Advertisement putting method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides an advertisement putting method and device, electronic equipment and a storage medium, and relates to the technical field of Internet. Obtaining an advertisement putting request in a preset time period and related information of the advertisement putting request; determining IP regional distribution information of the client; for each IP address, counting a first request number and a second request number in a preset time period; for each first IP address, calculating a first region drift probability of the first IP address; determining the first IP address of which the first regional drift probability is greater than a second preset threshold as a risk IP address, and adding the risk IP address into a risk IP address list; in response to a target advertisement putting request sent by a target client, under the condition that the target IP address is a risk IP address, determining a target advertisement in advertisements without target IP regional orientation; and putting the target advertisement to the target client. By applying the embodiment of the invention, the cost of determining the risk IP address can be reduced.

Description

Technical Field

[0001] This invention relates to the field of Internet technology, and in particular to an advertising delivery method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the continuous development of internet technology, the application of the internet for commercial promotion is becoming increasingly widespread. Advertisers place advertisements through media, where advertisers are generally businesses promoting their own products and services; media outlets are advertising platforms that deliver advertisements to clients according to the advertisers' needs. A common billing method for internet advertising is based on the number of ad impressions. Due to the possibility of errors and fraud in data from media outlets alone, advertisers usually entrust third-party monitoring companies (hereinafter referred to as "third parties") to monitor the number of ad impressions placed on the media servers through third-party servers, and use the monitored number of impressions as the basis for final settlement.

[0003] When advertisers request that media outlets target specific regions for their ads (i.e., different ads are placed in different regions), after the ad is delivered to the client by the media server and triggered for exposure, various factors such as differences in deployment location and DNS (Domain Name System) settings between the media server and the third-party server, DNS resolution errors, policy routing, and hijacking can cause the IP (Internet Protocol) address of the client to be determined by the media server and the third-party server to be different. Consequently, the third-party server may determine that the region where the client is located is different from that of the media server, resulting in a regional discrepancy. Therefore, the third-party server will consider that the ad has not been delivered to the region requested by the advertiser and will not count the ad's exposure.

[0004] In related technologies, media outlets and third parties reduce IP address differences caused by policy-based routing by building shared data centers, thereby minimizing regional discrepancies. Furthermore, for the client's IP address carried in ad delivery requests, if the media server and the third-party server determine that the IP address belongs to a different region, it can be considered a risky IP address. Based on this, the media outlet can further identify easily shifting IP addresses by building a risky IP server within the shared data center. During ad delivery, these risky IP addresses are used to avoid region-targeted ads, thus reducing the regional statistical discrepancies between the media server and the third-party server.

[0005] However, risky IP servers need to be set up in the same data center, which increases the additional server deployment and maintenance costs, thus making it more expensive to identify risky IP addresses. Summary of the Invention

[0006] The purpose of this invention is to provide an advertising delivery method, apparatus, electronic device, and storage medium to reduce the cost of identifying risky IP addresses. The specific technical solution is as follows:

[0007] In a first aspect of this invention, an advertising delivery method is provided, applied to an advertising delivery server, the method comprising:

[0008] Obtain advertising delivery requests within a preset time period and related information of the advertising delivery requests, wherein the related information of the advertising delivery requests includes the IP address of the client that sent the advertising delivery requests;

[0009] Based on the IP address of the client that sent the ad delivery request, the IP geographic distribution information of the client is determined, wherein the IP geographic distribution information represents the percentage of times the client sends the ad delivery request in the corresponding IP geographic region;

[0010] For each IP address, the number of first requests to that IP address within the preset time period and the number of second requests to each client corresponding to that IP address are counted. The IP address with a first request count greater than a first preset threshold is referred to as the first IP address.

[0011] For each first IP address, based on the first number of requests to the first IP address, the second number of requests to each client corresponding to the first IP address, and the IP geographic distribution information of each client corresponding to the first IP address, the first geographic drift probability of the first IP address is calculated, wherein the first geographic drift probability is: the probability that the client corresponding to the first IP address is not in the first IP region, and the first IP region is the IP region to which the first IP address belongs;

[0012] The first IP address whose first regional drift probability is greater than the second preset threshold is identified as a risky IP address and added to the risky IP address list.

[0013] In response to a target ad delivery request sent by a target client, if the target IP address is a risky IP address in the risky IP address list, the target ad is determined from ads without target IP region targeting, wherein the target IP address is the IP address carried in the target ad delivery request, and the target IP region is the IP region to which the target IP address belongs;

[0014] The target advertisement is delivered to the target client.

[0015] In one possible embodiment, determining the IP geographic distribution information of the client based on the IP address of the client that sent the advertising delivery request includes:

[0016] The number of times each client sends an ad delivery request within the preset time period is counted to obtain the third request count;

[0017] Determine the IP region to which the IP address of the client that sent the ad delivery request belongs;

[0018] The number of times the client sends advertising delivery requests in each IP region within the preset time period is counted to obtain the fourth request count;

[0019] Based on the third and fourth request counts, the percentage of times the client sends the advertising delivery request in each IP region is calculated to obtain the client's IP region distribution information.

[0020] In one possible embodiment, calculating the first geographic drift probability of each first IP address based on the first number of requests to the first IP address, the second number of requests to each client corresponding to the first IP address, and the geographic distribution information of the IP addresses of each client corresponding to the first IP address includes:

[0021] For each of the first IP addresses, obtain the first IP region to which the first IP address belongs, and the IP region distribution information of the clients corresponding to the first IP address;

[0022] Based on the number of second requests of each client corresponding to the first IP address and the IP geographic distribution information of each client corresponding to the first IP address, the mathematical expectation of the first geographic drift of each client corresponding to the first IP address is calculated, wherein the mathematical expectation of the first geographic drift is: the mathematical expectation that the client corresponding to the first IP address is not in the first IP region;

[0023] Based on the mathematical expectation of the first regional drift of each client corresponding to the first IP address, and the first number of requests to the first IP address, the probability of the first regional drift of the first IP address is calculated.

[0024] In one possible embodiment, calculating the mathematical expectation of the first regional drift of each client corresponding to the first IP address, based on the second request count of each client corresponding to the first IP address and the IP geographic distribution information of the clients corresponding to the first IP address, includes:

[0025] For each client corresponding to the first IP address, with 1 as the minuend and the proportion of times the client sends the advertising request in the first IP region as the subtrahend, calculate the difference between 1 and the proportion of times the client sends the advertising request in the first IP region, and use this difference as the first difference.

[0026] Based on the client's second request count and the first difference, calculate the product of the client's second request count and the first difference as the mathematical expectation of the client's first regional drift;

[0027] The mathematical expectation of the first regional drift for each client corresponding to the first IP address is obtained by summing the results.

[0028] The calculation of the first-region drift probability of the first IP address based on the mathematical expectation of the first-region drift of each client corresponding to the first IP address and the first number of requests to the first IP address includes:

[0029] The mathematical expectation of the first regional drift of each client corresponding to the first IP address is summed to obtain the mathematical expectation of the first regional drift of the first IP address, which is used as the first mathematical expectation;

[0030] Using the first expected value as the dividend and the first number of requests from the first IP address as the divisor, the quotient of the first expected value and the first number of requests from the first IP address is calculated and used as the first regional drift probability of the first IP address.

[0031] In one possible embodiment, after determining the first IP address with a first geographical drift probability greater than a second preset threshold as a risky IP address, the method further includes:

[0032] Based on the first number of requests to the risky IP address, the second number of requests to each client corresponding to the risky IP address, and the IP geographic distribution information of each client corresponding to the risky IP address, the second geographic drift probability of the risky IP address is calculated. The second geographic drift probability is the probability of the client corresponding to the risky IP address in each IP geographic region, and each second geographic drift probability corresponds to one IP geographic region.

[0033] Compare the magnitudes of the drift probabilities of each second region, and determine the IP region corresponding to the one with the highest drift probability and a drift probability greater than a third preset threshold as the correction region corresponding to the risky IP address;

[0034] The risky IP addresses whose correction regions have been identified are removed from the list of risky IP addresses, resulting in a second IP address.

[0035] The method further includes:

[0036] In response to a target ad delivery request sent by a target client, if the target IP address carried in the target ad delivery request is the second IP address, the target ad is determined from the ads with corrected geographic targeting corresponding to the target IP address.

[0037] In one possible embodiment, calculating the second regional drift probability of the risky IP address based on the first number of requests to the risky IP address, the second number of requests to each client corresponding to the risky IP address, and the IP geographic distribution information of each client corresponding to the risky IP address includes:

[0038] Obtain the number of second requests for each client corresponding to the risky IP address, and the percentage of times each client corresponding to the risky IP address sends the advertising request in the corresponding IP region;

[0039] Based on the number of second requests of each client corresponding to the risky IP address, and the percentage of times each client corresponding to the risky IP address sends the advertising request in the corresponding IP region, the mathematical expectation of the second region drift of each client corresponding to the risky IP address is calculated. The mathematical expectation of the second region drift is: the mathematical expectation of the client corresponding to the risky IP address in each IP region, and each mathematical expectation of the second region drift corresponds to one IP region.

[0040] Based on the mathematical expectation of the second-region drift of each client corresponding to the risky IP address, and the first request number of the risky IP address, calculate the second-region drift probability of each risky IP address.

[0041] In one possible embodiment, calculating the expected value of the second-region drift of each client corresponding to the risky IP address, based on the number of second requests from each client corresponding to the risky IP address and the percentage of times each client sent the advertising request in the corresponding IP region, includes:

[0042] For each client corresponding to the risky IP address, based on the client's second request count and the proportion of times the client sends the advertising request in the corresponding IP region, the product of the client's second request count and the proportion of times the client sends the advertising request in the corresponding IP region is calculated as the mathematical expectation of the client's second region drift.

[0043] The mathematical expectation of the second-region drift of each client corresponding to the risky IP address is obtained by summarizing the results.

[0044] The calculation of the second-region drift probability of the risky IP address based on the mathematical expectation of the second-region drift of each client corresponding to the risky IP address and the first number of requests to the risky IP address includes:

[0045] The mathematical expectation of the second regional drift of each client corresponding to the risky IP address is summed to obtain the mathematical expectation of the second regional drift of the risky IP address, which is used as the second mathematical expectation;

[0046] Using the second expected value as the dividend and the first number of requests from the risky IP address as the divisor, the quotient of the second expected value and the first number of requests from the risky IP address is calculated, and this quotient is used as the second regional drift probability of the risky IP address.

[0047] In a second aspect of the invention, an advertising delivery device is also provided, applied to an advertising delivery server, the device comprising:

[0048] The first acquisition module is used to acquire advertising delivery requests within a preset time period and related information of the advertising delivery requests, wherein the related information of the advertising delivery requests includes the IP address of the client that sent the advertising delivery requests;

[0049] The first determining module is used to determine the IP geographic distribution information of the client based on the IP address of the client that sent the advertising delivery request, wherein the IP geographic distribution information represents the percentage of times the client sent the advertising delivery request in the corresponding IP geographic region;

[0050] The first statistics module is used to count the number of first requests for each IP address within the preset time period, and the number of second requests for each client corresponding to the IP address, wherein the IP address with the number of first requests greater than the first preset threshold is called the first IP address;

[0051] The first calculation module is used to calculate the first regional drift probability of each first IP address based on the first number of requests to the first IP address, the second number of requests to each client corresponding to the first IP address, and the IP regional distribution information of each client corresponding to the first IP address. The first regional drift probability is the probability that the client corresponding to the first IP address is not in the first IP region, and the first IP region is the IP region to which the first IP address belongs.

[0052] The second determination module is used to determine the first IP address whose first regional drift probability is greater than the second preset threshold as a risky IP address and add it to the risky IP address list.

[0053] The third determination module is used to respond to the target advertisement delivery request sent by the target client, and determine the target advertisement from the advertisements without target IP region targeting if the target IP address is a risk IP address in the risk IP address list. The target IP address is the IP address carried in the target advertisement delivery request, and the target IP region is the IP region to which the target IP address belongs.

[0054] The target delivery module is used to deliver the target advertisement to the target client.

[0055] In one possible embodiment, the first determining module includes:

[0056] The first statistics submodule is used to count the number of times each client sends an advertising delivery request within the preset time period to obtain the third request count;

[0057] The first determining submodule is used to determine the IP region to which the IP address of the client that sent the advertising delivery request belongs;

[0058] The second statistics submodule is used to count the number of times the client sends advertising delivery requests in each IP region within the preset time period, and obtain the fourth request count;

[0059] The first calculation submodule is used to calculate the percentage of times the client sends the advertising delivery request in each IP region based on the third request number and the fourth request number, so as to obtain the IP region distribution information of the client.

[0060] In one possible embodiment, the first computing module includes:

[0061] The first acquisition submodule is used to acquire, for each first IP address, the first IP region to which the first IP address belongs, and the IP region distribution information of the clients corresponding to the first IP address.

[0062] The second calculation submodule is used to calculate the expected value of the first regional drift of each client corresponding to the first IP address based on the number of second requests of each client corresponding to the first IP address and the IP regional distribution information of each client corresponding to the first IP address. The expected value of the first regional drift is the expected value of the client corresponding to the first IP address not being in the first IP region.

[0063] The third calculation submodule is used to calculate the first regional drift probability of the first IP address based on the mathematical expectation of the first regional drift of each client corresponding to the first IP address and the first number of requests of the first IP address.

[0064] In one possible embodiment, the second computing submodule is specifically used for:

[0065] For each client corresponding to the first IP address, with 1 as the minuend and the proportion of times the client sends the advertising request in the first IP region as the subtrahend, calculate the difference between 1 and the proportion of times the client sends the advertising request in the first IP region, and use this difference as the first difference.

[0066] Based on the client's second request count and the first difference, calculate the product of the client's second request count and the first difference as the mathematical expectation of the client's first regional drift;

[0067] The mathematical expectation of the first regional drift for each client corresponding to the first IP address is obtained by summing the results.

[0068] The third calculation submodule is specifically used for:

[0069] The mathematical expectation of the first regional drift of each client corresponding to the first IP address is summed to obtain the mathematical expectation of the first regional drift of the first IP address, which is used as the first mathematical expectation;

[0070] Using the first expected value as the dividend and the first number of requests from the first IP address as the divisor, the quotient of the first expected value and the first number of requests from the first IP address is calculated and used as the first regional drift probability of the first IP address.

[0071] In one possible embodiment, the device further includes:

[0072] The second calculation module is used to calculate the second region drift probability of the risk IP address based on the first number of requests to the risk IP address, the second number of requests to each client corresponding to the risk IP address, and the IP region distribution information of each client corresponding to the risk IP address. The second region drift probability is the probability of the client corresponding to the risk IP address in each IP region, and each second region drift probability corresponds to one IP region.

[0073] The fourth determining module is used to compare the magnitude of the drift probability of each second region, and determine the IP region corresponding to the one with the largest drift probability of the second region and the second region drift probability greater than the third preset threshold as the correction region corresponding to the risky IP address.

[0074] The removal module is used to remove the risky IP addresses whose correction regions have been determined from the list of risky IP addresses, and obtain a second IP address;

[0075] The fifth determining module is used to respond to a target advertisement delivery request sent by the target client, and when the target IP address carried in the target advertisement delivery request is the second IP address, determine the target advertisement from the advertisements with corrected geographic targeting corresponding to the target IP address.

[0076] In one possible embodiment, the second computing module includes:

[0077] The second acquisition submodule is used to acquire the number of second requests of each client corresponding to the risky IP address, and the percentage of times each client corresponding to the risky IP address sends the advertising delivery request in the corresponding IP region;

[0078] The fourth calculation submodule is used to calculate the expected value of the second region drift of each client corresponding to the risky IP address based on the number of second requests of each client corresponding to the risky IP address and the proportion of times each client corresponding to the risky IP address sends the advertising delivery request in the corresponding IP region. The expected value of the second region drift is: the expected value of the client corresponding to the risky IP address in each IP region, and each expected value of the second region drift corresponds to one IP region.

[0079] The fifth calculation submodule is used to calculate the probability of each second-region drift of the risky IP address based on the mathematical expectation of the second-region drift of each client corresponding to the risky IP address and the first number of requests of the risky IP address.

[0080] In one possible embodiment, the fourth computing submodule is specifically used for:

[0081] For each client corresponding to the risky IP address, based on the client's second request count and the proportion of times the client sends the advertising request in the corresponding IP region, the product of the client's second request count and the proportion of times the client sends the advertising request in the corresponding IP region is calculated as the mathematical expectation of the client's second region drift.

[0082] The mathematical expectation of the second-region drift of each client corresponding to the risky IP address is obtained by summarizing the results.

[0083] The fifth calculation submodule is specifically used for:

[0084] The mathematical expectation of the second regional drift of each client corresponding to the risky IP address is summed to obtain the mathematical expectation of the second regional drift of the risky IP address, which is used as the second mathematical expectation;

[0085] Using the second expected value as the dividend and the first number of requests from the risky IP address as the divisor, the quotient of the second expected value and the first number of requests from the risky IP address is calculated, and this quotient is used as the second regional drift probability of the risky IP address.

[0086] In a third aspect of the present invention, an electronic device is also provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus.

[0087] Memory, used to store computer programs;

[0088] When a processor executes a program stored in memory, it implements any of the steps described in the first aspect.

[0089] In a fourth aspect of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, it implements any of the above-described advertising delivery methods.

[0090] In a fifth aspect of the invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the advertising delivery methods described above.

[0091] This invention provides an advertising delivery method, apparatus, electronic device, and storage medium. The method is applied to an advertising delivery server and includes: acquiring advertising delivery requests and related information of the advertising delivery requests within a preset time period, wherein the related information includes the IP address of the client sending the advertising delivery request; determining the IP geographic distribution information of the client based on the IP address of the client sending the advertising delivery request, wherein the IP geographic distribution information represents the proportion of times the client sends advertising delivery requests in the corresponding IP geographic region; for each IP address, counting the first request count of that IP address and the second request count of each client corresponding to that IP address within the preset time period, wherein IP addresses with a first request count greater than a first preset threshold are referred to as first IP addresses; for each first IP address, calculating the first request count of that first IP address and the second request count of each client corresponding to that first IP address... The system calculates the first region drift probability of the first IP address based on the number of IP addresses and the IP geographic distribution information of the clients corresponding to the first IP address. The first region drift probability is the probability that the client corresponding to the first IP address is not in the first IP region, where the first IP region is the IP region to which the first IP address belongs. First IP addresses with a first region drift probability greater than a second preset threshold are identified as risk IP addresses and added to the risk IP address list. In response to a target ad delivery request sent by the target client, if the target IP address is a risk IP address in the risk IP address list, a target ad is determined from ads without target IP region targeting, where the target IP address is the IP address carried in the target ad delivery request and the target IP region is the IP region to which the target IP address belongs. The target ad is then delivered to the target client.

[0092] This invention identifies risky IP addresses by statistically analyzing and processing ad delivery requests and related information. In other words, this invention relies solely on data from the ad delivery server to determine risky IP addresses, eliminating the need for a separate risk IP server and thus avoiding additional server deployment and maintenance costs, thereby reducing the overall cost of identifying risky IP addresses. When the target IP address carried in the ad delivery request sent by the target client is a risky IP address, delivering ads without targeted IP geographic location to the target client reduces the discrepancies in geographic statistics between the ad delivery server and third-party servers. Attached Figure Description

[0093] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0094] Figure 1 This is a schematic diagram of the first process of the advertising delivery method provided in an embodiment of the present invention;

[0095] Figure 2 This is a schematic diagram of a second process for the advertising delivery method provided in an embodiment of the present invention;

[0096] Figure 3 This is a schematic diagram of the third process of the advertising delivery method provided in the embodiments of the present invention;

[0097] Figure 4 This is a schematic diagram of the fourth process of the advertising delivery method provided in the embodiments of the present invention;

[0098] Figure 5 A schematic diagram of the fifth process of the advertising delivery method provided in the embodiments of the present invention;

[0099] Figure 6 A schematic diagram of the sixth process of the advertising delivery method provided in the embodiments of the present invention;

[0100] Figure 7 A schematic diagram of the seventh process of the advertising delivery method provided in the embodiments of the present invention;

[0101] Figure 8 This is a schematic diagram of the eighth process of the advertising delivery method provided in the embodiments of the present invention;

[0102] Figure 9 This is a schematic diagram of a first structure of an advertising delivery device provided in an embodiment of the present invention;

[0103] Figure 10 This is a schematic diagram of a second structure of the advertising delivery device provided in an embodiment of the present invention;

[0104] Figure 11 This is a schematic diagram of a third structure of the advertising delivery device provided in an embodiment of the present invention;

[0105] Figure 12 This is a schematic diagram of a fourth structure of the advertising delivery device provided in an embodiment of the present invention;

[0106] Figure 13 This is a fifth structural schematic diagram of the advertising delivery device provided in an embodiment of the present invention;

[0107] Figure 14 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0108] The technical solutions of the present invention will now be described with reference to the accompanying drawings in the embodiments of the present invention.

[0109] To address the problems existing in the prior art, embodiments of the present invention provide an advertising delivery method, apparatus, electronic device, and storage medium. The advertising delivery method provided by the embodiments of the present invention will be described first below.

[0110] The advertising delivery method provided in this embodiment of the invention is applied to an advertising delivery server, and the method includes:

[0111] Obtain advertising delivery requests within a preset time period and related information of the advertising delivery requests, wherein the related information of the advertising delivery requests includes the IP address of the client that sent the advertising delivery requests;

[0112] Based on the IP address of the client that sent the ad delivery request, the IP geographic distribution information of the client is determined, wherein the IP geographic distribution information represents the percentage of times the client sends the ad delivery request in the corresponding IP geographic region;

[0113] For each IP address, the number of first requests to that IP address within the preset time period and the number of second requests to each client corresponding to that IP address are counted. The IP address with a first request count greater than a first preset threshold is referred to as the first IP address.

[0114] For each first IP address, based on the first number of requests to the first IP address, the second number of requests to each client corresponding to the first IP address, and the IP geographic distribution information of each client corresponding to the first IP address, the first geographic drift probability of the first IP address is calculated, wherein the first geographic drift probability is: the probability that the client corresponding to the first IP address is not in the first IP region, and the first IP region is the IP region to which the first IP address belongs;

[0115] The first IP address whose first regional drift probability is greater than the second preset threshold is identified as a risky IP address and added to the risky IP address list.

[0116] In response to a target ad delivery request sent by a target client, if the target IP address is a risky IP address in the risky IP address list, the target ad is determined from ads without target IP region targeting, wherein the target IP address is the IP address carried in the target ad delivery request, and the target IP region is the IP region to which the target IP address belongs;

[0117] The target advertisement is delivered to the target client.

[0118] This invention identifies risky IP addresses by statistically analyzing and processing ad delivery requests and related information. In other words, this invention relies solely on data from the ad delivery server to determine risky IP addresses, eliminating the need for a separate risk IP server and thus avoiding additional server deployment and maintenance costs, thereby reducing the overall cost of identifying risky IP addresses. When the target IP address carried in the ad delivery request sent by the target client is a risky IP address, delivering ads without targeted IP geographic location to the target client reduces the discrepancies in geographic statistics between the ad delivery server and third-party servers.

[0119] The following is a detailed explanation.

[0120] See Figure 1 This is a schematic diagram of the first type of advertising delivery method provided in the embodiment of the present invention. It is applied to an advertising delivery server, wherein the advertising delivery server is an advertising delivery server in an advertising delivery platform. The method includes the following steps S101-S107.

[0121] Step S101: Obtain the ad delivery requests and related information for the ad delivery requests within the preset time period.

[0122] The relevant information for the ad delivery request includes the IP address of the client that sent the ad delivery request.

[0123] It is understandable that the ad delivery request is sent by the client (user). According to existing communication protocols, the ad delivery request contains the IP address of the sender, which is the client's IP address.

[0124] When the ad delivery server receives an ad delivery request from a client, it records the current timestamp, IP address, DC-code (data center code), and client information. The DC-code represents the data center where the ad delivery server that received the client request is located, which can be understood as a data center. It records relevant information about the ad delivery request. The client information refers to a unique client ID (Identity Document) obtained by combining OAI (Open Anonymous Device Identifier), IDFA (Identifier for Advertising), cookies (small text files), etc. All of this information can be used for ad delivery requests. In addition, the ad delivery server will also resolve the IP address to the corresponding IP region based on the IP database provided by the advertising association.

[0125] In step S101, it is necessary to obtain the advertising delivery requests and related information for the advertising delivery requests within a preset time period. The preset time period can be set according to actual needs. For example, the preset time period can be set to the period closest to the current time. For instance, the preset time period can be set to the past 5 hours, the past 12 hours, the past 24 hours, etc.

[0126] Step S102: Determine the IP geographic distribution information of the client based on the IP address of the client that sent the ad delivery request.

[0127] Among them, IP geographic distribution information indicates the percentage of times the client sends advertising delivery requests in the corresponding IP geographic region.

[0128] Understandably, within a preset time period, the number of ad delivery requests sent by the client is usually multiple. Furthermore, due to factors such as the client's variable location, the same client may send ad delivery requests from multiple geographical regions within the preset time period. Therefore, in step S102, it is necessary to determine the client's IP geographical distribution information based on the IP address of the client sending the ad delivery request. Understandably, the percentage of ad delivery requests sent by the client from each IP region indicates the probability of the client appearing in each IP region.

[0129] Step S103: For each IP address, count the number of first requests for that IP address within a preset time period, and the number of second requests for each client corresponding to that IP address.

[0130] The IP address whose first request count is greater than the first preset threshold is referred to as the first IP address.

[0131] Understandably, within a preset time period, the total number of ad delivery requests on an IP address is usually multiple, and these requests may originate from multiple clients. In step S103, for each IP address, it is necessary to count the first number of requests for that IP address within the preset time period, as well as the second number of requests from each client corresponding to that IP address. In one example, the preset time period is the past 24 hours. In the past 24 hours, the total number of requests (i.e., the first number of requests) on a certain IP address is 100. Among these, client A made 20 second requests on that IP address, client B made 30 second requests on that IP address, client C made 15 second requests on that IP address, and client D made 35 second requests on that IP address.

[0132] Furthermore, to ensure the accuracy of the statistical results, IP addresses with low request counts need to be filtered out, and IP addresses with a first request count greater than a first preset threshold are referred to as the first IP address. The first preset threshold can be set according to the actual access volume. When the actual access volume is high, the first preset threshold can be set to a larger value, and when the actual access volume is low, the first preset threshold can be set to a smaller value. This invention does not impose specific limitations on this. For example, the first preset threshold can be set to 50 times, 100 times, 200 times, etc.

[0133] Step S104: For each first IP address, calculate the first regional drift probability of the first IP address based on the first number of requests to the first IP address, the second number of requests to each client corresponding to the first IP address, and the IP regional distribution information of each client corresponding to the first IP address.

[0134] The first region drift probability is the probability that the client corresponding to the first IP address is not in the first IP region, where the first IP region is the IP region to which the first IP address belongs.

[0135] The IP region to which the first IP address belongs can be obtained through the IP database provided by the Advertising Association, which records the correspondence between IP addresses and IP regions.

[0136] Step S105: Identify the first IP address whose first regional drift probability is greater than the second preset threshold as a risky IP address and add it to the risky IP address list.

[0137] In step S105, after obtaining the first regional drift probability of each first IP address, the first IP addresses with a first regional drift probability greater than a second preset threshold are identified as risky IP addresses and added to the risky IP address list. The second preset threshold can be set manually based on experience or according to actual circumstances; this invention does not impose specific limitations on it.

[0138] Step S106: In response to the target client's target ad delivery request, if the target IP address is a risky IP address in the risky IP address list, identify the target ad among ads without target IP geographic targeting.

[0139] Among them, the target IP address is the IP address carried in the target ad delivery request, and the target IP region is the IP region to which the target IP address belongs.

[0140] After identifying the risky IP address in the aforementioned steps, when a target ad delivery request is received from the target client, if the target IP address it carries is a risky IP address, the target ad will not be selected from the ads targeted by the target IP region, but rather from the ads that are not targeted by the target IP region.

[0141] Step S107: Targeted advertisements are delivered to the target client.

[0142] In this embodiment of the invention, since the target advertisement is determined from advertisements without target IP region targeting, if the target IP address carried in the target advertisement delivery request is a risky IP address, the delivered target advertisement will be a non-risky IP region targeted advertisement, meaning that risky IP region targeted advertisements will not be delivered. This can reduce the regional statistical differences between the advertisement delivery server (media party) and the third-party server, and reduce losses caused by the regional differences between the media and the third party.

[0143] By applying this embodiment of the invention, risky IP addresses are determined through data statistics and analysis of advertising delivery requests and related information. In other words, this embodiment of the invention uses data from the advertising delivery server; the risky IP address can be determined solely by the advertising delivery server, without relying on data from third-party servers. Therefore, there is no need to build a separate risky IP server, thus avoiding additional server deployment and maintenance costs and reducing the cost of determining risky IP addresses. When the target IP address carried in the target advertising delivery request sent by the target client is a risky IP address, delivering ads without target IP geographic targeting to the target client can reduce the geographic statistical discrepancies between the advertising delivery server and third-party servers. Furthermore, in related technologies, building a risky IP service in a data center of the same origin is less important than advertising delivery services and log collection services, and its maintenance priority is also weaker. Stability can affect the generation of risky IP addresses. However, this embodiment of the invention uses data collected from the advertising delivery server, which is highly important and stable, thus improving the generation rate of risky IP addresses.

[0144] In one possible embodiment, see Figure 2 This is a schematic diagram of the second process of the advertising delivery method provided in the embodiments of the present invention. Figure 1 Compared to the embodiment shown, step S102 can be implemented by the following steps S102A-S102D.

[0145] Step S102A: Count the number of times each client sends an ad delivery request within a preset time period to obtain the third request count.

[0146] It is understandable that within a preset time period, the ad delivery server will usually receive ad delivery requests from multiple clients. In step S102A, it is necessary to count the number of ad delivery requests sent by each client within the preset time period to obtain the number of third requests for each client.

[0147] Step S102B: Determine the IP region to which the IP address of the client that sent the ad delivery request belongs.

[0148] In step S102B, the IP address of the client that sent the advertising placement request can be resolved using the IP database provided by the advertising association to obtain the IP region to which the IP address of the client that sent the advertising placement request belongs.

[0149] Step S102C: Count the number of times the client sends advertising delivery requests in each IP region within the preset time period to obtain the fourth request count.

[0150] It is understandable that, within a preset time period, due to factors such as the client's location not being fixed, the same client may send advertising requests to multiple regions within that time period. In step S102C, it is necessary to count the number of times the client sends advertising requests to each IP region within the preset time period, thus obtaining the fourth request count of the client sending advertising requests to each IP region.

[0151] Step S102D: Based on the number of third and fourth requests, calculate the percentage of times the client sends advertising delivery requests in each IP region to obtain the client's IP region distribution information.

[0152] For example, the preset time period is the past 24 hours. In the past 24 hours, the total number of advertising requests sent by client A (i.e., the third request count) is 50. The IP addresses obtained are identified as belonging to 4 IP regions, namely region a, region b, region c, and region d. The fourth request counts of client A sending advertising requests in each IP region are as follows: 8 advertising requests in region a, 15 advertising requests in region b, 10 advertising requests in region c, and 17 advertising requests in region d. Therefore, the IP region distribution information of client A obtained based on the third and fourth request counts is as follows: the proportion of advertising requests sent in region a is 16%, the proportion of advertising requests sent in region b is 30%, the proportion of advertising requests sent in region c is 20%, and the proportion of advertising requests sent in region d is 34%. It should be noted that the purpose of this example is only to explain the solution of the present invention and make it clear. In practical applications, the number of requests obtained within the preset time period should be based on the actual number. The present invention does not make any specific limitations in this regard.

[0153] By applying the embodiments of the present invention, by separately counting the number of third requests sent by each client to place advertisements within a set time period, and the number of fourth requests sent by each client to place advertisements in each IP region, the IP region distribution information of each client can be accurately obtained. This IP region distribution information can represent the probability of a client appearing in each region, providing conditions for subsequently identifying risky IP addresses.

[0154] In one possible embodiment, see Figure 3 This is a schematic diagram of the third process of the advertising delivery method provided in the embodiments of the present invention. Figure 1 Compared to the embodiment shown, step S104 can be implemented by the following steps S104A-S104C.

[0155] Step S104A: For each first IP address, obtain the first IP region to which the first IP address belongs, and the IP region distribution information of the clients corresponding to the first IP address.

[0156] In the preceding steps, the advertising server has already resolved the IP regions corresponding to each IP address based on the IP database provided by the advertising association, and determined the IP region distribution information of each client. In step S104A, it is only necessary to obtain the first IP region to which each first IP address belongs, and the IP region distribution information of each client corresponding to that first IP address.

[0157] Step S104B: Based on the number of second requests of each client corresponding to the first IP address and the IP geographic distribution information of each client corresponding to the first IP address, calculate the mathematical expectation of the first geographic drift of each client corresponding to the first IP address.

[0158] In one possible embodiment, see Figure 4 This is a schematic diagram of the fourth process of the advertising delivery method provided in the embodiments of the present invention. Figure 3 Compared to the embodiment shown, step S104B can be implemented by the following steps S104B1-S104B3.

[0159] Step S104B1: For each client corresponding to the first IP address, take 1 as the minuend and the proportion of times the client sends advertising requests in the first IP region as the subtrahend. Calculate the difference between 1 and the proportion of times the client sends advertising requests in the first IP region, and use this difference as the first difference.

[0160] Step S104B2: Based on the second number of requests and the first difference of the client, calculate the product of the second number of requests and the first difference of the client as the mathematical expectation of the first regional drift of the client.

[0161] Step S104B3: Summarize the mathematical expectations of the first regional drift of each client corresponding to the first IP address.

[0162] The mathematical expectation of the first region drift is: the mathematical expectation that the client corresponding to the first IP address is not in the first IP region.

[0163] Step S104C: Based on the mathematical expectation of the first regional drift of each client corresponding to the first IP address and the first number of requests for the first IP address, calculate the first regional drift probability of the first IP address.

[0164] exist Figure 4In the embodiment shown, step S104C can be implemented by steps S104C1 and S104C2.

[0165] Step S104C1: Sum the mathematical expectations of the first regional drift of each client corresponding to the first IP address to obtain the mathematical expectation of the first regional drift of the first IP address, which is used as the first mathematical expectation.

[0166] Step S104C2: Using the first mathematical expectation as the dividend and the first request number of the first IP address as the divisor, calculate the quotient of the first mathematical expectation and the first request number of the first IP address, and use it as the first regional drift probability of the first IP address.

[0167] For example, given a first IP address A, there are two clients corresponding to the first IP address A, namely client a and client b. Based on the aforementioned step S103, the first request count for the first IP address A is 50. For client a corresponding to the first IP address A, based on the aforementioned step S102, the percentage of times client a sends advertising requests in the first IP region is 20%, and based on the aforementioned step S103, the second request count for client a in the first IP address A is 30. Therefore, the first difference for client a is: 1 - 20% = 80%, and the expected value of the first region drift for client a is: 30 × 80% = 24 times. For client b corresponding to the first IP address A, based on the aforementioned step S102, the percentage of times client b sends advertising requests in the first IP region is 40%, and based on the aforementioned step S103, the second request count for client b in the first IP address A is 20. Therefore, the first difference for client b is: 1 - 40% = 60%, and the expected value of the first region drift for client b is: 20 × 60% = 12 times. The expected first-region migration times for each client corresponding to the first IP address A are: 24 migration times for client a and 12 migration times for client b. Further, summing the expected first-region migration times for clients a and b yields the expected first-region migration times for the first IP address A: 24 + 12 = 36 migration times. Therefore, the probability of first-region migration for the first IP address A is 36 / 50 = 72%. Based on this, if the second preset threshold is set to 60%, the probability of first-region migration for the first IP address A is greater than the second preset threshold. Therefore, the first IP address A can be identified as a risky IP address.

[0168] It should be noted that the above examples are only intended to explain the solution of the present invention and make it clear. In practical applications, the number of clients corresponding to the first IP address and the number of requests obtained within the preset time period should be based on the actual situation. The present invention does not make any specific limitations in this regard.

[0169] By applying the embodiments of the present invention, for each first IP address, data analysis and calculation are performed on the second request count of each client corresponding to the first IP address and the IP geographic distribution information of each client corresponding to the first IP address. This allows for the mathematical expectation of the first geographic drift of each client corresponding to the first IP address. Furthermore, by using the mathematical expectation of the first geographic drift of each client corresponding to the first IP address and the first request count of the first IP address, the first geographic drift probability of the first IP address can be calculated, which is the probability that the client corresponding to the first IP address is not in the first IP region. By comparing this probability with a second preset threshold, risky IP addresses can be accurately identified without the need for additional server deployment, thus reducing the cost of identifying risky IP addresses.

[0170] In one possible embodiment, see Figure 5 This is a schematic diagram of the fifth process of the advertising delivery method provided in the embodiments of the present invention, and... Figure 1 Compared to the illustrated embodiment, the steps S108-S110 are also included.

[0171] Step S108: Based on the first number of requests for the risky IP address, the second number of requests for each client corresponding to the risky IP address, and the IP geographic distribution information of each client corresponding to the risky IP address, calculate the second geographic drift probability of each risky IP address.

[0172] The second region drift probability is the probability of the client corresponding to the risky IP address being in each IP region, with each second region drift probability corresponding to one IP region.

[0173] It is understandable that there may be multiple clients corresponding to a risky IP address, and each client may send advertising requests in multiple regions within a preset time period due to reasons such as the client's location not being fixed. In step S108, it is necessary to calculate the second regional drift probability of each risky IP address based on the first number of requests from the risky IP address, the second number of requests from each client corresponding to the risky IP address, and the IP regional distribution information of each client corresponding to the risky IP address.

[0174] Step S109: Compare the magnitudes of the drift probabilities of each second region, and determine the IP region corresponding to the one with the highest drift probability and a second region drift probability greater than the third preset threshold as the correction region corresponding to the risky IP address.

[0175] In step S109, after obtaining the drift probability of each second region for the risky IP address, the magnitude of each second region drift probability is compared, and the IP region corresponding to the one with the largest second region drift probability and the second region drift probability greater than the third preset threshold is determined as the correction region corresponding to the risky IP address. The magnitude of the third preset threshold can be manually set based on experience or based on actual conditions. This invention does not impose specific limitations on this.

[0176] Step S110: Remove the risky IP addresses whose correction regions have been determined from the risky IP address list to obtain the second IP address.

[0177] After determining the correction region corresponding to the risky IP address, the risky IP address is removed from the list of risky IP addresses. At this point, it is no longer a risky IP address, but a secondary IP address.

[0178] In this embodiment of the invention, the following step S111 is also included.

[0179] Step S111: In response to the target ad delivery request sent by the target client, if the target IP address is the second IP address, determine the target ad in the ads with corrected geographic targeting corresponding to the target IP address.

[0180] The target IP address is the IP address carried in the target ad delivery request.

[0181] After identifying the risky IP addresses and their corresponding correction regions in the aforementioned steps, the risky IP addresses with their correction regions identified are removed from the risky IP address list, resulting in the second IP address. When a target ad delivery request is received from the target client, and the target IP address it carries is the second IP address, the target ad is selected from the ads targeted by the correction region corresponding to that target IP address.

[0182] Therefore, when the target IP address carried in the target ad delivery request is the second IP address, the delivered target ad is the corrected geographic targeting ad corresponding to the target IP address. This can improve the inventory utilization rate of the ad and reduce inventory loss by reducing the geographic statistical differences between the ad delivery server (media party) and the third-party server.

[0183] By applying the embodiments of the present invention, after identifying a risky IP address, the corrective region for that risky IP address can be determined through data statistics and analysis of advertising delivery requests and related information. In other words, the embodiments of the present invention, based on data from the advertising delivery server, can determine the corrective region for a risky IP address solely through the advertising delivery server, without relying on data from a third-party server. Therefore, there is no need to build a separate risky IP server, thus avoiding additional server deployment and maintenance costs and reducing the cost of determining the corrective region for a risky IP address.

[0184] In one possible embodiment, see Figure 6 This is a schematic diagram of the sixth process of the advertising delivery method provided in the embodiments of the present invention, and... Figure 5 Compared to the embodiment shown, step S108 can be implemented by the following steps S108A-S108C.

[0185] Step S108A: Obtain the number of second requests for each client corresponding to the risky IP address, and the percentage of times each client corresponding to the risky IP address sends advertising requests in the corresponding IP region.

[0186] In the preceding steps, the ad delivery server has resolved the IP regions corresponding to each IP address based on the IP database provided by the advertising association, determined the IP region distribution information of each client, and counted the number of second requests for each client corresponding to each IP address. In step S108A, it is only necessary to obtain the number of second requests for each client corresponding to the risky IP address, and the percentage of times each client corresponding to the risky IP address sends ad delivery requests in the corresponding IP region.

[0187] Step S108B: Based on the number of second requests of each client corresponding to the risky IP address and the percentage of times each client corresponding to the risky IP address sends advertising requests in the corresponding IP region, calculate the expected value of the second region drift of each client corresponding to the risky IP address.

[0188] The mathematical expectation of the second region drift is: the mathematical expectation of the client corresponding to the risky IP address in each IP region, and each mathematical expectation of the second region drift corresponds to one IP region.

[0189] In one possible embodiment, see Figure 7 This is a schematic diagram of the seventh process of the advertising delivery method provided in the embodiments of the present invention, and... Figure 6 Compared to the embodiment shown, step S108B can be implemented by the following steps S108B1 and S108B2.

[0190] Step S108B1: For each client corresponding to a risky IP address, based on the number of second requests of that client and the proportion of times that client sends advertising requests in the corresponding IP region, calculate the product of the number of second requests of that client and the proportion of times that client sends advertising requests in the corresponding IP region, and use it as the mathematical expectation of the client's second region drift.

[0191] Step S108B2: Summarize the mathematical expectations of the second-region drift of each client corresponding to the risky IP address.

[0192] The mathematical expectation of the second region drift is: the mathematical expectation of the client corresponding to the risky IP address in each IP region, and each mathematical expectation of the second region drift corresponds to one IP region.

[0193] Step S108C: Based on the mathematical expectation of the second-region drift of each client corresponding to the risky IP address and the first request number of the risky IP address, calculate the second-region drift probability of each risky IP address.

[0194] exist Figure 7 In the embodiment shown, step S108C can be implemented by steps S108C1 and S108C2.

[0195] Step S108C1: Sum the mathematical expectations of the second regional drift of each client corresponding to the risky IP address to obtain the mathematical expectation of the second regional drift of the risky IP address, which is used as the second mathematical expectation.

[0196] Step S108C2: Using the second mathematical expectation as the dividend and the first request number of the risky IP address as the divisor, calculate the quotient of the second mathematical expectation and the first request number of the risky IP address, which is used as the second regional drift probability of the risky IP address.

[0197] For example, given a risky IP address A, there are two clients corresponding to risky IP address A, namely client c and client d. Based on the aforementioned step S103, the first request count for risky IP address A is 100. For client a corresponding to risky IP address A, based on the aforementioned step S102, the percentage of times client c sends advertising requests in IP region 1 is 40%, and the percentage of times client c sends advertising requests in IP region 2 is 60%. Based on the aforementioned step S103, the second request count for client c at risky IP address A is 60. Therefore, the expected number of times client c drifts in the second region corresponding to IP region 1 is 60 × 40% = 24 times, and the expected number of times client c drifts in the second region corresponding to IP region 2 is 60 × 60% = 36 times. For client d corresponding to risky IP address A, based on the aforementioned step S102, the percentage of times client d sends advertising requests in IP region 3 is 50%, and the percentage of times client d sends advertising requests in IP region 4 is 50%. Based on the aforementioned step S103, the number of second requests by client d to risky IP address A is 80. Therefore, the expected number of second-region drifts for client d in IP region 2 is 80 × 50% = 40 times, and the expected number of second-region drifts for client d in IP region 3 is also 80 × 50% = 40 times. Therefore, the summed expected numbers of second-region drifts for each client corresponding to risky IP address A are: client c in IP region 1 has an expected number of 24 drifts, client c in IP region 2 has an expected number of 36 drifts, client d in IP region 2 has an expected number of 40 drifts, and client d in IP region 3 has an expected number of 40 drifts. Furthermore, by summing the expected values ​​of the second-region drift for clients c and d, the expected value of the risky IP address A is obtained as follows: the expected value in IP region 1 is 24 times, in IP region 2 is 76 times, and in IP region 3 is 40 times. Therefore, the probability of the risky IP address A drifting in the second region is: 24 / 100 = 24% in IP region 1, 76 / 100 = 76% in IP region 2, and 40 / 100 = 40% in IP region 3. Based on this, if the third preset threshold is set to 50%, the highest probability of the risky IP address A drifting in the second region is 76%, which is greater than the third preset threshold. Therefore, IP region 2, corresponding to a second-region drift probability of 76%, can be determined as the correction region for the risky IP address A.

[0198] It should be noted that the above examples are only intended to explain the solution of the present invention and make it clear. In practical applications, the number of clients corresponding to risky IP addresses, the geographical distribution of each client, and the number of requests obtained within a preset time period should all be based on the actual situation. The present invention does not impose any specific limitations on these aspects.

[0199] By applying the embodiments of the present invention, through data analysis and calculation of the number of second requests of each client corresponding to a risky IP address and the proportion of times each client corresponding to a risky IP address sends advertising requests in the corresponding IP region, the expected value of the second region drift of each client corresponding to the risky IP address can be obtained. Furthermore, by using the expected value of the second region drift of each client corresponding to the risky IP address and the number of first requests of the risky IP address, the probability of the second region drift of the risky IP address can be calculated, which is the probability of the client corresponding to the risky IP address in each IP region. By comparing the magnitude of each second region drift probability and comparing the second region drift probability with the largest value with a third preset threshold, the correction region of the risky IP address can be accurately determined without the need for additional server deployment, thus reducing the cost of determining the correction region of the risky IP address.

[0200] The following describes a complete implementation process of the advertising delivery method provided in the embodiments of the present invention.

[0201] See Figure 8 This is a schematic diagram of the eighth process of the advertising delivery method provided in this embodiment of the invention. The first step is to collect data. Specifically, when the advertising delivery server receives an advertising delivery request initiated by the client, it records relevant information about the advertising delivery request, such as the current timestamp, IP address, DC-code (the data center where the advertising server that received the client request is located), and client information (a unique client ID obtained by combining OAI, IDFA, cookies, etc.). Furthermore, it resolves the corresponding IP region according to the IP database provided by the advertising association.

[0202] The second step is to determine the geographical distribution of clients and the distribution of clients with the first IP address within a preset time period (e.g., the past day). Specifically, based on dc-code information (e.g., dc-code = s, indicating the advertising server is located in data center s), the geographical distribution information of clients within the preset time period (e.g., the past day) is first determined. For example, if the total number of third requests received from client u in the past day is userreq(s,u), the IP addresses carried in the requests are resolved using the IP database provided by the advertising association, and n IP regions are counted. These IP regions can include city information. The distribution of clients within the IP regions and cities is then determined. iThe fourth request number is req(s,u,city) i ), obviously userreq(s,u)=∑req(s,u,city i Therefore, the percentage of requests made by client u in the i-th city can be calculated as p(s,u,city). i = req(s,u,city) i ) / userreq(s,u), p(s,u,city i Let $\frac{ ...

[0203] The third step is to calculate the first-region drift probability of the first IP address and output the risky IP addresses. Specifically, for a given first IP address, the IP region resolved from the IP database provided by the advertising association is ipcity. The expected value of a specific client u on this first IP address not being in that ipcity (i.e., regional drift) is draft(s,ip,u) = REQ(s,ip,u) * (1 - p(s,u,ipcity)). The data for each client under this first IP address is summarized to obtain the first-region drift probability of this first IP address as ip_draft(s,ip) = ∑draft(s,ip,u) / ipreq(s,ip). Finally, first IP addresses whose ip_draft(s,ip) exceeds a second preset threshold are output as risky IP addresses. When the advertising server in data center s receives an advertising request from a risky IP address, it does not serve ads targeted by that risky IP address's region.

[0204] The fourth step is to calculate the second regional drift probability of the identified risky IP addresses and further determine the correction region for the risky IP addresses. Specifically, similar to step three, the expected value of a specific client u on a risky IP address migrating to a certain IP region (city) is calculated as draft_city(s,ip,u,city) = REQ(s,ip,u)*p(s,u,city). The probability of this risky IP address migrating to a certain city is then calculated as ip_draft_city(s,ip,city) = ∑draft_city(s,ip,u,city) / ipreq(s,ip). The probability of this risky IP address migrating to each IP region is compared, and the IP region with the highest probability of migrating to a certain city and exceeding a third preset threshold is selected as the correction region for this risky IP address. When the advertising server in data center s receives an advertising request from a risky IP address, it needs to avoid advertising targeted by that risky IP address region, but can still deliver advertising targeted by that risky IP address in the correction region, thereby improving the utilization rate of advertising inventory and reducing losses.

[0205] Corresponding to the aforementioned advertising delivery method, this embodiment of the invention also provides an advertising delivery device applied to an advertising delivery server. See also... Figure 9 This is a schematic diagram of a first structure of an advertising delivery device provided in an embodiment of the present invention. The device includes:

[0206] The first acquisition module 201 is used to acquire advertising delivery requests within a preset time period and related information of the advertising delivery requests, wherein the related information of the advertising delivery requests includes the IP address of the client that sent the advertising delivery requests;

[0207] The first determining module 202 is used to determine the IP geographic distribution information of the client based on the IP address of the client that sent the advertising delivery request, wherein the IP geographic distribution information represents the percentage of times the client sent the advertising delivery request in the corresponding IP geographic region;

[0208] The first statistics module 203 is used to count the number of first requests for each IP address within the preset time period, and the number of second requests for each client corresponding to the IP address, wherein the IP address with the number of first requests greater than the first preset threshold is called the first IP address;

[0209] The first calculation module 204 is used to calculate the first regional drift probability of each first IP address based on the first number of requests to the first IP address, the second number of requests to each client corresponding to the first IP address, and the IP regional distribution information of each client corresponding to the first IP address. The first regional drift probability is the probability that the client corresponding to the first IP address is not in the first IP region, and the first IP region is the IP region to which the first IP address belongs.

[0210] The second determination module 205 is used to determine the first IP address whose first regional drift probability is greater than the second preset threshold as a risky IP address and add it to the risky IP address list.

[0211] The third determining module 206 is used to respond to a target advertisement delivery request sent by the target client, and when the target IP address carried in the target advertisement delivery request is the risk IP address, determine the target advertisement among the advertisements without target IP region targeting, wherein the target IP region is the IP region to which the target IP address belongs;

[0212] The target delivery module 207 is used to deliver the target advertisement to the target client.

[0213] By applying this embodiment of the invention, risky IP addresses are determined through data statistics and analysis of ad delivery requests and related information. In other words, this embodiment of the invention relies solely on data from the ad delivery server to determine risky IP addresses, without depending on data from third-party servers. Therefore, there is no need to build a separate risky IP server, thus avoiding additional server deployment and maintenance costs and reducing the cost of determining risky IP addresses. When the target IP address carried in the target ad delivery request sent by the target client is a risky IP address, delivering ads without target IP geographic targeting to the target client can reduce the geographic statistical discrepancies between the ad delivery server and third-party servers.

[0214] In one possible embodiment, see Figure 10 This is a schematic diagram of a second structure of the advertising delivery device provided in an embodiment of the present invention. The first determining module 202 includes:

[0215] The first statistics submodule 2021 is used to count the number of times each client sends an advertising delivery request within the preset time period to obtain the third request count;

[0216] The first determining submodule 2022 is used to determine the IP region to which the IP address of the client that sent the advertising delivery request belongs;

[0217] The second statistics submodule 2023 is used to count the number of times the client sends advertising delivery requests in each IP region within the preset time period, and obtain the fourth request count;

[0218] The first calculation submodule 2024 is used to calculate the percentage of times the client sends the advertising delivery request in each IP region based on the third request number and the fourth request number, so as to obtain the IP region distribution information of the client.

[0219] By applying the embodiments of the present invention, by separately counting the number of third requests sent by each client to place advertisements within a set time period, and the number of fourth requests sent by each client to place advertisements in each IP region, the IP region distribution information of each client can be accurately obtained. This IP region distribution information can represent the probability of a client appearing in each region, providing conditions for subsequently identifying risky IP addresses.

[0220] In one possible embodiment, see Figure 11 This is a schematic diagram of a third structure of the advertising delivery device provided in an embodiment of the present invention. The first calculation module 204 includes:

[0221] The first acquisition submodule 2041 is used to acquire, for each first IP address, the first IP region to which the first IP address belongs, and the IP region distribution information of the clients corresponding to the first IP address.

[0222] The second calculation submodule 2042 is used to calculate the mathematical expectation of the first regional drift of each client corresponding to the first IP address based on the second request number of each client corresponding to the first IP address and the IP regional distribution information of each client corresponding to the first IP address. The mathematical expectation of the first regional drift is the mathematical expectation that the client corresponding to the first IP address is not in the first IP region.

[0223] The third calculation submodule 2043 is used to calculate the first regional drift probability of the first IP address based on the mathematical expectation of the first regional drift of each client corresponding to the first IP address and the first number of requests of the first IP address.

[0224] By applying the embodiments of the present invention, for each first IP address, data analysis and calculation are performed using the second request count of each client corresponding to the first IP address and the IP geographic distribution information of each client corresponding to the first IP address. This allows for the mathematical expectation of the first geographic drift of each client corresponding to the first IP address. Furthermore, by using the mathematical expectation of the first geographic drift of each client corresponding to the first IP address and the first request count of the first IP address, the first geographic drift probability of the first IP address can be calculated, which is the probability that the client corresponding to the first IP address is not in the first IP region. By comparing this probability with a second preset threshold, risky IP addresses can be accurately identified without the need for additional server deployment, thus reducing the cost of identifying risky IP addresses.

[0225] In one possible embodiment, the second computing submodule 2042 is specifically used for:

[0226] For each client corresponding to the first IP address, with 1 as the minuend and the proportion of times the client sends the advertising request in the first IP region as the subtrahend, calculate the difference between 1 and the proportion of times the client sends the advertising request in the first IP region, and use this difference as the first difference.

[0227] Based on the client's second request count and the first difference, calculate the product of the client's second request count and the first difference as the mathematical expectation of the client's first regional drift;

[0228] The mathematical expectation of the first regional drift for each client corresponding to the first IP address is obtained by summing the results.

[0229] The third calculation submodule 2043 is specifically used for:

[0230] The mathematical expectation of the first regional drift of each client corresponding to the first IP address is summed to obtain the mathematical expectation of the first regional drift of the first IP address, which is used as the first mathematical expectation;

[0231] Using the first expected value as the dividend and the first number of requests from the first IP address as the divisor, the quotient of the first expected value and the first number of requests from the first IP address is calculated and used as the first regional drift probability of the first IP address.

[0232] By applying the embodiments of the present invention, a specific algorithm is provided for calculating the first region drift probability of a first IP address, thereby improving the accuracy of calculating the first region drift probability of a first IP address.

[0233] In one possible embodiment, see Figure 12This is a fourth structural diagram of the advertising delivery device provided in an embodiment of the present invention. The device further includes:

[0234] The second calculation module 208 is used to calculate the second region drift probability of the risk IP address based on the first number of requests to the risk IP address, the second number of requests to each client corresponding to the risk IP address, and the IP region distribution information of each client corresponding to the risk IP address. The second region drift probability is the probability of the client corresponding to the risk IP address in each IP region, and each second region drift probability corresponds to one IP region.

[0235] The fourth determining module 209 is used to compare the magnitude of the drift probability of each second region, and determine the IP region corresponding to the one with the largest second region drift probability and the second region drift probability greater than the third preset threshold as the correction region corresponding to the risky IP address.

[0236] The removal module 210 is used to remove the risky IP addresses whose correction regions have been determined from the risky IP address list to obtain a second IP address;

[0237] The fifth determining module 211 is used to respond to a target advertisement delivery request sent by the target client, and when the target IP address carried in the target advertisement delivery request is the second IP address, determine the target advertisement from the advertisements with corrected geographic targeting corresponding to the target IP address.

[0238] By applying the embodiments of the present invention, after identifying a risky IP address, the corrective region for that risky IP address can be determined through data statistics and analysis of advertising delivery requests and related information. In other words, the embodiments of the present invention, based on data from the advertising delivery server, can determine the corrective region for a risky IP address solely through the advertising delivery server, without relying on data from a third-party server. Therefore, there is no need to build a separate risky IP server, thus avoiding additional server deployment and maintenance costs and reducing the cost of determining the corrective region for a risky IP address.

[0239] In one possible embodiment, see Figure 13 This is a fifth structural schematic diagram of the advertising delivery device provided in an embodiment of the present invention. The second calculation module 208 includes:

[0240] The second acquisition submodule 2081 is used to acquire the number of second requests of each client corresponding to the risky IP address, and the percentage of times each client corresponding to the risky IP address sends the advertising delivery request in the corresponding IP region.

[0241] The fourth calculation submodule 2082 is used to calculate the expected value of the second region drift of each client corresponding to the risk IP address based on the number of second requests of each client corresponding to the risk IP address and the proportion of times each client corresponding to the risk IP address sends the advertising delivery request in the corresponding IP region. The expected value of the second region drift is: the expected value of the client corresponding to the risk IP address in each IP region, and each expected value of the second region drift corresponds to one IP region.

[0242] The fifth calculation submodule 2083 is used to calculate the probability of each second region drift of the risky IP address based on the mathematical expectation of the second region drift of each client corresponding to the risky IP address and the first number of requests of the risky IP address.

[0243] By applying the embodiments of the present invention, through data analysis and calculation of the number of second requests of each client corresponding to a risky IP address and the proportion of times each client corresponding to a risky IP address sends advertising requests in the corresponding IP region, the expected value of the second region drift of each client corresponding to the risky IP address can be obtained. Furthermore, by using the expected value of the second region drift of each client corresponding to the risky IP address and the number of first requests of the risky IP address, the probability of the second region drift of the risky IP address can be calculated, which is the probability of the client corresponding to the risky IP address in each IP region. By comparing the magnitude of each second region drift probability and comparing the second region drift probability with the largest value with a third preset threshold, the correction region of the risky IP address can be accurately determined without the need for additional server deployment, thus reducing the cost of determining the correction region of the risky IP address.

[0244] In one possible embodiment, the fourth computing submodule 2082 is specifically used for:

[0245] For each client corresponding to the risky IP address, based on the client's second request count and the proportion of times the client sends the advertising request in the corresponding IP region, the product of the client's second request count and the proportion of times the client sends the advertising request in the corresponding IP region is calculated as the mathematical expectation of the client's second region drift.

[0246] The mathematical expectation of the second-region drift of each client corresponding to the risky IP address is obtained by summarizing the results.

[0247] The fifth calculation submodule 2083 is specifically used for:

[0248] The mathematical expectation of the second regional drift of each client corresponding to the risky IP address is summed to obtain the mathematical expectation of the second regional drift of the risky IP address, which is used as the second mathematical expectation;

[0249] Using the second expected value as the dividend and the first number of requests from the risky IP address as the divisor, the quotient of the second expected value and the first number of requests from the risky IP address is calculated, and this quotient is used as the second regional drift probability of the risky IP address.

[0250] The present invention provides a specific algorithm for calculating the second-region drift probability of risky IP addresses, thereby improving the accuracy of calculating the second-region drift probability of risky IP addresses.

[0251] This invention also provides an electronic device, such as... Figure 14 As shown, it includes a processor 301, a communication interface 302, a memory 303, and a communication bus 304, wherein the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304.

[0252] Memory 303 is used to store computer programs;

[0253] When processor 301 executes a program stored in memory 303, it performs the following steps:

[0254] Obtain advertising delivery requests within a preset time period and related information of the advertising delivery requests, wherein the related information of the advertising delivery requests includes the IP address of the client that sent the advertising delivery requests;

[0255] Based on the IP address of the client that sent the ad delivery request, the IP geographic distribution information of the client is determined, wherein the IP geographic distribution information represents the percentage of times the client sends the ad delivery request in the corresponding IP geographic region;

[0256] For each IP address, the number of first requests to that IP address within the preset time period and the number of second requests to each client corresponding to that IP address are counted. The IP address with a first request count greater than a first preset threshold is referred to as the first IP address.

[0257] For each first IP address, based on the first number of requests to the first IP address, the second number of requests to each client corresponding to the first IP address, and the IP geographic distribution information of each client corresponding to the first IP address, the first geographic drift probability of the first IP address is calculated, wherein the first geographic drift probability is: the probability that the client corresponding to the first IP address is not in the first IP region, and the first IP region is the IP region to which the first IP address belongs;

[0258] The first IP address whose first regional drift probability is greater than the second preset threshold is identified as a risky IP address and added to the risky IP address list.

[0259] In response to a target ad delivery request sent by a target client, if the target IP address is a risky IP address in the risky IP address list, the target ad is determined from ads without target IP region targeting, wherein the target IP address is the IP address carried in the target ad delivery request, and the target IP region is the IP region to which the target IP address belongs;

[0260] The target advertisement is delivered to the target client.

[0261] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0262] The communication interface is used for communication between the aforementioned terminal and other devices.

[0263] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0264] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0265] By applying the embodiments of the present invention, risky IP addresses are determined through data statistics and analysis of advertising delivery requests and related information. In other words, the embodiments of the present invention determine risky IP addresses solely based on data from the advertising delivery server, without relying on data from third-party servers. Therefore, there is no need to build a separate risky IP server, thus avoiding additional server deployment and maintenance costs and reducing the cost of determining risky IP addresses.

[0266] In another embodiment of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, it implements any of the advertising delivery methods described in the above embodiments.

[0267] By applying the embodiments of the present invention, risky IP addresses are determined through data statistics and analysis of advertising delivery requests and related information. In other words, the embodiments of the present invention determine risky IP addresses based on analysis of existing data, eliminating the need for a separate server and thus avoiding additional server deployment and maintenance costs. This solves the problem of high costs in determining risky IP addresses in related technologies, thereby reducing the overall cost of risky IP address determination.

[0268] In another embodiment of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the advertising delivery methods described in the above embodiments.

[0269] By applying the embodiments of the present invention, risky IP addresses are determined through data statistics and analysis of advertising delivery requests and related information. In other words, the embodiments of the present invention determine risky IP addresses solely based on data from the advertising delivery server, without relying on data from third-party servers. Therefore, there is no need to build a separate risky IP server, thus avoiding additional server deployment and maintenance costs and reducing the cost of determining risky IP addresses.

[0270] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0271] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0272] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, electronic devices, storage media, and computer program products are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0273] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. An advertising placement method, characterized in that, Applied to an ad serving server, the method includes: Obtain advertising delivery requests within a preset time period and related information of the advertising delivery requests, wherein the related information of the advertising delivery requests includes the IP address of the client that sent the advertising delivery requests; Based on the IP address of the client that sent the ad delivery request, the IP geographic distribution information of the client is determined, wherein the IP geographic distribution information represents the percentage of times the client sends the ad delivery request in the corresponding IP geographic region; For each IP address, the number of first requests to that IP address within the preset time period and the number of second requests to each client corresponding to that IP address are counted. The IP address with a first request count greater than a first preset threshold is referred to as the first IP address. For each first IP address, based on the first number of requests to the first IP address, the second number of requests to each client corresponding to the first IP address, and the IP geographic distribution information of each client corresponding to the first IP address, the first geographic drift probability of the first IP address is calculated, wherein the first geographic drift probability is: the probability that the client corresponding to the first IP address is not in the first IP region, and the first IP region is the IP region to which the first IP address belongs; The first IP address whose first regional drift probability is greater than the second preset threshold is identified as a risky IP address and added to the risky IP address list. In response to a target ad delivery request sent by a target client, if the target IP address is a risky IP address in the risky IP address list, the target ad is determined from ads without target IP region targeting, wherein the target IP address is the IP address carried in the target ad delivery request, and the target IP region is the IP region to which the target IP address belongs; The target advertisement is delivered to the target client.

2. The method according to claim 1, characterized in that, Determining the IP geographic distribution information of the client based on the IP address of the client that sent the advertising delivery request includes: The number of times each client sends an ad delivery request within the preset time period is counted to obtain the third request count; Determine the IP region to which the IP address of the client that sent the ad delivery request belongs; The number of times the client sends advertising delivery requests in each IP region within the preset time period is counted to obtain the fourth request count; Based on the third and fourth request counts, the percentage of times the client sends the advertising delivery request in each IP region is calculated to obtain the client's IP region distribution information.

3. The method according to claim 1, characterized in that, The step of calculating the first geographical drift probability of each first IP address based on the first number of requests to that first IP address, the second number of requests to each client corresponding to that first IP address, and the geographical distribution information of the IP addresses of each client corresponding to that first IP address includes: For each of the first IP addresses, obtain the first IP region to which the first IP address belongs, and the IP region distribution information of the clients corresponding to the first IP address; Based on the number of second requests of each client corresponding to the first IP address and the IP geographic distribution information of each client corresponding to the first IP address, the mathematical expectation of the first geographic drift of each client corresponding to the first IP address is calculated, wherein the mathematical expectation of the first geographic drift is: the mathematical expectation that the client corresponding to the first IP address is not in the first IP region; Based on the mathematical expectation of the first regional drift of each client corresponding to the first IP address, and the first number of requests to the first IP address, the probability of the first regional drift of the first IP address is calculated.

4. The method according to claim 3, characterized in that, The step of calculating the mathematical expectation of the first regional drift of each client corresponding to the first IP address, based on the number of second requests of each client corresponding to the first IP address and the IP geographic distribution information of the clients corresponding to the first IP address, includes: For each client corresponding to the first IP address, with 1 as the minuend and the proportion of times the client sends the advertising request in the first IP region as the subtrahend, calculate the difference between 1 and the proportion of times the client sends the advertising request in the first IP region, and use this difference as the first difference. Based on the client's second request count and the first difference, calculate the product of the client's second request count and the first difference as the mathematical expectation of the client's first regional drift; The mathematical expectation of the first regional drift for each client corresponding to the first IP address is obtained by summing the results. The calculation of the first-region drift probability of the first IP address based on the mathematical expectation of the first-region drift of each client corresponding to the first IP address and the first number of requests to the first IP address includes: The mathematical expectations of the first regional drift of each client corresponding to the first IP address are summed to obtain the mathematical expectation of the first regional drift of the first IP address, which is used as the first mathematical expectation; Using the first expected value as the dividend and the first number of requests from the first IP address as the divisor, the quotient of the first expected value and the first number of requests from the first IP address is calculated and used as the first regional drift probability of the first IP address.

5. The method according to claim 1, characterized in that, After determining the first IP address whose first geographical migration probability is greater than the second preset threshold as a risky IP address, the method further includes: Based on the first number of requests to the risky IP address, the second number of requests to each client corresponding to the risky IP address, and the IP geographic distribution information of each client corresponding to the risky IP address, the second geographic drift probability of the risky IP address is calculated. The second geographic drift probability is the probability of the client corresponding to the risky IP address in each IP geographic region, and each second geographic drift probability corresponds to one IP geographic region. Compare the magnitudes of the drift probabilities of each second region, and determine the IP region corresponding to the one with the highest drift probability and a drift probability greater than a third preset threshold as the correction region corresponding to the risky IP address; The risky IP addresses whose correction regions have been identified are removed from the list of risky IP addresses, resulting in a second IP address. The method further includes: In response to a target ad delivery request sent by a target client, if the target IP address carried in the target ad delivery request is the second IP address, the target ad is determined from the ads with corrected geographic targeting corresponding to the target IP address.

6. The method according to claim 5, characterized in that, The calculation of the second-region drift probability of the risky IP address based on the first number of requests to the risky IP address, the second number of requests to each client corresponding to the risky IP address, and the IP geographic distribution information of each client corresponding to the risky IP address includes: Obtain the number of second requests for each client corresponding to the risky IP address, and the percentage of times each client corresponding to the risky IP address sends the advertising request in the corresponding IP region; Based on the number of second requests of each client corresponding to the risky IP address, and the percentage of times each client corresponding to the risky IP address sends the advertising request in the corresponding IP region, the mathematical expectation of the second region drift of each client corresponding to the risky IP address is calculated. The mathematical expectation of the second region drift is: the mathematical expectation of the client corresponding to the risky IP address in each IP region, and each mathematical expectation of the second region drift corresponds to one IP region. Based on the mathematical expectation of the second-region drift of each client corresponding to the risky IP address, and the first request number of the risky IP address, calculate the second-region drift probability of each risky IP address.

7. The method according to claim 6, characterized in that, The step of calculating the expected value of the second-region drift of each client corresponding to the risky IP address, based on the number of second requests for each client corresponding to the risky IP address and the percentage of times each client sent the advertising request in the corresponding IP region, includes: For each client corresponding to the risky IP address, based on the client's second request count and the proportion of times the client sends the advertising request in the corresponding IP region, the product of the client's second request count and the proportion of times the client sends the advertising request in the corresponding IP region is calculated as the mathematical expectation of the client's second region drift. The mathematical expectation of the second-region drift of each client corresponding to the risky IP address is obtained by summarizing the results. The calculation of the second-region drift probability of the risky IP address based on the mathematical expectation of the second-region drift of each client corresponding to the risky IP address and the first number of requests to the risky IP address includes: The mathematical expectation of the second regional drift of each client corresponding to the risky IP address is summed to obtain the mathematical expectation of the second regional drift of the risky IP address, which is used as the second mathematical expectation; Using the second expected value as the dividend and the first number of requests from the risky IP address as the divisor, the quotient of the second expected value and the first number of requests from the risky IP address is calculated, and this quotient is used as the second regional drift probability of the risky IP address.

8. An advertising delivery device, characterized in that, The device, used in an ad delivery server, includes: The first acquisition module is used to acquire advertising delivery requests within a preset time period and related information of the advertising delivery requests, wherein the related information of the advertising delivery requests includes the IP address of the client that sent the advertising delivery requests; The first determining module is used to determine the IP geographic distribution information of the client based on the IP address of the client that sent the advertising delivery request, wherein the IP geographic distribution information represents the percentage of times the client sent the advertising delivery request in the corresponding IP geographic region; The first statistics module is used to count the number of first requests for each IP address within the preset time period, and the number of second requests for each client corresponding to the IP address, wherein the IP address with the number of first requests greater than the first preset threshold is called the first IP address; The first calculation module is used to calculate the first regional drift probability of each first IP address based on the first number of requests to the first IP address, the second number of requests to each client corresponding to the first IP address, and the IP regional distribution information of each client corresponding to the first IP address. The first regional drift probability is the probability that the client corresponding to the first IP address is not in the first IP region, and the first IP region is the IP region to which the first IP address belongs. The second determination module is used to determine the first IP address whose first regional drift probability is greater than the second preset threshold as a risky IP address and add it to the risky IP address list. The third determination module is used to respond to the target advertisement delivery request sent by the target client, and determine the target advertisement from the advertisements without target IP region targeting if the target IP address is a risk IP address in the risk IP address list. The target IP address is the IP address carried in the target advertisement delivery request, and the target IP region is the IP region to which the target IP address belongs. The target delivery module is used to deliver the target advertisement to the target client.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the steps of the method described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1-7.

Images