A large-scale precision marketing algorithm and system based on artificial intelligence

Abstract

This invention relates to the technical field of marketing push, and in particular to a large-scale precision marketing algorithm and system based on artificial intelligence. The algorithm includes: obtaining the names of applications installed on a user's terminal; inputting the installed application names into a first model to infer first user preferences, wherein the first user preferences include multiple first preference tags; calculating the matching degree between each advertisement in the advertising database and the first user preferences, wherein each advertisement is associated with a preference tag; selecting advertisements whose matching degree meets preset requirements as advertisements to be pushed; and pushing advertisements with matching degrees exceeding preset values ​​to the user's terminal. This application has the effect of enhancing the personalization of advertising marketing push and improving the accuracy of marketing.

Description

Technical Field

[0001] This invention relates to the technical field of marketing push, and in particular to a large-scale precision marketing algorithm and system based on artificial intelligence. Background Technology

[0002] Advertising and marketing refer to the activities by which enterprises promote their products through advertising, encourage direct consumer purchases, expand product sales, and enhance the enterprise's brand awareness, reputation, and influence. With the rapid development of economic globalization and the market economy, advertising and marketing activities are playing an increasingly important role in corporate marketing strategies and are an important component of the corporate marketing mix.

[0003] However, the current methods of pushing electronic advertisements to mobile devices by various software are relatively simple and lack personalization, which often leads to user resentment and therefore needs to be improved. Summary of the Invention

[0004] To enhance the personalization of advertising and marketing pushes and improve the accuracy of marketing, this application provides a large-scale precision marketing algorithm and system based on artificial intelligence.

[0005] The above-mentioned objective of this application is achieved through the following technical solution: A large-scale precision marketing algorithm based on artificial intelligence includes: Get the names of applications installed on the user's terminal; Input the name of the installed application into the first model, and infer the first user preference, which includes multiple first preference tags. Calculate the matching degree between each advertisement in the advertising database and the first user's preferences, with each advertisement associated with a preference tag; Ads that meet the preset matching requirements will be designated as ads to be pushed to users. Ads with a match rate exceeding a preset value are pushed to the user's device.

[0006] By adopting the above technical solution, user preferences are inferred based on the applications installed on the user's terminal, and then appropriate advertising information is matched and pushed to the user's terminal, thereby enhancing the personalization of advertising marketing push and improving the accuracy of marketing.

[0007] In a preferred embodiment, this application may be further configured to include: Get the name of the application uninstalled by the user's terminal; Input the name of the uninstalled application into the first model, and infer the second user preference, which includes multiple second preference tags. The uninstalled application name is input into the second model, and user aversion is inferred. The user aversion includes multiple aversion tags. The tags to be deleted are obtained based on the second user's preferences and user dislikes; Delete the first preference tag that is the same as the tag to be deleted in the first user preferences.

[0008] By adopting the above technical solution, the first user preference after deleting the tags to be deleted can reduce the amount of content that users no longer need or like when pushing marketing content, thereby increasing user acceptance and reducing user aversion.

[0009] In a preferred example, this application can be further configured to: calculate the matching degree between each advertisement in the advertisement database and the first user's preferences, including: The number of identical tags between the preference tags associated with the advertising information and multiple first preference tags in the first user's preferences; The smaller of the number of preference tags associated with the advertising information and the number of multiple first preference tags in the first user's preferences is used as the comparison value; The ratio between the number of identical labels and the value to be compared is used as the matching degree.

[0010] By adopting the above technical solution and using a smaller value as the denominator, the actual matching situation can be reflected more accurately.

[0011] In a preferred embodiment, this application can be further configured such that the installed application name does not include the basic application name in the basic application library.

[0012] By adopting the above technical solution, the impact of basic applications within the terminal on the analysis can be avoided.

[0013] In a preferred embodiment, this application can be further configured such that obtaining the tags to be deleted based on second user preferences and user aversions includes: Tags that are identical between multiple second-preference tags in the second user preferences list and multiple dislike tags in the user dislike list are designated as tags to be deleted.

[0014] By adopting the above technical solution, the tags to be deleted are accurately located by using multiple second-preference tags in the second user preferences and multiple dislike tags in the user dislikes, thus avoiding the situation where some tags are mistakenly deleted due to using only the second user preferences or only the user dislikes.

[0015] The second objective of this invention is achieved through the following technical solution: A large-scale precision marketing system based on artificial intelligence includes: The application acquisition module is used to obtain the names of applications already installed on the user's terminal. The first preference inference module is used to input the name of the installed application into the first model and infer the first user preference, which includes multiple first preference tags. The calculation module is used to calculate the matching degree between each advertisement in the advertising database and the first user's preferences. Each advertisement is associated with a preference tag. The push-to-ads module is used to select advertising information that meets preset matching requirements as ads to be pushed. The push module is used to push advertisements with a matching degree exceeding a preset value to the user's terminal.

[0016] In a preferred embodiment, this application may be further configured to include: The uninstalled application acquisition module is used to obtain the names of applications uninstalled by the user's terminal; The second preference reasoning module is used to input the name of the uninstalled application into the first model and reason to obtain the second user preference, which includes multiple second preference tags. The aversion reasoning module is used to input the name of the uninstalled application into the second model and reason to obtain the user's aversion, which includes multiple aversion tags. The "to be deleted" module is used to obtain tags to be deleted based on the second user's preferences and the user's dislikes; The deletion module is used to delete a first preference tag that is the same as the tag to be deleted in the first user preferences.

[0017] In a preferred embodiment, this application can be further configured such that the computing module includes: The quantity statistics unit is used to count the number of identical tags between the preference tags associated with the advertising information and multiple first preference tags in the first user's preferences; The unit for obtaining the comparison value is used to obtain the smaller value between the number of preference tags associated with the advertising information and the number of multiple first preference tags in the first user preferences as the comparison value; The matching degree calculation unit is used to take the ratio between the number of identical tags and the value to be compared as the matching degree.

[0018] In summary, this application includes at least one of the following beneficial technical effects: 1. Based on the applications installed on the user's terminal, infer the user's preferences, then match appropriate advertising information and push it to the user's terminal, thereby enhancing the personalization of advertising and marketing push and improving the accuracy of marketing. 2. Deleting the first user preference after deleting the tags to be deleted can reduce the amount of content that users no longer need or like when pushing marketing content in the future, thereby increasing user acceptance and reducing user aversion; 3. By using multiple second-preference tags in the second user preferences and multiple dislike tags in the user dislikes, the tags to be deleted can be accurately located, avoiding the situation where some tags are accidentally deleted due to using only the second user preferences or only the user dislikes. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the implementation of a large-scale precision marketing algorithm based on artificial intelligence in one embodiment of this application; Figure 2 This is a flowchart illustrating the implementation of a large-scale precision marketing algorithm based on artificial intelligence in another embodiment of this application; Figure 3 This is a flowchart illustrating the implementation of a large-scale precision marketing algorithm based on artificial intelligence in another embodiment of this application. Detailed Implementation

[0020] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0021] It should be noted that the terms "first," "second," etc., used in this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this disclosure.

[0022] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0023] Figure 1 This is a flowchart illustrating the implementation of a large-scale market precision marketing algorithm based on artificial intelligence in one embodiment of this application, as follows: Figure 1 As shown, this AI-based precision marketing algorithm for large-scale marketing includes: S1. Obtain the names of applications installed on the user's terminal; The application names include names such as "Weibo", "Himalaya", "Acfun", "Hema", "Xiaohongshu", "MiLe", etc. The installed application names do not include the basic application names in the basic application library. The basic applications in the basic application library can be applications that come pre-installed on the terminal, or applications used for daily life such as "QQ" and "WeChat".

[0024] The terminal can be an electronic device such as a mobile phone, tablet computer, multimedia playback device, wearable device, or vehicle terminal.

[0025] S2. Input the name of the installed application into the first model, and infer the first user preference. The first user preference includes multiple first preference tags. The first model is used to infer end-user preferences based on terminal applications. The first model is trained in the following way: Each application name sample in the application name sample training set is labeled to indicate the user preferences for each application name sample. The user preferences are associated with all or part of the information in the application name sample. The neural network is then trained using the labeled application name sample training set to obtain the model.

[0026] The application name sample contains at least one application name, meaning it can contain one or more application names. For example, a sample application name might contain the names of three applications already installed on a device: "Himalaya," "Acfun," and "Hema." User preferences can include tags such as "food," "beauty," "entertainment," "learning," "education and training," "scenery," and "cute pets." User preference data is obtained through daily data collection, such as data collected from users' searches on shopping apps. The search input includes, but is not limited to, text, voice, and images. The search input is then analyzed using mature text analysis, voice analysis, and image analysis models to obtain the corresponding user preferences.

[0027] S3. Calculate the matching degree between each advertisement in the advertisement database and the first user's preferences; The advertising information refers to the advertising segments provided by the supplier, which can be presented in the form of text or video. Each advertising information is associated with a preference tag. The preference tags can be "food", "travel", "scenery", "books", "preschool education", "myopia", "games", "cute pets", etc. Most preference tags and like tags are the same.

[0028] Combination Figure 2 S3 specifically includes: S31. Count the number of identical tags between the preference tags associated with the advertising information and multiple first preference tags in the first user's preferences; Suppose that the current system's database contains 50 advertising items, each of which is associated with a salary preference tag. Then, the number of identical tags between the preference tags associated with each advertising item and multiple first preference tags in the first user's preferences is counted. For example, if an advertising item is associated with 20 preference tags and the first user's preferences contain 16 first preference tags, then the number of identical tags between the advertising item and the first user's preferences is 10. S32. Obtain the smaller value between the number of preference tags associated with the advertising information and the number of multiple first preference tags in the first user's preferences as the comparison value; Continuing with the previous example, the smaller value of 16 items between 20 items and 16 items will be used as the comparison value.

[0029] S33. The ratio between the same number of tags and the value to be compared is used as the matching degree.

[0030] Then, the ratio of 62.5% between 10 items with the same label and 16 items to be compared is taken as the matching degree.

[0031] S4. Select advertising information that meets the preset matching requirements as the advertising to be pushed. Specifically, the preset requirement is to exceed the preset matching degree. For example, if the preset matching degree is set to 60%, then 62.5% of the matching degree exceeds the preset matching degree of 60%, and the corresponding advertising information will be used as the advertising to be pushed.

[0032] S5. Push ads with a matching degree exceeding the preset value to the user's terminal.

[0033] The specific push method can be to push advertising information on a periodic basis, or to sort multiple ads to be pushed and push them one by one.

[0034] Combination Figure 3 In one embodiment, the AI-based big data precision marketing algorithm further includes: S6. Obtain the name of the application uninstalled by the user's terminal; That is, the name of the application to be uninstalled is obtained each time the terminal executes the command to uninstall the application. For example, when the user uninstalls the "MiLe" application, the name of the application is obtained. S7. Input the name of the uninstalled application into the first model and infer the second user preference; The second user preference includes multiple second preference tags. The name of the "MiLe" application is input into the first model mentioned above to infer the second user preference. Since the second user preference is inferred from a single application name, the number of second preference tags in the second user preference will be less than that in the first user preference, which may be inferred from multiple application names. This is because a user having multiple applications indicates a finer granularity of user, a more accurate description of the user, and therefore fewer and more precise preference tags. On the other hand, a single application may correspond to many preferences, which is not a precise description of the user, and therefore has more tags.

[0035] S8. Input the name of the uninstalled application into the second model and infer the user's aversion. User aversion includes multiple aversion tags; Specifically, the second model is used to infer user aversion based on the applications on the terminal. For example, if a user uninstalls an application related to "preschool education," the application is retrieved. Aversion label data is obtained through routine data collection, such as changes in the content retrieved by the user in shopping apps after uninstalling the application. The input content for retrieval includes, but is not limited to, text, voice, and images. The preferences analyzed from the content retrieved before uninstallation are compared with those analyzed from the content retrieved after uninstallation, and the reduced labels are the aversion labels. The second model is trained in the following way: Each application name sample in the application name sample training set is labeled to identify user aversions associated with each application name sample, and user aversions are associated with all or part of the information in the application name sample; and a neural network is trained using the labeled application name sample training set to obtain a second model.

[0036] S9. Obtain the tags to be deleted based on the second user's preferences and user dislikes; Specifically, tags that are identical between multiple second-preference tags in the second user preferences and multiple dislike tags in the user dislikes are designated as tags to be deleted.

[0037] S10. Delete the first preference tag that is the same as the tag to be deleted in the first user preferences.

[0038] By removing the user's first preference after deleting the tags to be deleted, we can reduce the amount of content that users no longer need or like when pushing marketing content, thereby increasing user acceptance and reducing user aversion.

[0039] This application also provides a large-scale precision marketing system based on artificial intelligence, including: The application acquisition module is used to obtain the names of applications already installed on the user's terminal. The first preference inference module is used to input the name of the installed application into the first model and infer the first user preference, which includes multiple first preference tags. The calculation module is used to calculate the matching degree between each advertisement in the advertising database and the first user's preferences. Each advertisement is associated with a preference tag. The push-to-ads module is used to select advertising information that meets preset matching requirements as ads to be pushed. The push module is used to push advertisements with a matching degree exceeding a preset value to the user's terminal.

[0040] The calculation module includes: The quantity statistics unit is used to count the number of identical tags between the preference tags associated with the advertising information and multiple first preference tags in the first user's preferences; The unit for obtaining the comparison value is used to obtain the smaller value between the number of preference tags associated with the advertising information and the number of multiple first preference tags in the first user preferences as the comparison value; The matching degree calculation unit is used to take the ratio between the number of identical tags and the value to be compared as the matching degree.

[0041] In one embodiment, the AI-based large-scale precision marketing system further includes: The uninstalled application acquisition module is used to obtain the names of applications uninstalled by the user's terminal; The second preference reasoning module is used to input the name of the uninstalled application into the first model and reason to obtain the second user preference, which includes multiple second preference tags. The aversion reasoning module is used to input the name of the uninstalled application into the second model and reason to obtain the user's aversion. The user's aversion includes multiple aversion tags. The "to be deleted" module is used to obtain tags to be deleted based on the second user's preferences and the user's dislikes; The deletion module is used to delete first preference tags that are the same as the tag to be deleted in the first user preferences.

[0042] For specific limitations regarding AI-based large-scale precision marketing systems, please refer to the limitations of AI-based large-scale precision marketing methods mentioned above, which will not be repeated here. Each step of the aforementioned AI-based large-scale precision marketing method can be implemented entirely or partially through software, hardware, or a combination thereof.

[0043] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, application-specific integrated circuits (ASICs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0044] These computational programs (also referred to as programs, software, software applications, or code) include machine instructions for a programmable processor and can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. As used herein, the terms “machine-readable medium” and “computer-readable medium” refer to any computer program product, device, and / or apparatus (e.g., disk, optical disk, memory, programmable logic device (PLD)) used to provide machine instructions and / or data to a programmable processor, including machine-readable media that receive machine instructions as machine-readable signals. The term “machine-readable signal” refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0045] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0046] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0047] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application can be achieved, and this is not limited herein.

[0048] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A large-scale precision marketing algorithm based on artificial intelligence, characterized in that: include: Get the names of applications installed on the user's terminal; Input the name of the installed application into the first model, and infer the first user preference, which includes multiple first preference tags. Calculate the matching degree between each advertisement in the advertising database and the first user's preferences, with each advertisement associated with a preference tag; Ads that meet the preset matching requirements will be designated as ads to be pushed to users. Push ads with a match rate exceeding a preset value to the user's device; Also includes: Get the name of the application uninstalled by the user's terminal; Input the name of the uninstalled application into the first model, and infer the second user preference, which includes multiple second preference tags. The uninstalled application name is input into the second model, and user aversion is inferred. The user aversion includes multiple aversion tags. The tags to be deleted are obtained based on the second user's preferences and user dislikes; Delete the first preference tag that is the same as the tag to be deleted in the first user preferences.

2. The AI-based precision marketing algorithm for large-scale marketing as described in claim 1, characterized in that, Calculate the match degree between each advertisement in the advertising database and the first user's preferences, including: The number of identical tags between the preference tags associated with the advertising information and multiple first preference tags in the first user's preferences; The smaller of the number of preference tags associated with the advertising information and the number of multiple first preference tags in the first user's preferences is used as the comparison value; The ratio between the number of identical tags and the value to be compared is used as the matching degree.

3. The AI-based precision marketing algorithm for large-scale marketing as described in claim 1, characterized in that, The installed application names do not include the basic application names in the basic application library.

4. The AI-based precision marketing algorithm for large-scale marketing as described in claim 1, characterized in that, The process of obtaining the tags to be deleted based on second user preferences and user aversions includes: Tags that are identical between multiple second-preference tags in the second user preferences list and multiple dislike tags in the user dislike list are designated as tags to be deleted.

5. A large-scale precision marketing system based on artificial intelligence, characterized in that: include: The application acquisition module is used to obtain the names of applications already installed on the user's terminal. The first preference inference module is used to input the name of the installed application into the first model and infer the first user preference, which includes multiple first preference tags. The calculation module is used to calculate the matching degree between each advertisement in the advertisement database and the first user's preferences. Each advertisement is associated with a preference tag. The push-to-ads module is used to select advertising information that meets preset matching requirements as ads to be pushed. The push module is used to push advertisements with a matching degree exceeding a preset value to the user's terminal; Also includes: The uninstalled application acquisition module is used to obtain the names of applications uninstalled by the user's terminal; The second preference reasoning module is used to input the name of the uninstalled application into the first model and reason to obtain the second user preference, which includes multiple second preference tags. The aversion reasoning module is used to input the name of the uninstalled application into the second model and reason to obtain the user's aversion, which includes multiple aversion tags. The "to be deleted" module is used to obtain tags to be deleted based on the second user's preferences and the user's dislikes; The deletion module is used to delete a first preference tag that is the same as the tag to be deleted in the first user preferences.

6. The AI-based precision marketing system for large-scale markets as described in claim 5, characterized in that, Also includes: The uninstalled application acquisition module is used to obtain the names of applications uninstalled by the user's terminal; The second preference reasoning module is used to input the name of the uninstalled application into the first model and reason to obtain the second user preference, which includes multiple second preference tags. The aversion reasoning module is used to input the name of the uninstalled application into the second model and reason to obtain the user's aversion, which includes multiple aversion tags. The "to be deleted" module is used to obtain tags to be deleted based on the second user's preferences and the user's dislikes; The deletion module is used to delete a first preference tag that is the same as the tag to be deleted in the first user preferences.

7. The AI-based precision marketing system for large-scale markets as described in claim 5, characterized in that, The computing module includes: The quantity statistics unit is used to count the number of identical tags between the preference tags associated with the advertising information and multiple first preference tags in the first user's preferences; The unit for obtaining the comparison value is used to obtain the smaller value between the number of preference tags associated with the advertising information and the number of multiple first preference tags in the first user preferences as the comparison value; The matching degree calculation unit is used to take the ratio between the number of identical tags and the value to be compared as the matching degree.

Images