page, the applet, or the like.

In an alternative embodiment, the communication network is a wired network or a wireless network. The terminal device 110 and the server 130 may be directly or indirectly connected through wired or wireless communication, and the application is not limited herein.

In this embodiment, the terminal device 110 is an electronic device used by a user, and the electronic device may be a computer device having a certain computing capability and running instant messaging software and a website or social contact software and a website, such as a personal computer, a mobile phone, a tablet computer, a notebook, an e-book reader, and the like. Each terminal device 110 is connected to the server 130 through a wireless Network, and the server 130 may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a Network service, cloud communication, middleware service, a domain name service, a security service, a CDN (Content Delivery Network), and a big data and artificial intelligence platform.

In the embodiment of the present application, a user may log in the advertisement interface 120 through the terminal device 110, and the terminal device 110 responds to various operations triggered by the user at the advertisement interface 120, such as clicking, sliding, and the like.

After terminal device 110 responds to the user operation, there may be interaction with server 130, such as terminal device 110 sending an advertisement request to server 130, obtaining target anti-cheating rules from server 130, and so on.

Referring to fig. 2, an implementation flow chart of an anti-cheating behavior recognition method provided in the embodiment of the present application is applied to a client, and a specific implementation flow of the method is as follows:

s21: acquiring response behavior information of a target object to the released advertisements in the display page and advertisement picture information of the released advertisements;

in the embodiment of the present application, taking a target object as an example of a user, after a user requests an advertisement on a client side, the user may generate interactive behaviors, such as click, sliding, and the like, on the delivered advertisement displayed in a display page, where these behaviors are collectively referred to as response behaviors of the user, and the corresponding response behavior information refers to data related to these behaviors, such as the number of clicks, the click time, and the like. The advertisement picture information may be an advertisement picture obtained by screenshot, screen recording and the like. Hereinafter, the advertisement screen information mainly refers to a screenshot of an advertisement screen.

S22: analyzing advertisement picture information according to an advertisement display rule in a pre-acquired target anti-cheating rule to obtain a first identification result for indicating whether the delivered advertisement accords with the advertisement display rule; analyzing the response behavior information according to the behavior rules in the target anti-cheating rules to obtain a second identification result for indicating whether the response behavior of the target object meets the behavior rules;

in an optional implementation manner, the target anti-cheating rule obtained in advance by the client side may be pulled from the server side in advance, and the specific implementation process is as follows:

the client responds to the operation of requesting the advertisement of the target object and sends an advertisement request to the local anti-cheating rule file server, wherein the advertisement request comprises the hardware parameters of the terminal equipment where the client is located; after receiving an advertisement request sent by a client, a local anti-cheating rule file server searches a target anti-cheating rule matched with a hardware parameter from a pre-configured anti-cheating rule set according to the hardware parameter in the advertisement request, and issues the target anti-cheating rule to the client.

The anti-cheating rules in the anti-cheating rule set stored on the local anti-cheating rule file server side can be derived from any one or more of the following three ways:

in the first mode, the anti-cheating rules of the partial behavior data which is already run on the line can comprise at least one of advertisement display rules and behavior rules;

and secondly, collecting historical behavior data uploaded by the user, and formulating corresponding behavior rules by analyzing the historical behavior data. For example, the dwell time of the user opening the page may be recorded, and the mean, median, and variance of the dwell time may be viewed by the behavior rules. If the dwell time of each page is approximately the same or the data shows a periodic rule, the operation script of the machine is suspected to cheat. The assumption is that according to the statistical analysis of historical behavior data uploaded by users, the following is finally appointed: the dwell time of the page is longer than 1 second, and if the dwell time of the page is shorter than 1 second, cheating behavior exists;

and thirdly, manually giving advertisement display rules of different advertisement categories, for example, for a certain advertisement type, setting the height of the advertisement subject view to be not less than 75% of the height of the whole screen, and if the height is less than 75%, indicating that the condition of advertisement cheating exists.

The anti-cheating rule set constructed by the enumerated method comprises behavior rules corresponding to different hardware parameters and advertisement display rules, wherein rule thresholds in the behavior rules corresponding to different hardware parameters are different, or rule thresholds in the advertisement display rules corresponding to different hardware parameters are different.

The rule threshold refers to a reference standard set in the anti-cheating rule, for example, for a behavior rule that "the page stays for longer than 1 second", the rule threshold is 1 second, and the rule threshold is a reference standard of the user page browsing time, where the page staying time refers to the advertisement browsing time of the user. For different hardware parameters, for example, for some new models of terminal devices, since a Central Processing Unit (CPU) has a fast speed, a rule threshold of a page dwell time is low; for some terminal devices of old models, the rule threshold of the page stay time is higher because the CPU speed is slower.

For another example, for an advertisement presentation rule, "the height of the advertisement subject view must not be less than 75% of the overall screen height," where the rule threshold is 75%, which is a reference criterion for the height of the advertisement subject view.

It should be noted that, in the embodiment of the present application, each time the local anti-cheating rule file server updates the anti-cheating rule, the local anti-cheating rule file server will send the rule version number +1, and the final rule will be issued according to the request of the client, so that the client can update the rule in time.

When the user requests the advertisement, the user requests the latest anti-cheating rule in the local anti-cheating rule file server. The local anti-cheating rules file server will adapt different anti-cheating rules based on the hardware parameters in the ad request. For example, for the rule threshold of the page stay time, the rule threshold corresponding to the model a is 1 second, the rule threshold corresponding to the model B is 1.5 seconds, and the rule threshold corresponding to the model C is 2 seconds. For another example, for the rule threshold of the advertisement subject view height, the rule threshold corresponding to the model a is 75%, the rule threshold corresponding to the model B is 70%, and the rule threshold corresponding to the model C is 80%.

The terminal device requesting the advertisement at this time is of type a, so the behavior rules in the target anti-cheating rules issued by the server to the client include: the page dwell time is greater than 1 second. The advertisement display rules in the target anti-cheating rules include: the height of the ad subject view must be no less than 75% of the overall screen height.

After the client receives the target anti-cheating rule issued by the local anti-cheating rule file server, whether the local anti-cheating rule needs to be updated or not is determined according to comparison between the version number in the anti-cheating rule information and the local version number, if yes, the local anti-cheating rule is replaced and persisted, and then after the user request advertisement is detected again, the target anti-cheating rule does not need to be pulled from the local anti-cheating rule file server, but the local target anti-cheating rule is directly adopted.

For example, in the above embodiment, the version number of the target anti-cheating rule issued by the local anti-cheating file server is V2, and the version number of the anti-cheating rule locally stored at the client is V1, in this case, the target anti-cheating rule issued by the local anti-cheating file server is used to replace the local anti-cheating rule and persist the target anti-cheating rule. And then, performing anti-cheating behavior analysis on the target object based on the target anti-cheating rule to obtain a first recognition result and a second recognition result.

S23: and summarizing the first recognition result and the second recognition result to obtain an anti-cheating behavior recognition result aiming at the target object.

When a user triggers an advertisement operation, such as an exposure (advertisement is shown), a click (advertisement is clicked), a conversion (one-time downloading is completed) and the like, the local anti-cheating module judges the local anti-cheating behavior according to data collected by the local behavior data collector in the process, and the local anti-cheating behavior is mainly divided into two parts: the first part is to analyze the advertisement picture information locally at the client based on the advertisement display rule; the second part is to analyze the response behavior information of the user locally on the client based on the behavior rule.

The following first introduces the detailed process of analyzing the advertisement picture information based on the advertisement display rule:

in an optional implementation manner, the advertisement display rule is related to the advertisement type, and when the advertisement display rule in the target anti-cheating rule is used to analyze the advertisement picture information, it needs to determine whether the advertisement picture information of the delivered advertisement conforms to the target advertisement display rule corresponding to the corresponding advertisement type, and the specific process is as follows:

firstly, image recognition is carried out on advertisement picture information of released advertisements displayed in a display page, and an advertisement area in the released advertisements is determined; and further analyzing the picture proportion of the advertisement area in the released advertisement based on a target advertisement display rule corresponding to the advertisement type of the released advertisement in the advertisement display rule to obtain a first identification result.

In the embodiment of the application, the advertisement area occupation ratios required in the advertisement pictures of different advertisement types or other picture information may be different, so that when the analysis is performed, the application adopts the corresponding target advertisement display rule to perform judgment, and mainly judges whether the picture occupation ratio of the advertisement area in the delivered advertisement meets the regulation or not.

Taking advertisement picture information as a screenshot as an example, as shown in fig. 3, it is a schematic diagram of an open screen advertisement screenshot shown in the embodiment of the present application. Firstly, the method of OpenCV and the like is adopted to carry out picture marking and picture character recognition on the advertisement screenshot, and basic data and coordinates of the picture are divided. The basic data of the picture comprises characters in the picture, basic picture data in the picture and the like, and the coordinates mainly refer to the coordinates of the view part of the advertisement main body or can also comprise the coordinates of boundary points of the page foot part of the advertisement and the like.

Wherein for any one advertisement, the advertisement may be roughly divided into a maximum of three regions, depending on the different advertisement types. Taking the open screen advertisement as an example, the advertisement area may be divided into two parts, namely a Body part and a Footer part, and it is assumed that the advertisement display rule corresponding to the first screen advertisement in the advertisement display rule received by the client specifies that the Body area should not be less than 75% of the screen area, and the Footer area should not exceed 25% of the screen area.

After the data are obtained, matching is performed by combining the target advertisement display rule corresponding to the advertisement type of the advertisement shown in fig. 3, and whether the target advertisement display rule is satisfied is judged. Assume that the Body region of the advertisement screenshot shown in FIG. 3 occupies 80% of the screen area, over 75%; the Footer area occupies 20% of the screen area, and is less than 25%, that is, the Footer area meets the target advertisement display rule, and at this time, the corresponding first recognition result can be represented by 1; if the target advertisement display rule is not met, the corresponding first recognition result can be represented by 0.

Optionally, for the target type advertisement, some advertisement icons or character identifiers may exist in the advertisement picture, so that it is necessary to further determine whether the information of the target area in the advertisement picture conforms to the corresponding advertisement display rule, in addition to determining the advertisement area ratio. In this case, when the first recognition result is determined, it is necessary to analyze the screen occupation ratio of the advertisement area in the delivered advertisement and also to analyze the target area information, and to obtain the first recognition result from the two analysis results. If any one of the two parts does not accord with the target advertisement display rule, the corresponding first identification result can be represented by 0; if both of the two parts meet the target advertisement display rule, the corresponding first recognition result is represented by 1.

Taking the X-type advertisement targeting a certain advertisement delivery system as an example, it is assumed that there must be a logo (icon) and an "advertisement" two-word in the advertisement, as shown by the dashed box S30 in fig. 3, and these information are the target area information, which is the part of the target area shown by the dashed box S30, i.e. the lower right corner of Body area.

If the X-type open screen advertisement includes the target area information shown by the dashed box S30, and the Body area should not be less than 75% of the screen area, and the Footer area should not exceed 25% of the screen area, it indicates that the advertisement is a valid X-type advertisement, i.e., a normal advertisement, otherwise it may be determined that the advertisement is a cheating advertisement, i.e., an invalid advertisement, which cannot or affects the revenue of the advertisement delivery party. For example, for a normal open screen advertisement, a user can jump to a link corresponding to the advertisement by clicking, and the cheating advertisement cannot achieve the effect, and may only be a cheating advertisement picture. In the embodiment of the application, whether the user cheats or not is judged on the basis that the advertisement picture information accords with the advertisement display rule.

It should be noted that, the above embodiment implements analysis of comparison advertisement cheating behaviors through local advertisement screenshots, and can be used as an important supplement of anti-cheating capability based on online mobile phone data in the related art, so as to promote development of migrating an anti-cheating center of gravity from a server to a client and receive overhead of related resources of the server.

After the advertisement display rules are introduced, the following detailed description is given to the process of analyzing the response behavior information according to the behavior rules in the target anti-cheating rules:

in an optional implementation manner, when response behavior information of a target object is analyzed according to a behavior rule in a target anti-cheating rule to obtain a second recognition result for indicating whether the response behavior of the target object meets the behavior rule, the response behavior information of the target object is preferably needed to be analyzed, and various types of response behaviors of the target object for a released advertisement and response information corresponding to the various types of response behaviors are determined; and then, respectively comparing the response information corresponding to the various response behaviors with rule thresholds corresponding to the various response behaviors set in the behavior rule to obtain a second identification result for indicating whether the response behavior of the target object meets the behavior rule or not.

The response behavior information of the user to the delivered advertisement comprises data such as click coordinates, gesture sliding distance, historical browsing page information and the like, and the corresponding response behaviors are click, sliding and browsing pages respectively.

And respectively carrying out behavior rule matching on the response information corresponding to each type of response behaviors to see whether cheating problems exist. For example, it is determined whether the click coordinate of the user meets a coordinate rule threshold corresponding to the click coordinate, whether the gesture sliding distance of the user meets a distance rule threshold corresponding to the click coordinate, whether the time length for the user to browse the advertisement display page meets a time length rule threshold, and the like. If at least N pieces of response information corresponding to the response behaviors do not accord with the corresponding rule threshold, determining that the second recognition result represents that the response behaviors of the target object do not accord with the behavior rules, wherein the second recognition result can be represented by 0, N is a positive integer and can be determined according to the actual situation; otherwise, it may be determined that the response behavior of the second recognition result representing the target object meets the behavior rule, and the second recognition result may be represented by 1.

A target anti-cheating rule obtained by the client is listed below, and the specific rule format is as follows:

in the above enumerated target anti-cheating rules:

the Version field represents the Version number of the target anti-cheating rule, the BehaviorRules field represents the behavior rule, and the ScreenShotRules field represents the advertisement display rule.

Specifically, the rule format of the behavior rule is as follows: and the method comprises the following steps of [ < formula, value range >, … ], wherein each group of < formula, value range > represents behavior information of a type of response behavior and the value range (namely a rule threshold) corresponding to the type of response behavior. For example, > < the page stay time length, > < 1> indicates that the average page browsing time length is greater than 1s, and the time length rule threshold is 1 s.

The AdType1 field in the rule format of the advertisement display rule represents the advertisement type, which is exemplified by taking the advertisement type1 as an example, and for the advertisement of the type, the image mustlnclude field represents a picture URL, which can be used for judging whether the advertisement URL is correct or not; the includeDescription field represents that the text is contained and can be used for judging whether the advertisement screenshot contains the target text or not; the BodyViewHeightRange [ a, b ] can be used for judging whether the height of the view of the advertising subject is in accordance with the interval defined by the rule threshold value a and the rule threshold value b, for example, the value of a is 75, and the value of b is 100, which means that the height of the view of the advertising subject is not less than 75% of the height of the whole screen. Similarly, the HeaderViewHeightRange: [ a, b ] represents the duty cycle of the advertisement header view height compared to the entire screen height, and the FooterViewHeightRange: [ a, b ] represents the duty cycle of the advertisement footer view height compared to the entire screen height.

After the first recognition result and the second recognition result are obtained, the first recognition result and the second recognition result are summarized by the client. If the cheating is detected, the specific reason of the cheating is taken, and the operation is uploaded to the advertisement operation server along with exposure, clicking and conversion operation to serve as an important basis for judging whether the cheating is detected by the advertisement operation server. Meanwhile, local interaction data and historical behavior data can be desensitized and then uploaded to an advertisement operation server, or uploaded to a local anti-cheating rule file server and then reach a user interaction information database to serve as reference data when the local anti-cheating rule file server updates an anti-cheating rule set.

In the embodiment of the application, when the client performs the summary processing, for the first identification result, if the value of the first identification result is 1, it indicates that the delivered advertisement meets the advertisement display rule, and at this time, the anti-cheating behavior identification result includes that the delivered advertisement is a normal advertisement; if the first recognition result takes a value of 0, the delivered advertisement is not in accordance with the advertisement display rule, and the anti-cheating behavior recognition result comprises that the delivered advertisement is a cheating advertisement.

For the second recognition result, if the value of the second recognition result is 1, the target object is represented to conform to the behavior rule, and the anti-cheating behavior recognition result includes that the target object is a normal object; if the value of the second recognition result is 0, the target object does not conform to the behavior rule, and the anti-cheating behavior recognition result comprises that the target object is a cheating object.

Therefore, when the first recognition result and the second recognition result are summarized, the following four cases can be totally distinguished:

in case one, the first recognition result indicates that the delivered advertisement conforms to the advertisement display rule, and the second recognition result indicates that the target object conforms to the behavior rule, in this case, the anti-cheating behavior recognition result includes: the delivered advertisements are normal advertisements, the target objects are normal objects, namely the advertisements are delivered normally, and the target objects do not cheat.

And in case two, the first recognition result shows that the delivered advertisement accords with the advertisement display rule, and the second recognition result shows that the target object does not accord with the behavior rule, wherein the anti-cheating behavior recognition result comprises the following steps: the delivered advertisements are normal advertisements, the target objects are cheating objects, that is, the advertisements are normally delivered, and the behaviors of the target objects belong to the cheating behaviors.

And in a third case, the first recognition result shows that the delivered advertisement does not accord with the advertisement display rule, and the second recognition result shows that the target object accords with the behavior rule, wherein the anti-cheating behavior recognition result comprises the following steps: the delivered advertisements are cheating advertisements, the target objects are normal objects, namely the advertisement delivery is abnormal, the screenshot of the advertisement does not accord with the advertisement display rule and is a false advertisement, but the behavior of the target objects belongs to normal behavior, and the target objects are not cheated.

And in case that the first recognition result shows that the delivered advertisement does not accord with the advertisement display rule, and the second recognition result shows that the target object does not accord with the behavior rule, the anti-cheating behavior recognition result comprises the following steps: the released advertisement is a cheating advertisement, the target object is a cheating object, namely the advertisement is released abnormally, the screenshot of the advertisement does not accord with the advertisement display rule and is a false advertisement, and the behavior of the target object belongs to the cheating behavior.

In the embodiment of the application, after the client obtains the anti-cheating behavior recognition result for the target object through the summarizing processing of the first recognition result and the second recognition result, the anti-cheating behavior recognition result can be uploaded to the advertisement operation server and used for statistical analysis of the advertisement operation server.

In addition, the client analyzes the response behavior information of the user and the advertisement picture information of the delivered advertisement according to the target anti-cheating rule, and can further desensitize the response behavior information of the user and upload the desensitized response behavior information to a user interaction information database and an advertisement operation server, wherein the response behavior information of the user comprises local interaction data, historical behavior data and the like.

In an optional implementation manner, the client removes target information in the response behavior information according to a preset information removal rule, and obtains response behavior information with the target information removed, that is, response behavior information with desensitization processing performed on the response behavior information of the user; the response behavior information after the target information is removed and the anti-cheating behavior recognition result of the target object are uploaded to an advertisement operation server; and after receiving the response behavior information which is sent by each client and corresponds to each object and is subjected to the removal of the target information and the anti-cheating behavior identification result aiming at each object, the advertisement operation server performs statistical analysis on the response behavior information subjected to the removal of the target information based on the anti-cheating behavior identification result, determines a rule threshold value in a target anti-cheating rule, and respectively issues the determined rule threshold value to the clients of each object.

In the embodiment of the application, after receiving the rule threshold value issued by the advertisement operation server, the client compares the received rule threshold value with the rule threshold value in the locally stored target anti-cheating rule, and when the rule threshold value in the target anti-cheating rule is determined to be different from the rule threshold value issued by the advertisement operation server, the rule threshold value in the target anti-cheating rule is replaced by the rule threshold value issued by the advertisement operation server.

For example, the target anti-cheating rules issued by the local anti-cheating rule file server to the client include: the dwell time of the page is more than 1 second; the height of the ad subject view must be no less than 75% of the overall screen height. Wherein, the rule threshold corresponding to the page dwell time is 1 second, and the rule threshold corresponding to the advertisement subject view height. Assuming that the new page staying time rule threshold value issued by the advertisement operation server is 1.5 seconds, at this time, the client determines that the new page staying time rule threshold value is different from the page staying time rule threshold value in the target anti-cheating rule, and the behavior rule about the page staying time in the target anti-cheating rule is adjusted as follows: the page dwell time is greater than 1.5 seconds.

It should be noted that, in this embodiment of the application, the advertisement operation server may further adjust the rule threshold in combination with the anti-cheating behavior recognition result uploaded by the client and the response behavior information of the user, and then issue the rule threshold to the client, where the client adjusts the corresponding target anti-cheating rule, so as to improve the accuracy of anti-cheating rule behavior recognition.

The step of removing the target information refers to desensitizing the response behavior information of the user and removing sensitive information in the response behavior information. Referring to fig. 4, which is a flow chart of a desensitization method listed in the examples of the present application, the specific process is as follows:

in step S41, classifying the response behavior information according to preset response categories, determining original categories to which the various types of response behaviors belong, and recording response times corresponding to the various types of response behaviors;

where the original category of response behavior is also the specific class containing sensitive information. When classifying the response behavior information according to the preset response category, the classification mainly refers to classification according to the category name of the user staying or interacting, such as UserView (user view), ClickButton (click button), MessageDialogue (information dialogue), InputTextield (input field), and the like, and the specific time of staying or interacting is recorded, and the unit is ms.

For example, the retention time corresponding to the UserView behavior is 2 s; the retention time for the ClickButton behavior is 0.5 s; the MessageDialogue line corresponds to a dwell time of 5s of InputTextileld and to a dwell time of 3 s.

In step S42, removing target information in the category to which each type of response behavior belongs according to an information removal rule to obtain an abstract category of each type of response behavior;

the process is a process of abstracting user response behaviors, and specific categories of various types of response behaviors obtained by dividing based on the above manner are abstracted, for example, the four listed original categories can be abstracted into View, Button, dialog and Input. In the embodiment of the present application, the target information refers to sensitive information related to some behaviors of the user related to privacy.

In step S43, classifying and merging the abstract categories of each type of response behavior with a preset operation set to obtain target categories of each type of response behavior;

after the abstract category is obtained, the abstract category may be merged into a preset operation set, where the preset operation set enumerated in the embodiment of the present application is: after classifying and combining the four listed abstract categories and the preset operation set, the examples are classified as < static View element, Button interactive element, static View element, and Input box interactive element >, that is, the object category corresponding to View is static View element, the object category corresponding to Button is Button interactive element, the object category corresponding to Dialogue is static View element, and the object category corresponding to Input is Input box interactive element.

In step S44, the target category of each type of response behavior and the corresponding response time are bound and used as response behavior information from which the target information is removed.

In the embodiment of the present application, the binding of the target category and the dwell time may obtain: < static element, 2s >, < button interactive element, 0.5s >, < input box interactive element, 5s >, < static element, 3s >, upload advertisement operation server in chronological order, for example: { < static element, 2s >, < button interactive element, 0.5s >, < input box interactive element, 5s >, < static element, 3s > … }.

In the above embodiment, policy risk of future privacy information collection can be avoided by performing desensitization processing on response behavior information of a user.

The anti-cheating behavior identification method in the embodiment of the application can be realized by combining an artificial intelligence technology, and the specific realization mode is as follows: acquiring response behavior information of a target object aiming at the delivered advertisements and advertisement picture information of the delivered advertisements; and inputting the acquired response behavior information and the advertisement picture information into a trained anti-cheating model, and acquiring an anti-cheating behavior recognition result aiming at the target object and output by the anti-cheating model, wherein the trained anti-cheating model is obtained by machine learning training based on a training sample set. Wherein the training sample set includes a plurality of sets of data, and each set of data in the plurality of sets of data all includes:

first group data: the advertisement picture information of the sample advertisement and a first label, wherein the first label is used for marking whether the sample advertisement is a real sample advertisement or a cheating sample advertisement;

second-class group data: response behavior information of the sample object to the sample advertisement and a second label, wherein the second label is used for marking whether the sample object is a normal sample object or a cheating sample object.

Further, the anti-cheating model in the embodiment of the application is obtained by adopting multiple groups of data and utilizing machine learning training, based on the data, the anti-cheating model is trained into the anti-cheating model through machine learning and issued to the client side, and then data processing is carried out based on a machine learning framework on the client side.

In the embodiment of the application, when the local anti-cheating behavior is identified through the machine learning model, the identification efficiency and accuracy can be effectively improved.

After the anti-cheating behavior recognition method on the client side is introduced, the anti-cheating behavior recognition methods on the advertisement operation server side and the local anti-cheating rule file server side are briefly introduced as follows:

referring to fig. 5, an implementation flow chart of an anti-cheating behavior identification method provided in the embodiment of the present application is applied to a local anti-cheating rule file server, and a specific implementation flow of the method is as follows:

s51: receiving an advertisement request sent by a client, wherein the advertisement request is sent after the client responds to the operation of requesting the advertisement of a target object;

s52: according to hardware parameters in the advertisement request, a target anti-cheating rule matched with the hardware parameters is searched from a pre-configured anti-cheating rule set, and the target anti-cheating rule is issued to the client so that the client performs anti-cheating behavior identification on a target object according to the target anti-cheating rule, wherein the hardware parameters are hardware parameters of a terminal device where the client is located, the target anti-cheating rule comprises an advertisement display rule and a behavior rule, the advertisement display rule is used for analyzing advertisement picture information of released advertisements, and the behavior rule is used for analyzing response behavior information of the target object aiming at the released advertisements.

In the above embodiment, the local anti-cheating rule file server configures the set of anti-cheating rules, and continuously updates the anti-cheating rules corresponding to the hardware parameters, and in addition, the local anti-cheating rule file server does not need to receive the user response behavior information and the advertisement picture information uploaded by the client, but issues the target anti-cheating rules corresponding to the hardware parameters in the advertisement request sent by the client to the client, and migrates the anti-cheating gravity center from the server to the client, thereby effectively saving the overhead of resources related to the local anti-cheating rule file server. For a specific implementation manner, reference may be made to the above embodiments, and details are not repeated herein.

Referring to fig. 6, an implementation flow chart of an anti-cheating behavior recognition method provided in the embodiment of the present application is applied to an advertisement operation server, and a specific implementation flow of the method is as follows:

s61: receiving response behavior information which is sent by each client and corresponds to each object and is subjected to target information removal, and anti-cheating behavior identification results aiming at each object;

s62: and performing statistical analysis on the response behavior information after the target information is removed based on the anti-cheating behavior recognition result, determining a rule threshold value in the target anti-cheating rule, and respectively issuing the rule threshold value to the client side of each object, so that when the client side of each object determines that the rule threshold value in the pre-obtained target anti-cheating rule is different from the received rule threshold value, the rule threshold value in the pre-obtained target anti-cheating rule is replaced by the received rule threshold value, the target anti-cheating rule comprises an advertisement display rule and a behavior rule, the advertisement display rule is used for analyzing the advertisement picture information of the delivered advertisement, and the behavior rule is used for analyzing the response behavior information of the object aiming at the delivered advertisement.

In the above embodiment, when the client uploads the identification result, if the identification result indicates cheating, the client may further take the specific reason of the cheating, and upload response behavior information corresponding to the exposure, click, conversion, and the like, from which the target information is removed, to the advertisement operation server. The advertisement operation server can make anti-cheating behavior judgment on the side of the server according to the received anti-cheating behavior recognition result, can also perform statistical analysis on the received response behavior information according to the received anti-cheating behavior recognition result, for example, on the basis of a rule threshold corresponding to the retention time of the page in the statistical analysis, the original rule threshold is adjusted to 1.5 seconds for 1 second, and the newly adjusted rule threshold is issued to the client, and the client updates the threshold.

In the above embodiment, the advertisement operation server may further adjust the rule threshold in combination with the anti-cheating behavior recognition result uploaded by the client and the response behavior information of the user, and then issue the rule threshold to the client, and the client adjusts the corresponding target anti-cheating rule, so as to improve the accuracy of anti-cheating rule behavior recognition.

Fig. 7A is a schematic view of an interaction timing sequence of an anti-cheating behavior recognition method. The specific implementation flow of the method is as follows:

step S701: the client responds to the operation of requesting the advertisement by the user, and sends an advertisement request to the local anti-cheating rule file server, wherein the advertisement request comprises the hardware parameter of the terminal equipment where the client is located;

step S702: the local anti-cheating rule file server searches a target anti-cheating rule corresponding to the hardware parameter according to the hardware parameter in the advertisement request;

step S703: the local anti-cheating rule file server sends the target anti-cheating rule to the client;

step S704: the client analyzes the advertisement picture information of the target advertisement according to the advertisement display rule in the target anti-cheating rule to obtain a first identification result;

step S705: the client analyzes the response behavior information of the target object aiming at the target advertisement according to the behavior rules in the target anti-cheating rules to obtain a second identification result;

step S706: the client collects the first recognition result and the second recognition result to obtain an anti-cheating behavior recognition result aiming at the target object;

step S707: the client uploads the obtained anti-cheating behavior recognition result to an advertisement operation server;

step S708: desensitizing the response behavior information of the target object aiming at the target advertisement by the client to obtain desensitized response behavior information;

step S709: the client uploads the desensitized response behavior information to an advertisement operation server;

step S710: the advertisement operation server carries out statistical analysis on the response behavior information according to the received cheating behavior identification result, and determines a rule threshold value in the target anti-cheating rule;

step S711: the advertisement operation server issues the determined rule threshold to the client;

step S712: and when the client determines that the rule threshold value issued by the advertisement operation server is different from the rule threshold value in the local target anti-cheating rule, replacing the rule threshold value in the target anti-cheating rule with the rule threshold value issued by the advertisement operation server.

Fig. 7B is a schematic diagram of an interaction timing sequence of an anti-cheating behavior recognition method in another form according to an embodiment of the present application. The anti-cheating behavior recognition system in the embodiment of the application comprises: client, local anti-cheating rule file server and advertisement operation server.

When a user requests an advertisement, a client side pulls an anti-cheating rule from a local anti-cheating rule file server; after the anti-cheating rule is pulled, whether local updating is needed or not is judged, and if the local updating is needed, the local updating is carried out to the local storage. The process of comparing the version number of the target anti-cheating rule issued by the local anti-cheating rule file server with the version number of the local anti-cheating rule of the client listed in the above embodiment is to judge whether local updating is needed, and the process is not repeated.

In addition, the client side local anti-cheating module collects advertisement operation of the user, advertisement picture information and the like, wherein the advertisement operation of the user, namely data information related to response behaviors of the user to the delivered advertisements, including exposure, clicking and the like, is obtained through client side local behavior data collection, and the advertisement screenshot is obtained based on image processing and also comprises coordinates, basic data and the like obtained through image labeling and recognition.

Based on the information collected by the client local anti-cheating rule module, two parts of analysis can be executed: firstly, analyzing advertisement picture information based on advertisement display rules; and secondly, analyzing the response behavior information of the user based on the behavior rule. The specific analysis process can be seen in the above examples, and the limitation is not repeated here.

And finally, the client summarizes and processes the analysis results of the two parts and reports the analysis results to the advertisement operation server, and in addition, the client can also report the desensitization treatment to response behavior information of the user, such as advertisement operation and the like, to the advertisement operation server, and finally the advertisement operation server performs recording and statistical analysis.

In addition, the client side can also upload response behavior information such as advertisement operation after desensitization processing to a user interaction information database, the cloud side can make anti-cheating rules based on the information in the user interaction information database and store the anti-cheating rules in a local anti-cheating rule file server, and the anti-cheating rules in the local anti-cheating rule file server can also be derived from an anti-cheating strategy set obtained based on analysis of part of behavior data which is already run on line.

In the embodiment of the application, the policy risk of future privacy information collection can be avoided through desensitization processing on the response behavior information of the user; in addition, the anti-cheating gravity center is transferred from the server side to the client side through local screenshot comparison, local user behavior analysis and the like, and the expenditure of related resources of the server can be effectively saved.

Based on the same inventive concept, the embodiment of the application also provides an anti-cheating behavior recognition device. As shown in fig. 8, which is a schematic structural diagram of a first anti-cheating behavior recognition apparatus 800 in an embodiment of the present application, the apparatus may include:

an information obtaining unit 801, configured to obtain response behavior information of a target object to an advertisement delivered in a display page, and advertisement screen information of the delivered advertisement;

an analysis unit 802, configured to analyze the advertisement picture information according to an advertisement display rule in the pre-obtained target anti-cheating rule, and obtain a first recognition result indicating whether an advertisement that has been delivered meets the advertisement display rule; analyzing the response behavior information according to the behavior rules in the target anti-cheating rules to obtain a second identification result for indicating whether the response behavior of the target object meets the behavior rules or not;

the summarizing unit 803 is configured to obtain an anti-cheating behavior recognition result for the target object by summarizing the first recognition result and the second recognition result.

Optionally, the analysis unit 802 is specifically configured to:

and analyzing the picture proportion of the advertisement area in the released advertisement based on a target advertisement display rule corresponding to the advertisement type of the released advertisement in the advertisement display rule to obtain a first identification result.

the identifying unit 804 is further configured to:

acquiring target area information contained in the delivered advertisement;

the identifying unit 804 is specifically configured to:

and obtaining a first recognition result according to the analysis result of the target area information and the analysis result of the picture proportion of the advertisement area in the delivered advertisement.

Optionally, the response behavior information includes response information of the target object for various response behaviors of the delivered advertisement; the identifying unit 804 is specifically configured to:

and respectively comparing the response information corresponding to the various response behaviors with rule thresholds corresponding to the various response behaviors set in the behavior rule to obtain a second identification result for indicating whether the response behavior of the target object meets the behavior rule or not.

Optionally, the summarizing unit 803 is specifically configured to:

Optionally, the apparatus further comprises:

a transmission unit 805 configured to obtain a target anti-cheating rule based on the following:

responding to the operation of a target object for requesting the advertisement, and sending an advertisement request to a local anti-cheating rule file server, wherein the advertisement request comprises hardware parameters of a terminal device where a client is located;

and receiving a target anti-cheating rule matched with the hardware parameter, which is returned by the local anti-cheating rule file server.

Optionally, the summarizing unit 803 is further configured to:

Optionally, the summarizing unit 803 is specifically configured to:

classifying the response behavior information according to preset response categories, determining original categories to which various response behaviors belong, and recording response time corresponding to various response behaviors;

removing target information in the category to which each type of response behavior belongs according to an information removal rule to obtain abstract categories of each type of response behavior;

classifying and combining the abstract categories of various response behaviors with a preset operation set to obtain target categories of various response behaviors;

and binding the target types of the various response behaviors and the corresponding response time to obtain the response behavior information after the target information is removed.

Optionally, the apparatus further comprises:

an updating unit 806, configured to receive a rule threshold sent by the advertisement operation server;

and if the received rule threshold is different from the rule threshold in the target anti-cheating rule, replacing the rule threshold in the target anti-cheating rule with the received rule threshold, wherein the received rule threshold is obtained by carrying out statistical analysis on the response behavior information, corresponding to each object, of the advertisement operation server after the target information is removed according to the received anti-cheating behavior recognition result of each object.

Based on the same inventive concept, the embodiment of the application also provides an anti-cheating behavior recognition device. As shown in fig. 9, which is a schematic structural diagram of a second anti-cheating-behavior recognition apparatus 900 in the embodiment of the present application, the apparatus may include:

a first receiving unit 901, configured to receive an advertisement request sent by a client, where the advertisement request is sent by the client after an operation of requesting an advertisement in response to a target object;

the rule issuing unit 902 is configured to search a target anti-cheating rule matched with a hardware parameter from a pre-configured anti-cheating rule set according to the hardware parameter in the advertisement request, and issue the target anti-cheating rule to the client, so that the client performs anti-cheating behavior recognition on a target object according to the target anti-cheating rule, where the hardware parameter is a hardware parameter of a terminal device where the client is located, the target anti-cheating rule includes an advertisement display rule and a behavior rule, the advertisement display rule is used to analyze advertisement picture information of an advertisement that has been delivered, and the behavior rule is used to analyze response behavior information of the target object to the advertisement that has been delivered.

Based on the same inventive concept, the embodiment of the application also provides an anti-cheating behavior recognition device. As shown in fig. 10, which is a schematic structural diagram of a third anti-cheating-behavior recognition apparatus 1000 in the embodiment of the present application, the apparatus may include:

a second receiving unit 1001, configured to receive response behavior information, which is sent by each client and from which target information is removed, corresponding to each object, and an anti-cheating behavior recognition result for each object;

a statistical analysis unit 1002, configured to perform statistical analysis on the response behavior information after the target information is removed based on the anti-cheating behavior recognition result, determine a rule threshold in the target anti-cheating rule, and issue the rule threshold to the client of each object, so that when the rule threshold in the target anti-cheating rule obtained in advance is determined to be different from the received rule threshold, the client of each object replaces the rule threshold in the target anti-cheating rule obtained in advance with the received rule threshold, where the target anti-cheating rule includes an advertisement display rule and a behavior rule, the advertisement display rule is used to analyze advertisement picture information of the delivered advertisement, and the behavior rule is used to analyze response behavior information of the object for the delivered advertisement.

For convenience of description, the above parts are separately described as modules (or units) according to functional division. Of course, the functionality of the various modules (or units) may be implemented in the same one or more pieces of software or hardware when implementing the present application.

Having described the anti-cheating act recognition method and apparatus according to the exemplary embodiments of the present application, an electronic device according to another exemplary embodiment of the present application is described next.

As will be appreciated by one skilled in the art, aspects of the present application may be embodied as a system, method or program product. Accordingly, various aspects of the present application may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.

Fig. 11 is a block diagram illustrating an electronic device 1100 according to an example embodiment, the apparatus comprising:

a processor 1110;

a memory 1120 for storing instructions executable by the processor 1110;

wherein, the processor 1110 is configured to execute instructions to implement the anti-cheating behavior recognition method in the embodiment of the present application.

In an exemplary embodiment, a storage medium comprising instructions, such as the memory 1120 comprising instructions, executable by the processor 1110 of the electronic device 1100 to perform the method described above is also provided. Alternatively, the storage medium may be a non-transitory computer readable storage medium, for example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

In some possible implementations, the electronic device in the embodiments of the present application further includes a bus 1130 that connects the various system components, including the memory and the processor.

Bus 1130 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a processor, or a local bus using any of a variety of bus architectures.

The memory 1120 may include readable media in the form of volatile memory, such as Random Access Memory (RAM)1121 and/or cache memory 1122, and may further include Read Only Memory (ROM) 1123.

The memory 1120 may also include a program/utility 1125 having a set (at least one) of program modules 1124, including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.

The electronic device 1100 may also communicate with one or more external devices 1140 (e.g., keyboard, pointing device, etc.), with one or more devices that enable a user to interact with the electronic device 1100, and/or with any devices (e.g., router, modem, etc.) that enable the electronic device 1100 to communicate with one or more other electronic devices. Such communication may occur via an input/output (I/O) interface 1150. Also, the electronic device 1100 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the internet) via the network adapter 1160. As shown, the network adapter 1160 communicates with the other modules for the electronic device 1100 over the bus 1130. It should be understood that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device 1100, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

In some possible embodiments, the various aspects of the anti-cheating behavior recognition method provided by the present application may also be implemented in the form of a program product comprising program code for causing a computer device to perform the steps of the anti-cheating behavior recognition method according to various exemplary embodiments of the present application described above in this specification when the program product is run on the computer device, for example, the computer device may perform the steps as shown in fig. 2.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The program product of embodiments of the present application may employ a portable compact disc read only memory (CD-ROM) and include program code, and may be run on a computing device. However, the program product of the present application is not limited thereto, and in this document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with a command execution system, apparatus, or device.

While the preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all alterations and modifications as fall within the scope of the application.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

Claims

1. An anti-cheating behavior recognition method, comprising:

2. The method of claim 1, wherein the analyzing the advertisement picture information according to an advertisement display rule in pre-obtained target anti-cheating rules to obtain a first recognition result indicating whether the delivered advertisement meets the advertisement display rule specifically comprises:

3. The method of claim 2, wherein if the advertisement type of the served advertisement is a target advertisement type; the method further comprises the following steps:

acquiring target area information contained in the delivered advertisement;

the analyzing the picture proportion of the advertisement area in the released advertisement based on the target advertisement display rule corresponding to the advertisement type of the released advertisement in the advertisement display rule to obtain the first identification result specifically includes:

4. The method of claim 1, wherein the response behavior information comprises response information of various types of response behaviors of the target object with respect to the delivered advertisement; analyzing the response behavior information according to a behavior rule in the target anti-cheating rule to obtain a second recognition result used for indicating whether the response behavior of the target object meets the behavior rule, specifically comprising:

5. The method according to claim 1, wherein the obtaining of the anti-cheating behavior recognition result for the target object by summarizing the first recognition result and the second recognition result includes:

6. The method of claim 1, wherein the target anti-cheating rule is obtained based on:

7. The method of claim 1, wherein the method further comprises:

8. The method according to claim 7, wherein the removing target information in the response behavior information according to a preset information removal rule to obtain the response behavior information after the target information is removed specifically includes:

9. An anti-cheating behavior recognition method, comprising:

10. An anti-cheating behavior recognition method, comprising:

11. An anti-cheating behavior recognition apparatus, comprising:

12. An anti-cheating behavior recognition apparatus, comprising:

13. An anti-cheating behavior recognition apparatus, comprising:

14. An electronic device, comprising a processor and a memory, wherein the memory stores program code which, when executed by the processor, causes the processor to perform the steps of the method of any of claims 1 to 8 or the method of claim 9 or the method of claim 10.

15. A computer-readable storage medium, characterized in that it comprises program code for causing an electronic device to perform the steps of the method of any of claims 1-8 or the method of claim 9 or the method of claim 10, when said program code is run on the electronic device.