A Big Data-Driven Method and System for Efficient and Precise Online Advertising Targeting

Abstract

This application relates to a data-driven, efficient, and precise online advertising delivery method and system, belonging to the field of advertising delivery technology. It addresses the problem that different audience groups have varying levels of acceptance of ad pushes, and pushing ads at uniform intervals can easily lead some audiences to feel that the ads are too frequent, affecting their effectiveness. The method includes: calculating the total time a user has owned a relevant product based on the time of their most recent purchase and the current time; calculating the time difference between the total time a user has owned the relevant product and the average replacement time of the corresponding product; analyzing and determining the advertising delivery strategy based on the comparison between the time difference and a preset time difference and the corresponding advertising delivery strategy; and executing the advertising delivery strategy. This application has the following effects: making the ad push interval more suitable for each individual's situation, thereby making the advertising effect more efficient and precise.

Description

Technical Field

[0001] This application relates to the field of advertising delivery technology, and in particular to a method and system for efficient and accurate online advertising delivery based on big data. Background Technology

[0002] In recent years, social networks have gradually become an important means for people to communicate and disseminate information. Due to their rapid dissemination and wide reach, more and more advertisers are choosing to push advertisements through social networks. Currently, advertising is generally delivered via broadcast, meaning ads are pushed indiscriminately to all users of the social network.

[0003] However, since each advertisement targets a different audience, indiscriminate ad delivery may cause user resentment or even lead to users blocking ads, resulting in poor ad dissemination effectiveness.

[0004] In related technologies, advertising targeting specific audiences is typically conducted at uniform time intervals.

[0005] Regarding the aforementioned technologies, the inventors believe that they have the following drawbacks: different audience groups have different levels of acceptance of advertising pushes. Pushing ads at uniform intervals can easily lead some advertising audiences to feel that the ads are being pushed too frequently, thus affecting the effectiveness of the ads. Summary of the Invention

[0006] To make the ad delivery intervals more suitable for each individual's situation, thereby making the advertising more efficient and accurate, this application provides a method and system for efficient and accurate online ad delivery based on big data.

[0007] Firstly, this application provides a method for efficient and accurate online advertising delivery based on big data, employing the following technical solution:

[0008] A big data-driven method for efficient and precise online advertising delivery includes:

[0009] Acquire the time points when users clicked to view online advertising content related to the products in their historical click history, as well as the time point when the user last purchased a related type of product;

[0010] Based on the time points when users clicked to view online advertising content in their historical click history, we can analyze and obtain the time interval between the two most recent clicks to view the same product.

[0011] If the time interval between a user's two most recent clicks to view the same product is less than a preset time interval, then obtain the correspondence between the product and the average replacement time, and analyze the average replacement time of the corresponding product.

[0012] The total duration for which the user owns the relevant product category is calculated based on the time of the user's most recent purchase of the relevant product category and the current time.

[0013] Calculate the time difference between the total duration a user owns a particular type of product and the average replacement time of that product;

[0014] Based on the comparison results between the time difference and the preset time difference and the corresponding relationship with the advertising placement strategy, the advertising placement strategy is analyzed and determined.

[0015] Implement advertising placement strategies;

[0016] If the time interval between a user's two most recent clicks to view the same product is greater than or equal to the preset time interval, the advertising strategy for that product will not be executed.

[0017] Optionally, the preset time interval can be obtained by:

[0018] Query the time interval between a user's second purchase of the same type of product in their history;

[0019] If found, the time interval between the user's two previous purchases of the same product will be used as the preset time interval.

[0020] Conversely, the preset time interval is based on the average time interval between other users purchasing similar products.

[0021] Optionally, the preset time interval may include the average time interval between other users purchasing similar products:

[0022] Get the time interval between the current user's second purchases of other product types in their history;

[0023] Based on the time interval between the current user's second purchase of other types of products and the time interval between the second purchase of other types of products by other users, analyze and determine a preset number of other users whose time interval between purchasing other types of products is closest to the current user's time interval;

[0024] Obtain the average time interval between the second purchases of the same product by the other users in the past, and use it as the preset time interval.

[0025] Optionally, the correlation between products and average replacement time can be obtained, and the average replacement time of the corresponding products can be analyzed, including:

[0026] The data sources for obtaining preset time intervals are defined as including the time interval between a user's second purchase of the same type of product in history and the average time interval between other users' purchases of the same type of product.

[0027] Based on the correspondence between the data source of the preset time interval and the actual average replacement time deviation, the actual average replacement time deviation is analyzed and determined.

[0028] Obtain the correspondence between products and average replacement time, obtain the average replacement time, and add the actual average replacement time deviation to obtain the effective average replacement time, which is used as the average replacement time interval for the corresponding products.

[0029] Optionally, based on the comparison results between the time difference and the preset time difference and the correspondence with the advertising placement strategy, the advertising placement strategy is analyzed and determined, including:

[0030] The correlation between the comparison results of the time difference and the preset time difference and the advertising strategy;

[0031] If the time difference is greater than or equal to the first preset time difference, then ads containing the corresponding products will be delivered periodically according to the first preset interval until the user purchases the corresponding product again or the user sets the corresponding product ads to be blocked.

[0032] If the time difference is less than the first preset time difference and greater than the second preset time difference, then advertisements containing the corresponding products will be periodically delivered according to the second preset interval period until the user purchases the corresponding product again or the user sets the corresponding product advertisement to be blocked. The first preset interval period is less than the second preset interval period.

[0033] If the time difference is less than or equal to the second preset time difference, then advertisements containing the corresponding product repair tools will be periodically displayed according to the third preset interval until any one of the following three situations occurs: the user purchases the corresponding product again, the user purchases the corresponding product repair tools again, or the user sets up the blocking of the corresponding product advertisement.

[0034] Optionally, the acquisition of the third preset interval period includes:

[0035] Obtain the time interval between the user's historical purchases of similar product repair tools;

[0036] Based on the time intervals between users' historical purchases of similar product repair tools, the average time interval is calculated and used as the third preset interval period.

[0037] Optionally, the acquisition of the third preset interval period includes:

[0038] Obtain the time interval between the user's historical purchases of repair tools for similar products and the probability of purchasing repair tools. The probability of purchasing repair tools is the percentage of the number of times repair tools were purchased relative to the total number of times replacement products were purchased.

[0039] If the probability of purchasing repair tools exceeds the preset probability, the average time interval is calculated based on the time interval between the user's historical purchases of similar product repair tools, and this average time interval is used as the third preset time interval.

[0040] If the probability of purchasing repair tools is less than the preset probability, the third preset interval period will be doubled.

[0041] Optionally, the following advertising strategies can be implemented:

[0042] Obtain the probability distribution of the devices that users viewed ads on at different times in their history;

[0043] Based on the probability distribution of the terminals that users viewed ads on during the current time period and in different time periods in the user's history, we analyze and determine the terminal with the highest probability of users viewing ads during the current time period, and then execute the ad delivery strategy on the corresponding terminal.

[0044] Secondly, this application provides a big data-driven online advertising system for efficient and precise delivery, employing the following technical solution:

[0045] A big data-driven online advertising system for efficient and precise targeting includes a memory, a processor, and a program stored in the memory and executable on the processor. When the program is loaded and executed by the processor, it implements the big data-driven online advertising method for efficient and precise targeting as described in the first aspect. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0047] Figure 1 This is a flowchart illustrating a method for efficient and precise online advertising delivery based on big data, according to an embodiment of this application.

[0048] Figure 2 This is a schematic diagram of the process for obtaining a preset time interval according to another embodiment of this application.

[0049] Figure 3 This is a flowchart illustrating another embodiment of the present application, which uses the average time interval between other users purchasing similar products as a preset time interval.

[0050] Figure 4 This is a flowchart illustrating another embodiment of the present application for obtaining the correspondence between products and average replacement time, and analyzing the average replacement time of the corresponding products.

[0051] Figure 5 This is a flowchart illustrating the process of analyzing and determining an advertising placement strategy based on the correspondence between the comparison results of the time difference and the preset time difference and the advertising placement strategy, according to another embodiment of this application.

[0052] Figure 6 yes Figure 5 One implementation method is to obtain the third preset interval time period.

[0053] Figure 7 yes Figure 5 Another implementation method for obtaining the third preset interval period.

[0054] Figure 8 This is a schematic diagram illustrating the process of implementing an advertising delivery strategy according to another embodiment of this application. Detailed Implementation

[0055] The present application will be further described in detail below with reference to the accompanying drawings.

[0056] Reference Figure 1 This application discloses a method for efficient and precise online advertising delivery based on big data, comprising:

[0057] Step S100: Obtain the time points of the products involved in the user's historical clicks on online advertising content, as well as the time point of the user's most recent purchase of related types of products.

[0058] The products involved in the online advertising content can be electronic devices such as mobile phones and computers, or other products. The time points when users historically clicked to view the products involved in the online advertising content can be retrieved from a preset database that stores the time points when users historically clicked to view the products involved in the online advertising content. The time point when a user last purchased a related type of product can also be retrieved from a preset database that stores the time points when users historically clicked to view the products involved in the online advertising content.

[0059] Step S200: Based on the time points of the products involved in the online advertising content that the user clicked to view in the past, analyze and obtain the time interval between the user's two most recent clicks to view the same product.

[0060] Step S300: If the time interval between the user's two most recent clicks to view the same product is less than a preset time interval, then obtain the correspondence between the product and the average replacement time, and analyze the average replacement time of the corresponding product.

[0061] The preset time interval is 2 days or 3 days, or it can be a time interval set by other users according to their actual needs; the correspondence between products and average replacement time can be obtained by querying a preset database that stores the correspondence between products and average replacement time.

[0062] Step S400: Calculate the total duration for which the user has owned the relevant product category based on the time of the user's most recent purchase of the relevant product category and the current time.

[0063] Step S500: Calculate the time difference between the total duration of time a user owns the relevant type of product and the average replacement time of the corresponding product.

[0064] Step S600: Based on the comparison results of the time difference and the preset time difference and the correspondence with the advertising placement strategy, analyze and determine the advertising placement strategy.

[0065] The preset time difference can be 2 hours or 3 hours, or other times. The ad placement strategy can be retrieved from a database that stores the comparison results between the preset time difference and the corresponding relationship between the ad placement strategy and the time difference. The ad placement strategy refers to placement according to regular schedules or on terminals that users frequently view.

[0066] Step S700: Implement the advertising placement strategy.

[0067] Step S800: If the time interval between the user's two most recent clicks to view the same product is greater than or equal to the preset time interval, then the advertising strategy for the corresponding product will not be executed.

[0068] In step S300, the preset time interval can be further analyzed to better suit the user's situation. Therefore, further analysis of the preset time interval is required. See details below. Figure 2 The illustrated embodiments are described in detail.

[0069] Reference Figure 2 The acquisition of the preset time interval mentioned in step S300 includes:

[0070] Step S310: Query the time interval between the user's previous two purchases of the same type of product.

[0071] Step S320: If found, the time interval between the time interval of the user's two previous purchases of the same product is used as the preset time interval.

[0072] Step S330: Conversely, the average time interval between other users purchasing similar products is used as the preset time interval.

[0073] exist Figure 2In step S330, when using the average time interval between other users' purchases of similar products as the preset time interval, it is also necessary to consider filtering the other users to ensure that the purchasing habits of the filtered users are as similar as possible to the original users. Therefore, further analysis is needed on using the average time interval between other users' purchases of similar products as the preset time interval. See [link / reference] for details. Figure 3 The illustrated embodiments are described in detail.

[0074] Reference Figure 3 The preset time interval mentioned in step S330, based on the average time interval between other users' purchases of similar products, includes:

[0075] Step S331: Obtain the time interval between the current user's previous second purchases of other types of products.

[0076] The time interval between the current user's previous purchases of other types of products can be obtained by querying a pre-set database that stores the time interval between the current user's previous purchases of other types of products.

[0077] Step S332: Based on the time interval between the current user's second purchase of other types of products in history and the time interval between the other users' second purchase of other types of products in history, analyze and determine a preset number of other users whose time interval between purchasing other types of products is closest to the current user's time interval.

[0078] The time intervals between other users' previous purchases of other types of products can be obtained from a pre-set database that stores the time intervals between other users' previous purchases of other types of products; the pre-set number of other users can be 2 or 3, or it can be the number of other users set by the user.

[0079] Step S333: Obtain the average time interval between the second purchases of the same product by the other users in the past, and use it as the preset time interval.

[0080] exist Figure 1 Step S300 also needs to consider the replacement time set according to the data source, which may have time deviations. Therefore, further analysis of the average replacement time of the corresponding products is required. See [link / reference] for details. Figure 4 The illustrated embodiments are described in detail.

[0081] Reference Figure 4 The relationship between obtaining products and average replacement time mentioned in step S300, and the analysis of the average replacement time of the corresponding products, includes:

[0082] Step S3a0: Obtain the data source for the preset time interval. Define the data source as including the time interval between users' previous purchases of the same type of product and the average time interval between other users' purchases of the same type of product.

[0083] Step S3b0: Based on the correspondence between the data source of the preset time interval and the actual average replacement time deviation, analyze and determine the actual average replacement time deviation.

[0084] The correspondence between the data source of the preset time interval and the actual average replacement time deviation can be obtained by querying a preset database that stores the correspondence between the data source of the preset time interval and the actual average replacement time deviation.

[0085] Step S3c0: Obtain the correspondence between the product and the average replacement time, obtain the average replacement time, and add the actual average replacement time deviation to obtain the effective average replacement time, which is used as the average replacement time interval for the corresponding product.

[0086] The correspondence between products and average replacement time can be obtained by querying a pre-set database that stores the correspondence between products and average replacement time.

[0087] exist Figure 1 In step S600, it is also necessary to consider different advertising strategies corresponding to different comparison results, so as to achieve more accurate and effective advertising delivery. Therefore, further analysis and determination of the advertising strategy are required, as detailed in [reference needed]. Figure 5 The illustrated embodiments are described in detail.

[0088] Reference Figure 5 The step S600, which involves analyzing and determining the advertising strategy based on the comparison results between the time difference and the preset time difference, includes:

[0089] Step S610: Obtain the comparison result of the time difference and the preset time difference and the correspondence with the advertising placement strategy.

[0090] Step S620: If the time difference is greater than or equal to the first preset time difference, then periodically deliver advertisements containing the corresponding products according to the first preset interval period until the user purchases the corresponding product again or the user sets the corresponding product advertisement to be blocked.

[0091] The first preset time difference can be 10 days.

[0092] Step S630: If the time difference is less than the first preset time difference and greater than the second preset time difference, then ads containing the corresponding products are periodically delivered according to the second preset interval period until the user purchases the corresponding product again or the user sets the corresponding product ads to be blocked, wherein the first preset interval period is less than the second preset interval period.

[0093] The second preset time difference can be 12 days.

[0094] Step S640: If the time difference is less than or equal to the second preset time difference, then periodically deliver advertisements containing the corresponding product repair tools according to the third preset time interval until any one of the following three situations occurs: the user purchases the corresponding product again, the corresponding product repair tools are delivered, or the user sets up the blocking of the corresponding product advertisement.

[0095] For example, if the original product was a ballpoint pen, then the repair tools refer to the tools used to repair the ballpoint pen.

[0096] Reference Figure 6 The acquisition of the third preset interval period includes:

[0097] Step S641: Obtain the time interval between the user's historical purchases of similar product repair tools.

[0098] Step S642: Analyze and calculate the average time interval based on the time interval between the user's historical purchases of similar product repair tools, and use this as the third preset time interval.

[0099] Reference Figure 7 The acquisition of the third preset interval period includes:

[0100] Step S64a: Obtain the time interval between the user's historical purchases of repair tools for similar products and the probability of purchasing repair tools. The probability of purchasing repair tools is the percentage of the number of times repair tools are purchased relative to the total number of times replacement products are purchased.

[0101] Step S64b: If the probability of purchasing a repair tool exceeds the preset probability, then the average time interval is calculated based on the time interval of the user's historical purchase of repair tools for similar products, and this average time interval is used as the third preset time interval.

[0102] The preset probability can be 30%.

[0103] In step S64c, if the probability of purchasing repair tools is less than the preset probability, the third preset interval period is doubled.

[0104] Reference Figure 8 The advertising delivery strategy mentioned in step S700 includes:

[0105] Step S710: Obtain the probability distribution of the terminals that the user viewed the advertisement on in different time periods in history.

[0106] The probability distribution of the terminals that users viewed ads on in different time periods can be obtained by querying a database that stores the probability distribution of the terminals that users viewed ads on in different time periods.

[0107] Step S720: Based on the probability distribution of terminals that users view advertisements in different time periods in the current time period and in the user's history, analyze and determine the terminal with the highest probability of viewing advertisements in the current time period, and execute the advertisement delivery strategy on the corresponding terminal.

[0108] The terminal can be a mobile phone, computer, or other platform.

[0109] Based on the same inventive concept, embodiments of the present invention provide a big data-driven, highly efficient and precise online advertising delivery system, including a memory and a processor. The memory stores data that can run on the processor to implement the following... Figures 1 to 8 The procedure for any method.

[0110] The embodiments described in this specific implementation are preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for efficient and precise online advertising delivery based on big data, characterized in that, include: Acquire the time points when users clicked to view online advertising content related to the products in their historical click history, as well as the time point when the user last purchased a related type of product; Based on the time points when users clicked to view online advertising content in their historical click history, we can analyze and obtain the time interval between the two most recent clicks to view the same product. If the time interval between a user's two most recent clicks to view the same product is less than a preset time interval, then obtain the correspondence between the product and the average replacement time, and analyze the average replacement time of the corresponding product. The total duration for which the user owns the relevant product category is calculated based on the time of the user's most recent purchase of the relevant product category and the current time. Calculate the time difference between the total duration a user owns a particular type of product and the average replacement time of that product; Based on the comparison results between the time difference and the preset time difference and the corresponding relationship with the advertising placement strategy, the advertising placement strategy is analyzed and determined. Implement advertising placement strategies; If the time interval between a user's two most recent clicks to view the same product is greater than or equal to the preset time interval, the advertising strategy for the corresponding product will not be executed. The acquisition of the preset time interval includes: Query the time interval between a user's second purchase of the same type of product in their history; If found, the time interval between the user's two previous purchases of the same product will be used as the preset time interval. Conversely, the average time interval between other users purchasing similar products is used as the preset time interval. The preset time interval is based on the average time interval between other users' purchases of similar products, including: Get the time interval between the current user's second purchases of other product types in their history; Based on the time interval between the current user's second purchase of other types of products and the time interval between the second purchase of other types of products by other users, analyze and determine a preset number of other users whose time interval between purchasing other types of products is closest to the current user's time interval; Obtain the average time interval between the second purchases of the same product by the other users in the past, and use it as the preset time interval; Obtain the correlation between products and average replacement time, and analyze the average replacement time of the corresponding products, including: The data sources for obtaining preset time intervals are defined as including the time interval between a user's second purchase of the same type of product in history and the average time interval between other users' purchases of the same type of product. Based on the correspondence between the data source of the preset time interval and the actual average replacement time deviation, the actual average replacement time deviation is analyzed and determined. Obtain the correspondence between products and average replacement time, obtain the average replacement time, and add the actual average replacement time deviation to obtain the effective average replacement time, which is used as the average replacement time for the corresponding product.

2. The method for efficient and accurate online advertising delivery based on big data as described in claim 1, characterized in that, Based on the comparison results between the time difference and the preset time difference and the corresponding relationship with the advertising placement strategy, the advertising placement strategy is analyzed and determined to include: The correlation between the comparison results of the time difference and the preset time difference and the advertising strategy; If the time difference is greater than or equal to the first preset time difference, then ads containing the corresponding products will be delivered periodically according to the first preset interval until the user purchases the corresponding product again or the user sets the corresponding product ads to be blocked. If the time difference is less than the first preset time difference and greater than the second preset time difference, then advertisements containing the corresponding products will be periodically delivered according to the second preset interval period until the user purchases the corresponding product again or the user sets the corresponding product advertisement to be blocked. The first preset interval period is less than the second preset interval period. If the time difference is less than or equal to the second preset time difference, then advertisements containing the corresponding product repair tools will be periodically displayed according to the third preset interval until any one of the following three situations occurs: the user purchases the corresponding product again, the user purchases the corresponding product repair tools again, or the user sets up the blocking of the corresponding product advertisement.

3. The method for efficient and accurate online advertising delivery based on big data as described in claim 2, characterized in that, The acquisition of the third preset interval time period includes: Obtain the time interval between the user's historical purchases of similar product repair tools; Based on the time intervals between users' historical purchases of similar product repair tools, the average time interval is calculated and used as the third preset interval period.

4. The method for efficient and accurate online advertising delivery based on big data as described in claim 2, characterized in that, The acquisition of the third preset interval time period includes: Obtain the time interval between the user's historical purchases of repair tools for similar products and the probability of purchasing repair tools. The probability of purchasing repair tools is the percentage of the number of times repair tools were purchased relative to the total number of times replacement products were purchased. If the probability of purchasing repair tools exceeds the preset probability, the average time interval is calculated based on the time interval between the user's historical purchases of repair tools for similar products, and this average time interval is used as the third preset time interval. If the probability of purchasing repair tools is less than the preset probability, the third preset interval period will be doubled.

5. The method for efficient and accurate online advertising delivery based on big data as described in claim 1, characterized in that, Implementing an advertising strategy includes: Obtain the probability distribution of the devices that users viewed ads on at different times in their history; Based on the probability distribution of the terminals that users viewed ads on during the current time period and in different time periods in the user's history, we analyze and determine the terminal with the highest probability of users viewing ads during the current time period, and then execute the ad delivery strategy on the corresponding terminal.

6. A big data-driven online advertising system for efficient and precise delivery, characterized in that: It includes a memory, a processor, and a program stored in the memory and executable on the processor, which, when loaded and executed by the processor, implements a method for efficient and precise online advertising based on big data as described in any one of claims 1 to 5.

Images