Information pushing method and system, and object recommendation point description information pool generation method

Abstract

An information pushing method and system, and an object recommendation point description information pool generation method, relating to the technical field of artificial intelligence. The information pushing method comprises: in response to an object operation request, acquiring an object operation scenario and a target object which match the object operation request, and user information of a user; acquiring, from a preset object recommendation point description information pool, a specified number of pieces of object recommendation point description information matching the target object and the object operation scenario; selecting, from the specified number of pieces of object recommendation point description information, one piece of object recommendation point description information matching the user information as target recommendation point information; and pushing the object information to the user on the basis of the target recommendation point information. The information pushing method implements information pushing by combining a full-link scenario of a specified platform with the user information, thus effectively improving the matching level of pushed information, and improving the decision-making efficiency of users.

Description

Information push methods, systems, object recommendation point descriptions, and information pool generation methods

[0001] This disclosure claims priority to Chinese Patent Application No. 202510107792.0, filed with the China Patent Office on January 22, 2025, entitled "Information Push Method, System, and Method for Generating Object Recommendation Point Description Information Pool", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This disclosure relates to the field of artificial intelligence technology, and in particular to an information push method, an information push system, a method for generating an object recommendation point description information pool, an electronic device, a storage medium, and a computer program product. Background Technology

[0003] In e-commerce platforms, pushing information related to target products to users can improve the efficiency of users discovering target products and submitting orders. The timing and content of this information push are particularly important. Analysis of existing cross-border e-commerce platforms reveals that buyers repeatedly check product details (an average of 60+ times) or communicate (an average of 90+ times) to supplement their order decision-making information during the order conversion process, resulting in low decision-making efficiency. The main reason for this problem is the low match between the timing and content of the information pushed to buyers and their needs.

[0004] It is evident that existing information push methods still need improvement. Summary of the Invention

[0005] This disclosure provides an information push method that can improve the efficiency and accuracy of information push, thereby improving order generation efficiency.

[0006] Accordingly, this disclosure also provides a method for generating an object recommendation point description information pool, an information push system, an electronic device, a storage medium, and a computer program product to ensure the implementation and application of the above-mentioned information push method.

[0007] To address the aforementioned problems, this disclosure provides an information push method applied to a server, the method comprising:

[0008] In response to an object operation request, obtain the object operation scenario matching the object operation request, the target object, and the user's user information;

[0009] Obtain a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool;

[0010] Select one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions, and use it as the target recommendation point information;

[0011] Based on the target recommendation point information, object information is pushed to the user.

[0012] This disclosure provides an information push method applied to a client, the method comprising:

[0013] In response to a page operation performed by a user on the client, an object operation request is generated based on the user's user information, the target object matched by the page operation, and the object operation scenario.

[0014] The system sends the object operation request to a preset server to trigger the server to perform the following operations: In response to the object operation request, the system obtains the object operation scenario, target object, and user information that match the object operation request; it obtains a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool; it selects one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information; and it generates page display data that matches the object operation request based on the target recommendation point information.

[0015] Based on the page display data, the page display content of the client is updated to push the target recommendation point information to the user.

[0016] This disclosure provides a method for generating an object recommendation point description information pool, the method comprising:

[0017] From the data source corresponding to the preset object recommendation point type, obtain the source data associated with the target object, and obtain the source data associated with the target object and corresponding to the preset object recommendation point type;

[0018] Extract the descriptive text corresponding to the preset object recommendation point type from the source data;

[0019] Using artificial intelligence technology, object recommendation point description information matching the corresponding preset object recommendation point type is generated based on the description text corresponding to the corresponding preset object recommendation point type.

[0020] Based on the generated object recommendation point description information, an object recommendation point description information pool for the target object is created.

[0021] This disclosure provides an information push system, including: a client, a server, and a pre-created pool of object recommendation point description information corresponding to a target object.

[0022] The client is configured to respond to a page operation performed by a user on the client, and generate an object operation request based on the user's user information, the target object matched by the page operation, and the object operation scenario.

[0023] The client is also used to send the object operation request to the server;

[0024] The server is configured to respond to the object operation request by obtaining the object operation scenario, target object, and user information that match the object operation request; obtaining a specified number of object recommendation point descriptions that match the target object and the object operation scenario from the object recommendation point description information pool; selecting one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information; and generating page display data that matches the object operation request based on the target recommendation point information.

[0025] The server is also used to send the page display data to the client;

[0026] The client is also configured to update the page display content of the client based on the page display data, so as to push the target recommendation point information to the user.

[0027] This disclosure also discloses a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the methods described in this disclosure.

[0028] This disclosure also discloses a computer program product, including a computer program / computer executable instructions, wherein the computer program / computer executable instructions, when executed by a processor in an electronic device, implement the method described in this disclosure.

[0029] Compared with the prior art, the embodiments of this disclosure have the following advantages:

[0030] In response to an object operation request, the server obtains the object operation scenario, target object, and user information that match the request. It then retrieves a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a pre-defined object recommendation point description information pool. Finally, it selects one object recommendation point description that matches the user information from these specified number of descriptions as the target recommendation point. Based on this target recommendation point, it pushes object information to the user. This achieves information push based on user information across the entire process on a specified platform, effectively improving the matching accuracy of the pushed information and thus enhancing the user's decision-making efficiency. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this disclosure 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 some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 is a flowchart of the information push method disclosed in an embodiment of this disclosure;

[0033] Figure 2 is a schematic diagram of an application scenario of the information push method disclosed in this embodiment.

[0034] Figure 3 is a schematic diagram of the creation principle of the object recommendation point description information pool disclosed in this embodiment;

[0035] Figure 4 is a flowchart of another step of the information push method disclosed in this embodiment;

[0036] Figure 5 is a flowchart of the steps of the object recommendation point description information pool generation method disclosed in this embodiment of the present disclosure;

[0037] Figure 6 is a schematic diagram of the interaction of the information push system disclosed in this embodiment;

[0038] Figure 7 is a schematic diagram of the structure of an exemplary device provided in an embodiment of this disclosure. Detailed Implementation

[0039] Many specific details are set forth in the following description to provide a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0040] The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this specification. The singular forms “a,” “described,” and “the” as used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0041] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this specification, and similarly, second may also be referred to as first. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0042] Furthermore, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties. Moreover, the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0043] The information push method disclosed in this embodiment can be applied to e-commerce platforms, information service platforms, and other platforms. This method constructs a multi-dimensional information push system that integrates data from both inside and outside the platform, is based on artificial intelligence technology, matches specific application scenarios, and fully considers category advantages. Taking the application of the information push method to an e-commerce platform as an example, specifically referring to the application scenario shown in Figure 2, the information push method disclosed in this embodiment first integrates platform-internal data such as product titles, prices, reviews, descriptions, and sales volume, as well as external data such as downstream e-commerce data, market trend data, holiday and promotional activity data, customs and regulations data, and social media topics, as source data. Then, using artificial intelligence technology, object recommendation point description information is generated based on the source data to establish an object recommendation point description information pool including several object recommendation point description information. During the operation of an e-commerce platform, buyers perform a series of operations through the platform's client, such as querying products, browsing product details, submitting orders, and making payments. After obtaining user information such as the buyer's country, user preferences, and logistics preferences, the client uses artificial intelligence technology to select a recommended item description from the pool of recommended item description information and push it to the buyer, thereby effectively improving the buyer's decision-making efficiency during the order submission process.

[0044] Referring to Figure 1, the information push method disclosed in this embodiment is applied to the server of an information push system, and the method includes steps 102 to 106.

[0045] Step 102: In response to the object operation request, obtain the object operation scenario, target object, and user information that match the object operation request.

[0046] As shown in Figure 6, the information push system includes a client 602 and a server 604. The client 602 responds to user operations by acquiring the object operation scenario, the target object being operated on, and the user's user information. Then, based on the object operation scenario, the target object, and the user's user information, it generates an object operation request and sends the object operation request to the server 604.

[0047] The user information can be obtained by client 602 and / or server 604 when a user logs into the information push system through client 602. The user information includes, but is not limited to, user identifier, matched country, currently selected language, user profile information, and user behavior information. The user identifier uniquely identifies the user in the information push system; the matched country and currently selected language can be obtained by client 602 through real-time collection of client 602's geographical location, or determined based on user settings on client 602, or obtained from user settings pre-stored by server 604.

[0048] The server 604 is used to respond to the object operation request sent by the client 602, obtain the object operation scenario, target object, and user information matched by the object operation request, and perform subsequent object recommendation point description information acquisition and push operation based on the object operation scenario, the target object, and the user information.

[0049] The object operation scenario is determined by the client 602. For information push methods applied in different scenarios, the target object and the object operation scenario differ. Taking the application of the information push method disclosed in this embodiment to an e-commerce platform as an example, the target object can be a product within the e-commerce platform. Correspondingly, the object operation scenario can be each scenario in the entire product operation chain, including but not limited to one or more of the following scenarios: third-party platform traffic generation scenario, platform-wide search scenario, product details scenario, order placement scenario, payment scenario, etc. Taking the application of the information push method disclosed in this embodiment to an information service platform as an example, the target object can be a topic on the information service platform. Correspondingly, the object operation scenario can be each scenario in the entire topic operation chain, including but not limited to one or more of the following scenarios: platform-wide topic search scenario, topic details scenario, subtopic selection scenario, question submission scenario, etc.

[0050] The information push method disclosed in the embodiments of this disclosure can also be applied to other application scenarios, which will not be listed one by one in the embodiments of this disclosure. In order to facilitate readers' understanding of this solution, the following example illustrates the specific implementation of each step of the information push method by applying the information push method disclosed in the embodiments of this disclosure to an e-commerce platform.

[0051] In the embodiments disclosed herein, there are no restrictions on the specific implementation of obtaining object operation requests, nor are there any restrictions on the specific implementation of obtaining the object operation scenario, target object, and user information matched with the object operation request.

[0052] Step 104: Obtain a specified number of object recommendation point descriptions that match the target object and the object operation scenario from the preset object recommendation point description information pool.

[0053] In the embodiments of this disclosure, in order to push matching information to users by combining multiple dimensions of information, including specific application scenarios, object information, and user information, several object recommendation points are obtained, and an object recommendation point description information pool is pre-established. The object recommendation point description information pool includes multiple object recommendation point descriptions. The object recommendation point description information pool can correspond one-to-one with each object, or it can correspond one-to-one with each type of object.

[0054] In some optional embodiments, the preset object recommendation point description information pool is created by the following method: obtaining source data associated with the target object from data sources corresponding to preset object recommendation point types, thereby obtaining various source data associated with the target object corresponding to the preset object recommendation point types; extracting descriptive text corresponding to the corresponding preset object recommendation point types from the source data corresponding to the preset object recommendation point types; using artificial intelligence technology, generating object recommendation point description information matching the corresponding preset object recommendation point types based on the descriptive text corresponding to the corresponding preset object recommendation point types; and creating the object recommendation point description information pool for the target object based on the generated object recommendation point description information.

[0055] The type of preset object recommendation points is determined based on factors influencing the operation of the preset object. For example, in an e-commerce platform, when the target object is a product, the preset object recommendation point type includes one or more of the following: market trend type, price advantage type, major event type, hot event type, positive C-end reviews type, and platform guarantee type. Specifically, the market trend type object recommendation point description information describes the sales trend of the target product; the price trend type recommendation point information describes the price change trend of the target product; the major event type recommendation point information describes events related to the order quantity of the target product; the positive C-end reviews type recommendation point information describes the positive reviews of the target product from users on the e-commerce platform; and the platform guarantee type object recommendation point description information describes the guarantee policies of the e-commerce platform.

[0056] The following section, using the schematic diagram of the object recommendation point description information pool creation principle shown in Figure 3, further elaborates on the creation method of the object recommendation point description information pool.

[0057] As shown in Figure 3, creating an object recommendation point description information pool includes two stages: the first stage is to acquire source data; the second stage is to generate object recommendation point description information and establish an object recommendation point description information pool corresponding to the preset object recommendation point type.

[0058] The first stage involves acquiring source data used to generate object recommendation point description information.

[0059] The data source corresponding to the preset object recommendation point type includes one or more of the following: e-commerce platforms, social media platforms, market trend analysis platforms, search engines, holiday databases, and recommendation script template libraries. In some optional embodiments, source data for generating object recommendation point description information can be obtained from the above data sources.

[0060] Optionally, the source data includes: platform-internal data and external data of the e-commerce platform implementing the information push method disclosed in this embodiment. The platform-internal data includes, but is not limited to, one or more of the following: site search logs, basic product information, marketing activities, merchant recruitment activities, product selection collections, and basic transaction data of the e-commerce platform; the external data includes, but is not limited to, one or more of the following: search data from search engines, product data from other e-commerce platforms, analysis data from market trend analysis platforms, C-end evaluation data from social media, recommendation script template libraries, and holiday databases. In the embodiments of this disclosure, the data source corresponding to the recommendation point description information of different types of objects can be determined according to specific application requirements, thereby obtaining the source data corresponding to the recommendation point description information of different types of objects.

[0061] For example: the data sources corresponding to market trend-related object recommendation points include, but are not limited to: market trend analysis data from industry analysis platforms, industry forum data, and downstream e-commerce data; the data sources corresponding to price advantage-related object recommendation point descriptions include, but are not limited to: price trend analysis data from industry analysis platforms and product data from e-commerce websites; the data sources corresponding to major event-related object recommendation point descriptions include, but are not limited to, one or more of the following: holiday databases and promotional activity log databases; the data sources corresponding to hot event-related object recommendation point descriptions include, but are not limited to: local customs databases and holiday databases; the data sources corresponding to C-end positive review-related object recommendation points include, but are not limited to: e-commerce website review databases; and the data sources corresponding to platform guarantee-related object recommendation point descriptions include, but are not limited to: e-commerce platform platform guarantee rules.

[0062] The data sources corresponding to the above object recommendation point types are merely illustrative and are not intended to limit the correspondence between object recommendation point types and data sources.

[0063] In some optional embodiments, an automated acquisition mechanism for external data (such as market trend analysis data from industry analysis platforms, industry forum data, and downstream e-commerce data) can be built through content aggregation, data services, and third-party data procurement to automatically acquire external data. Then, the source data from each data source can be preprocessed according to preset rules to facilitate the subsequent extraction of recommendation point information from the source data.

[0064] On the other hand, by fully leveraging the capabilities of artificial intelligence in text understanding, image-text matching, and knowledge base mapping, the platform performs data cleaning and coarse classification on both internal and external data through operations such as category matching, normalization based on shared keywords, normalization based on shared objects, and correlation based on major events. This results in source data indexed by objects. Each piece of source data is then tagged to indicate its data source.

[0065] The second stage involves generating recommendation point information.

[0066] In some optional embodiments, an object recommendation point description information pool is pre-created for each object. This object recommendation point description information pool is generated after analyzing and extracting information from data on the object from multiple data sources.

[0067] Below, we will take product A as an example to illustrate how to generate the object recommendation point description information for product A.

[0068] First, retrieve the associated data for product A from each of the aforementioned data sources, using it as the source data for the corresponding data source for product A. For example, retrieve the associated data for product A from market trend analysis data from industry analysis platforms, industry forum data, downstream e-commerce data, price trend analysis data, basic product data from e-commerce websites, promotional activity data, review data from e-commerce websites, and platform guarantee rules from e-commerce platforms, using it as the source data for product A corresponding to each of the aforementioned data sources.

[0069] Subsequently, the source data for product A obtained from each data source is processed to extract descriptive text for each preset object recommendation point type. In some optional embodiments, a pre-tuned second generative large language model can be invoked to process the source data obtained from each data source, extracting words and / or phrases corresponding to the corresponding preset object recommendation point type as descriptive text. For example, the pre-tuned second generative large language model can be invoked to extract market trend words and / or phrases from market trend data sources as descriptive text corresponding to market trends. Another example is invoking the pre-tuned second generative large language model to extract words and / or phrases evaluating product A from e-commerce website review data as descriptive text corresponding to positive C-end reviews.

[0070] When the second generative large language model is invoked to process the source data obtained from the specified data source and extract words and / or phrases corresponding to the preset object recommendation point type from the source data, prompt words can be generated based on the source data and the demand description text containing the words and / or phrases corresponding to the specified object recommendation point type extracted from the source data. The generated prompt words are then used to invoke the second generative large language model, triggering the generation of words and / or phrases corresponding to the specified object recommendation point type. For example, the following prompt words can be generated: "The following are the collected evaluation information for product A. Please extract positive C-end evaluation information for product A based on this evaluation information: ...."

[0071] The descriptive text corresponding to each object recommendation point type for product A is extracted using the method described above. Then, artificial intelligence technology is used to generate object recommendation point description information matching the corresponding preset object recommendation point type for each of the descriptive texts. In some optional embodiments, a pre-tuned third generative large language model can be invoked to generate object recommendation point description information matching the corresponding preset object recommendation point type based on the descriptive texts corresponding to each preset object recommendation point type. For example, the pre-tuned third generative large language model can be invoked to generate object recommendation point description information matching market trends based on the descriptive texts corresponding to market trends. Another example is the invocation of the pre-tuned third generative large language model to generate object recommendation point description information matching C-end positive reviews based on the descriptive texts corresponding to C-end positive reviews.

[0072] When invoking a pre-tuned third generative large language model to generate object recommendation point description information matching the specified object recommendation point type based on the description text corresponding to that object recommendation point type, prompt words can be generated based on the description text corresponding to the specified object recommendation point type and the requirement description text for generating the object recommendation point description information. The third generative large language model is then invoked based on these prompt words to trigger the generation of object recommendation point description information corresponding to the specified object recommendation point type. For example, the following prompt words can be generated: "The following are the collected C-end review description texts for product A. Please generate C-end positive review descriptions for product A based on these words. Product A C-end review description text:..."

[0073] Following the method described above, object recommendation point description information for each object recommendation point type of product A is generated. The object recommendation point description information for all object recommendation point types of product A constitutes the object recommendation point description information pool for product A. For example, first, description information for positive C-end reviews of product A is generated; then, based on generating multiple descriptions of positive C-end reviews, a pool of positive C-end reviews is created. As another example, first, description information for market trend-related selling points of product A is generated; then, based on generating multiple descriptions of market trend-related selling points, a pool of market trend-related selling points is created.

[0074] In some alternative embodiments, an object recommendation point description information pool is pre-created for each type of object. This object recommendation point description information pool is generated after analyzing and extracting information from data of the corresponding category of object on multiple data sources.

[0075] Accordingly, the step of obtaining source data associated with the target object from the data source corresponding to the preset object recommendation point type, and obtaining various source data associated with the target object corresponding to the preset object recommendation point type, includes: based on the second preset object information, performing clustering processing on the objects in the data source corresponding to the preset object recommendation point type to obtain several object clusters; obtaining the source data associated with each object in the object cluster where the target object is located in each data source corresponding to the preset object recommendation point type, as the various source data associated with the target object corresponding to the preset object recommendation point type. Wherein, the second preset object information may be: object image and object title; the target platform may be an e-commerce platform, an information service platform, etc.

[0076] First, for the source data from all data sources corresponding to the preset object recommendation point type obtained using the method described in the first stage above, clustering is performed based on the similarity between object images and object titles, grouping similar objects of the same category and with the same search keywords into the same object cluster. Then, for each object cluster, the association data of all objects in that cluster across various data sources is obtained, serving as the association data between that object cluster and each data source. Finally, for each object cluster, the association data between that object cluster and each data source is used as the association data between each object in that cluster and each data source.

[0077] Taking product A as an example again, the second preset object information includes: the image and title of product A. During the clustering process based on the second preset object information, the products in the data source corresponding to the preset object recommendation point type can be clustered based on the product's image similarity and title similarity. With the goal of grouping similar products by keywords, similar products of the same category and with the same keywords are clustered into the same cluster, resulting in several product clusters. Then, the association data of all products in the product cluster where product A is located in various data sources is obtained, serving as the source data for product A in the corresponding type of data source.

[0078] Next, referring to the method described above, descriptive text corresponding to the respective preset object recommendation point types is extracted from the source data corresponding to those preset object recommendation point types. Then, using artificial intelligence technology, object recommendation point description information matching the corresponding preset object recommendation point type is generated based on the descriptive text corresponding to each preset object recommendation point type. Based on the generated object recommendation point description information, an object recommendation point description information pool for the target object is created. According to the aforementioned clustering process, the object recommendation point description information pool for the target object is the object recommendation point description information pool for all objects in the same object cluster as the target object.

[0079] In some optional embodiments, obtaining a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool includes: obtaining first preset object information of the target object, and obtaining object recommendation point descriptions from the preset object recommendation point description information pool that match the target object; using artificial intelligence technology, filtering from the obtained object recommendation point descriptions to obtain a specified number of object recommendation point descriptions that match both the first preset and the object operation scenario, as the specified number of object recommendation point descriptions that match the target object and the object operation scenario.

[0080] In some optional embodiments, prompt words can be generated based on the first preset object information, the acquired object recommendation point description information, and the object operation scenario. These prompt words guide the first generative large language model to select object recommendation point description information from the acquired object recommendation point description information that matches the first preset object information and the object operation scenario. Then, the first generative large language model is invoked based on the prompt words. The first generative large language model will select a specified number of object recommendation point descriptions from the acquired object recommendation point description information based on the matching relationship between the first preset object information, the object recommendation point description information, and the object operation scenario, as the specified number of object recommendation point descriptions matching the target object and the object operation scenario.

[0081] Taking the first preset object information as including: object image and / or object title, for a product details browsing scenario, prompts in the following form can be generated based on the object information, the obtained object recommendation point description information, and the object operation scenario:

[0082] "Object Image: Picture"

[0083] Object title: Title

[0084] The above are the object image and object title. Please select approximately three object recommendation point descriptions from the following list that best match the object image and object title and are suitable for browsing product details:

[0085] 'Description information for object recommendation point 1'

[0086] 'Description information for object recommendation point 2'

[0087] 'Object Recommendation Point 3 Description Information'

[0088] 'Object Recommendation Point 4 Description Information'

[0089] 'Object Recommendation Point 5 Description Information'.

[0090] Subsequently, based on the prompt words in the above form, the first generative large language model is invoked. The first generative large language model will select 3 object recommendation point descriptions from the description information of the above object recommendation points 1 to 5 as object recommendation point descriptions that match the target object and the object operation scenario.

[0091] The aforementioned first, second, and third generative large language models can be obtained by using existing large language models as a base and fine-tuning them with data from specific application scenarios. The structures and fine-tuning methods of the first, second, and third generative large language models can be found in existing technologies and will not be repeated in this embodiment.

[0092] In some optional embodiments, the first, second, and third generative large language models obtained through fine-tuning can be further evaluated and corrected based on manually labeled training samples to improve the understanding of object recommendation point information by the large language models in the corresponding scenarios.

[0093] Step 106: Select one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions, and use it as the target recommendation point information.

[0094] After obtaining multiple object recommendation points that match the application scenario and target object, further filtering is performed to select object recommendation points that are suitable for the current user, thereby improving the user's decision-making efficiency.

[0095] In some optional embodiments, the user information includes: user profile information and / or user behavior information. Selecting one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information includes: obtaining a first user feature of the user based on the user profile information, and / or obtaining a second user feature of the user based on the user behavior information; and using artificial intelligence technology to select one object recommendation point description that matches the first user feature and / or the second user feature from the specified number of object recommendation point descriptions as the target recommendation point information.

[0096] Optionally, the user profile information can be user-preset preference information, or it can be collected based on the user's historical behavior data after the user authorizes the information push system to enable personalized information push. The user profile information includes, but is not limited to, one or more of the following: basic information, behavioral data, consumption habits, social information, geographical information, etc. The server of the information push system performs user preference identification based on the user profile information to obtain the user's first user characteristics.

[0097] On the other hand, the user behavior information includes, but is not limited to, browsing operations, search operations, order submission operations, and other behaviors performed by the user on the client within a specified time period, as well as the behavior trajectory. The server of the information push system mines the user's operational behavior characteristics towards the target object based on the user behavior information to obtain a second user characteristic.

[0098] For specific implementation methods of obtaining the user's first user feature based on the user profile information and obtaining the user's second user feature based on the user behavior information, please refer to the prior art, and will not be repeated in the embodiments of this disclosure.

[0099] After obtaining the user's first user characteristics and second user characteristics, the server further employs artificial intelligence technology to select one object recommendation point description that matches the first user characteristics and / or the second user characteristics from the specified number of object recommendation point descriptions obtained in the aforementioned steps, based on the matching degree between the object recommendation point description information and the first user characteristics and / or the second user characteristics, as the target recommendation point information matching the user.

[0100] For example, prompt words can be generated based on the first user characteristic and the specified number of object recommendation point descriptions. Based on these prompt words, a pre-tuned fourth generative model can be triggered to select one object recommendation point description from the specified number of object recommendation point descriptions that matches the first user characteristic. Taking a product as the target object, assuming the user's first user characteristic indicates a preference for highly rated products, the following prompt words can be generated: "The following are several selling points of a product. Now, a buyer who prefers highly rated products should select the most suitable selling point description: Object Recommendation Point 1, Object Recommendation Point 2, and Object Recommendation Point 3." After calling the pre-tuned fourth generative model based on these prompt words, the fourth generative model will select and output the object recommendation point description that best matches the first user characteristic from the prompt words.

[0101] For example, prompt words can be generated based on the first user feature, the second user feature, and the specified number of object recommendation point descriptions. Based on these prompt words, a pre-tuned fourth generative model is triggered to select one object recommendation point description from the specified number of object recommendation point descriptions that matches the first user feature. Taking the target object as a product as an example, assuming the first user feature indicates that the user prefers highly rated products, and the second user feature indicates that the user has repeatedly exited the order page, then when the user exits the order page, the following prompt words can be generated: "The following are several selling points of a product. One selling point has already been included in the order page. Now there is a buyer who prefers products with high ratings, but after looking for a while, they are considering leaving. Please select one selling point description to retain the buyer: Object Recommendation Point 1, Object Recommendation Point 2, Object Recommendation Point 3." After calling the pre-tuned fourth generative model based on these prompt words, the fourth generative model will select one object recommendation point description from the prompt words to retain the user and output it.

[0102] In some optional embodiments, the user information further includes: the language of the user matching; after selecting one object recommendation point description that matches the first user feature and / or the second user feature from the specified number of object recommendation point description information using artificial intelligence technology as the target recommendation point information, the method further includes: optimizing the target recommendation point information based on the object operation scenario and the language to obtain optimized target recommendation point information.

[0103] In some optional embodiments, optimizing the target recommendation point information based on the object operation scenario and the language to obtain optimized target recommendation point information includes: optimizing the content of the target recommendation point information based on the information display length matched by the object operation scenario, so that the length of the optimized target recommendation point information matches the information display length; and converting the target recommendation point information to the language to obtain target recommendation point information in the specified language.

[0104] In practical implementation, different information display pages have limited content length displayed in information display slots, while the length of object recommendation point information generated by the generative large model is uncertain, which may result in incomplete display of recommendation point information. In the embodiments of this disclosure, artificial intelligence technology can be used to optimize the content of the target recommendation point information, ensuring that the length of the optimized target recommendation point information is equal to or less than the information display length. In some optional embodiments, when the language used by the user differs from the language of the object recommendation point information generated by the generative large model, the server can call the generative large language model to convert the target recommendation point information into the user's language, thereby improving the user adaptability of the target recommendation point information.

[0105] Step 108: Based on the target recommendation point information, push object information to the user.

[0106] Next, the server pushes the target recommendation point information to the user through the client.

[0107] In some optional embodiments, the step of pushing object information to the user based on the target recommendation point information includes: dynamically obtaining the display slot matching the target recommendation point information; assembling the preset display information of the target object and the target recommendation point information according to the display slot to obtain information to be displayed; and displaying the information to be displayed to the user to push the target recommendation point information to the user.

[0108] For example, the server retrieves the page data corresponding to the object operation scenario and the display slot matching the target recommendation point information. Then, it inserts the target recommendation point information into the display slot to update the information to be displayed on the page corresponding to the object operation scenario. Next, the server sends the updated information to be displayed on the page corresponding to the object operation scenario to the user's currently logged-in client, and the client displays the page corresponding to the object operation scenario based on the updated information. Thus, the server completes the process of pushing the target recommendation point information to the user through the page corresponding to the object operation scenario.

[0109] In summary, the information push method disclosed in this embodiment involves the server responding to an object operation request by obtaining the object operation scenario, target object, and user information matching the object operation request; retrieving a specified number of object recommendation point descriptions matching the target object and the object operation scenario from a preset object recommendation point description information pool; selecting one object recommendation point description matching the user information from the specified number of object recommendation point descriptions as the target recommendation point information; and pushing object information to the user based on the target recommendation point information. This achieves information push combined with user information across the entire process on a specified platform, effectively improving the matching degree of the pushed information and thus enhancing the user's decision-making efficiency.

[0110] On the other hand, the information push method disclosed in this embodiment combines various data from within and outside the platform to generate object recommendation point description information, making the object recommendation points richer. Furthermore, from acquiring source data to extracting object recommendation points and generating object recommendation point description information, the entire process requires no manual intervention, resulting in high efficiency and high stability of the information push results.

[0111] After testing, the information push method disclosed in this embodiment, when applied to an e-commerce platform, effectively improves the efficiency of user order placement decisions by integrating data from within and outside the platform to construct an AI-based multi-dimensional selling point pool and business opportunity pool generation mechanism with e-commerce characteristics and category advantages, as well as a full-link personalized intelligent distribution system for selling points and business opportunities. For example, in the product details page scenario, the user's order submission decision efficiency increased by +5.42%.

[0112] Accordingly, as shown in Figure 4, this disclosure also discloses an information push method applied to a client, the method including steps 402, 404 and 406.

[0113] Step 402: In response to the page operation performed by the user on the client, an object operation request is generated based on the user's user information, the target object matched by the page operation, and the object operation scenario.

[0114] Taking the information push method applied to an e-commerce platform as an example, when a user browses the client page of the e-commerce platform, such as when the user clicks on a product link and enters the product details page, the client first obtains the user's user information, and at the same time, obtains the product clicked by the user as the target product, and takes the scenario of entering the details page as the object operation scenario. Then, based on the user's user information, the target object and the object operation scenario, an object operation request is generated according to the preset communication protocol between the client and server of the e-commerce platform.

[0115] Step 404: Send the object operation request to a preset server to trigger the server to perform the following operations: In response to the object operation request, obtain the object operation scenario, target object, and user information that match the object operation request; obtain a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool; select one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information; generate page display data that matches the object operation request based on the target recommendation point information.

[0116] Subsequently, the client sends the generated object operation request to the e-commerce platform's server. The server then performs a page data generation operation corresponding to the object operation request based on the information carried in the request. During the server-side page data generation process, in addition to generating page data according to existing data processing procedures, it also needs to obtain target recommendation point information based on the information carried in the object operation request and insert this information into the page data to generate page data containing the target recommendation point information.

[0117] In response to the object operation request, the server obtains the object operation scenario, target object, and user information that match the object operation request; and retrieves a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool; and selects one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as a specific implementation of the target recommendation point information, as described in the previous embodiments, which will not be repeated here.

[0118] Subsequently, the server generates page display data that matches the object operation request based on the target recommendation point information, and sends the generated page display data to the client.

[0119] Step 406: Based on the page display data, update the page display content of the client to push the target recommendation point information to the user.

[0120] After receiving the page display data sent by the server, the client updates the page display content based on the page display data to display the target recommendation point information generated by the server to the user.

[0121] In summary, the information push method disclosed in this embodiment involves the server responding to an object operation request by obtaining the object operation scenario, target object, and user information matching the object operation request; retrieving a specified number of object recommendation point descriptions matching the target object and the object operation scenario from a preset object recommendation point description information pool; selecting one object recommendation point description matching the user information from the specified number of object recommendation point descriptions as the target recommendation point information; generating page display data based on the target recommendation point information; and then displaying a client page to the user based on the page display data to show the target recommendation point information to the user. This achieves information push combined with user information across the entire process on a specified platform, effectively improving the matching degree of the pushed information and thus enhancing the user's decision-making efficiency.

[0122] On the other hand, as can be seen from the method for obtaining target recommendation points, the information push method disclosed in this embodiment combines various data from within and outside the platform to generate object recommendation point description information, making the object recommendation points richer. Furthermore, from obtaining source data to extracting object recommendation points and generating object recommendation point description information, the entire process requires no manual intervention, resulting in high efficiency and high stability of the information push results.

[0123] Accordingly, this disclosure also discloses a method for generating an object recommendation point description information pool, as shown in Figure 5. The method includes steps 502 to 508.

[0124] Step 502: Obtain source data associated with the target object from the data source corresponding to the preset object recommendation point type, and obtain the source data associated with the target object and corresponding to the preset object recommendation point type.

[0125] Step 504: Extract the description text corresponding to the preset object recommendation point type from the source data corresponding to the preset object recommendation point type.

[0126] Step 506: Using artificial intelligence technology, object recommendation point description information matching the corresponding preset object recommendation point type is generated based on the description text corresponding to the preset object recommendation point type.

[0127] Step 508: Based on the generated object recommendation point description information, create an object recommendation point description information pool for the target object.

[0128] Optionally, the data source corresponding to the preset object recommendation point type includes one or more of the following: e-commerce platforms, social media platforms, market trend analysis platforms, search engines, holiday databases, and recommendation script template libraries.

[0129] For the specific implementation of steps 502 to 508 above, please refer to the relevant descriptions in the previous embodiments, which will not be repeated here.

[0130] The object recommendation point description information pool generation method disclosed in this embodiment combines various data from within and outside the platform and uses artificial intelligence technology to generate object recommendation point description information, making the object recommendation points richer. Furthermore, the entire process, from acquiring source data to extracting object recommendation points and generating object recommendation point description information, requires no manual intervention, resulting in high efficiency and high stability of information push results.

[0131] To implement the above embodiments, this disclosure also discloses an information push system, as shown in FIG6. The information push system includes: a client 602 and a server 604, and a pre-created object recommendation point description information pool 606 corresponding to the target object, wherein...

[0132] The client 602 is used to respond to the page operation performed by the user on the client 602, and generate an object operation request based on the user's user information, the target object matched by the page operation and the object operation scenario;

[0133] The client 602 is also used to send the object operation request to the server 604;

[0134] The server 604 is configured to respond to the object operation request by obtaining the object operation scenario, target object, and user information that match the object operation request; obtaining a specified number of object recommendation point descriptions that match the target object and the object operation scenario from the object recommendation point description information pool 606; selecting one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information; and generating page display data that matches the object operation request based on the target recommendation point information.

[0135] The server 604 is also used to send the page display data to the client 602;

[0136] The client 602 is further configured to update the page display content of the client 602 based on the page display data, so as to push the target recommendation point information to the user.

[0137] For a detailed implementation of the object recommendation point description information pool 606 corresponding to the target object, please refer to the relevant description in the previous embodiments, which will not be repeated here.

[0138] For specific implementation details of the client 602 and server 604, please refer to the relevant descriptions in the previous embodiments, which will not be repeated here.

[0139] In summary, this disclosure also discloses an information push system in which the server responds to an object operation request by obtaining the object operation scenario, target object, and user information that match the object operation request; obtaining a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool; selecting one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information; generating page display data that matches the object operation request based on the target recommendation point information; and updating the client's page display content based on the page display data to push the target recommendation point information to the user. This system achieves information push combined with user information across the entire process on a specified platform, effectively improving the matching degree of the pushed information and thus improving the user's decision-making efficiency.

[0140] On the other hand, the information push system disclosed in this embodiment combines various data from within and outside the platform to generate object recommendation point description information, making the object recommendation points richer. Furthermore, from acquiring source data to extracting object recommendation points and generating object recommendation point description information, the entire process requires no manual intervention, resulting in high efficiency and high stability of the information push results.

[0141] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this disclosure are not limited to the described order of actions, because according to the embodiments of this disclosure, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of this disclosure.

[0142] Based on the above embodiments, this embodiment also provides an information push device applied to a server, the device comprising:

[0143] The object operation request parameter acquisition module is used to respond to an object operation request and acquire the object operation scenario, target object, and user information that match the object operation request.

[0144] The first object recommendation point description information acquisition module is used to acquire a specified number of object recommendation point description information that match the target object and the object operation scenario from a preset object recommendation point description information pool.

[0145] The second object recommendation point description information acquisition module is used to select one object recommendation point description information that matches the user information from the specified number of object recommendation point description information, and use it as the target recommendation point information;

[0146] The information push module is used to push object information to the user based on the target recommendation point information.

[0147] Optionally, the first object recommendation point description information acquisition module is further used for:

[0148] Obtain first preset object information of the target object, and obtain object recommendation point description information from a preset object recommendation point description information pool that matches the target object;

[0149] Artificial intelligence technology is used to filter out a specified number of object recommendation point descriptions that match both the first preset and the object operation scenario from the acquired object recommendation point description information, and these are used as the specified number of object recommendation point descriptions that match the target object and the object operation scenario.

[0150] Optionally, the preset object recommendation point description information pool is created using the following method:

[0151] From the data source corresponding to the preset object recommendation point type, obtain the source data associated with the target object respectively, and obtain the various source data associated with the target object corresponding to the preset object recommendation point type;

[0152] Extract the descriptive text corresponding to the preset object recommendation point type from the source data;

[0153] Using artificial intelligence technology, object recommendation point description information matching the corresponding preset object recommendation point type is generated based on the description text corresponding to the corresponding preset object recommendation point type.

[0154] Based on the generated object recommendation point description information, an object recommendation point description information pool for the target object is created.

[0155] Optionally, obtaining source data associated with the target object from the data source corresponding to the preset object recommendation point type, and obtaining various source data associated with the target object corresponding to the preset object recommendation point type, includes:

[0156] Based on the second preset object information, the objects in the data source corresponding to the preset object recommendation point type are clustered to obtain several object clusters;

[0157] Obtain the source data associated with each object in the object cluster where the target object is located in each data source corresponding to the preset object recommendation point type, and use it as the various source data associated with the target object corresponding to the preset object recommendation point type.

[0158] Optionally, the data source corresponding to the preset object recommendation point type includes one or more of the following: e-commerce platforms, social media platforms, market trend analysis platforms, search engines, holiday databases, and recommendation script template libraries.

[0159] Optionally, the user information includes: user profile information and / or user behavior information, and the step of selecting one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information includes:

[0160] The user's first user characteristics are obtained based on the user profile information, and / or the user's second user characteristics are obtained based on the user behavior information;

[0161] Using artificial intelligence technology, one object recommendation point description that matches the first user feature and / or the second user feature is selected from the specified number of object recommendation point descriptions as the target recommendation point information.

[0162] Optionally, the user information further includes: the language of the user matching; after selecting one object recommendation point description that matches the first user feature and / or the second user feature from the specified number of object recommendation point descriptions using artificial intelligence technology as the target recommendation point information, the method further includes:

[0163] Based on the object operation scenario and the language, the target recommendation point information is optimized to obtain optimized target recommendation point information.

[0164] Optionally, the information push module is further used for:

[0165] Dynamically acquire the display slots that match the target recommendation point information;

[0166] According to the display slot, the preset display information of the target object and the target recommendation point information are assembled to obtain the information to be displayed;

[0167] The information to be displayed is shown to the user in order to push the target recommendation point information to the user.

[0168] Based on the above embodiments, this embodiment also provides an information push device applied to a client, the device comprising:

[0169] The object operation request generation module is used to respond to the page operation performed by the user on the client, and generate an object operation request based on the user's user information, the target object matched by the page operation, and the object operation scenario.

[0170] The page display data acquisition module is used to send the object operation request to a preset server to trigger the server to perform the following operations: in response to the object operation request, acquire the object operation scenario, target object, and user information that match the object operation request; acquire a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool; select one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information; and generate page display data that matches the object operation request based on the target recommendation point information.

[0171] The display module is used to update the page display content of the client based on the page display data, so as to push the target recommendation point information to the user.

[0172] The information push device disclosed in this embodiment is used to implement the above-described information push method. For the specific implementation of each module of the device, please refer to the specific implementation of the corresponding steps in the foregoing method embodiment, which will not be repeated here.

[0173] In summary, the information push device disclosed in this embodiment, by responding to an object operation request from the server, obtains the object operation scenario, target object, and user information matching the object operation request; retrieves a specified number of object recommendation point descriptions matching the target object and the object operation scenario from a preset object recommendation point description information pool; selects one object recommendation point description matching the user information from the specified number of object recommendation point descriptions as the target recommendation point information; and pushes object information to the user based on the target recommendation point information. This achieves information push combined with user information in a full-link scenario on a specified platform, effectively improving the matching degree of the pushed information and thus improving the user's decision-making efficiency.

[0174] Based on the foregoing embodiments, this disclosure also discloses an object recommendation point description information pool generation device, the device comprising:

[0175] The source data acquisition module is used to acquire source data associated with the target object from the data source corresponding to the preset object recommendation point type, and obtain the source data associated with the target object and corresponding to the preset object recommendation point type;

[0176] The recommendation point description text acquisition module is used to extract description text corresponding to the corresponding preset object recommendation point type from the source data corresponding to the preset object recommendation point type;

[0177] The object recommendation point description information generation module is used to generate object recommendation point description information that matches the corresponding preset object recommendation point type based on the description text corresponding to the corresponding preset object recommendation point type, using artificial intelligence technology.

[0178] The object recommendation point description information pool generation module is used to create an object recommendation point description information pool for the target object based on the generated object recommendation point description information.

[0179] The object recommendation point description information pool generation device disclosed in this embodiment is used to implement the above-mentioned information push method. For the specific implementation of each module of the device, please refer to the specific implementation of the corresponding steps in the foregoing method embodiment, which will not be repeated here.

[0180] In summary, the object recommendation point description information pool generation device disclosed in this embodiment combines various data from both within and outside the platform to generate object recommendation point description information, thus enriching the object recommendation points. Furthermore, the entire process, from acquiring source data to extracting object recommendation points and generating object recommendation point description information, requires no manual intervention, resulting in high efficiency and high stability of information push results.

[0181] This disclosure also provides a non-volatile readable storage medium storing one or more modules (programs) that, when applied to a device, enable the device to execute instructions for the method steps in this disclosure.

[0182] This disclosure also provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the methods described in this disclosure.

[0183] This disclosure also provides an electronic device, including: a processor and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method described in this disclosure. In this disclosure, the electronic device includes devices such as servers and terminal devices.

[0184] This disclosure also discloses a computer program product, including a computer program / computer executable instructions, wherein the computer program / computer executable instructions, when executed by a processor in an electronic device, implement the method described in this disclosure.

[0185] The embodiments of this disclosure can be implemented as an apparatus configured using any suitable hardware, firmware, software, or any combination thereof, and may include electronic devices such as servers (clusters) and terminals. Figure 7 schematically illustrates an exemplary apparatus 700 that can be used to implement the various embodiments described in this disclosure.

[0186] For one embodiment, FIG7 illustrates an exemplary device 700 having one or more processors 702, a control module (chipset) 704 coupled to at least one of the processors 702, a memory 706 coupled to the control module 704, a non-volatile memory (NVM) / storage device 708 coupled to the control module 704, one or more input / output devices 710 coupled to the control module 704, and a network interface 712 coupled to the control module 704.

[0187] Processor 702 may include one or more single-core or multi-core processors, and processor 702 may include any combination of general-purpose processors or special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In some embodiments, device 700 can serve as a server, terminal, or other device as described in the embodiments of this disclosure.

[0188] In some embodiments, the apparatus 700 may include one or more computer-readable media (e.g., memory 706 or NVM / storage device 708) having instructions 714 and one or more processors 702 that are combined with the one or more computer-readable media and configured to execute the instructions 714 to implement the module and thus perform the actions described in this disclosure.

[0189] In one embodiment, the control module 704 may include any suitable interface controller to provide any suitable interface to at least one of the processors 702 and / or any suitable device or component communicating with the control module 704.

[0190] The control module 704 may include a memory controller module to provide an interface to the memory 706. The memory controller module may be a hardware module, a software module, and / or a firmware module.

[0191] Memory 706 may be used, for example, to load and store data and / or instructions 714 for device 700. In one embodiment, memory 706 may include any suitable volatile memory, such as suitable DRAM. In some embodiments, memory 706 may include double data rate type 4 synchronous dynamic random access memory (DDR4 SDRAM).

[0192] In one embodiment, the control module 704 may include one or more input / output controllers to provide an interface to the NVM / storage device 708 and (one or more) input / output devices 710.

[0193] For example, NVM / storage device 708 may be used to store data and / or instructions 714. NVM / storage device 708 may include any suitable non-volatile memory (e.g., flash memory) and / or may include any suitable (one or more) non-volatile storage devices (e.g., one or more hard disk drives (HDDs), one or more optical disc drives (CDs), and / or one or more digital universal optical disc (DVD) drives).

[0194] NVM / storage device 708 may include storage resources that are part of a device on which device 700 is mounted, or that are accessible to the device but do not necessarily have to be part of the device. For example, NVM / storage device 708 may be accessed via a network through one or more input / output devices 710.

[0195] One or more input / output devices 710 may provide an interface for device 700 to communicate with any other suitable device. Input / output devices 710 may include communication components, audio components, sensor components, etc. A network interface 712 may provide an interface for device 700 to communicate via one or more networks. Device 700 may wirelessly communicate with one or more components of a wireless network according to any of one or more wireless network standards and / or protocols, such as accessing wireless networks based on communication standards, such as Bluetooth, WiFi, 2G, 3G, 4G, 5G, etc., or combinations thereof.

[0196] In one embodiment, at least one of the processors 702 may be logically packaged with one or more controllers (e.g., memory controller modules) of the control module 704. In one embodiment, at least one of the processors 702 may be logically packaged with one or more controllers of the control module 704 to form a system-in-package (SiP). In one embodiment, at least one of the processors 702 may be integrated with the logic of one or more controllers of the control module 704 on the same die. In one embodiment, at least one of the processors 702 may be integrated with the logic of one or more controllers of the control module 704 on the same die to form a system-on-a-chip (SoC).

[0197] In various embodiments, device 700 may be, but is not limited to, a server, desktop computing device, or mobile computing device (e.g., laptop computing device, handheld computing device, tablet computer, netbook, etc.). In various embodiments, device 700 may have more or fewer components and / or different architectures. For example, in some embodiments, device 700 includes one or more cameras, a keyboard, a liquid crystal display (LCD) screen (including a touchscreen display), a non-volatile memory port, multiple antennas, a graphics chip, an application-specific integrated circuit (ASIC), and a speaker.

[0198] The detection device can use a main control chip as a processor or control module, and sensor data, position information, etc. can be stored in a memory or NVM / storage device. The sensor group can be used as an input / output device, and the communication interface can include a network interface.

[0199] This disclosure also provides an electronic device, including: a processor; and a memory storing executable code thereon, which, when executed, causes the processor to perform one or more methods as described in this disclosure. In this disclosure, the memory can store various types of data, such as target files, file-application association data, and user behavior data, thereby providing a data foundation for various processing methods.

[0200] This disclosure also provides one or more machine-readable media having executable code stored thereon, which, when executed, causes a processor to perform one or more of the methods described in this disclosure.

[0201] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0202] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0203] This disclosure describes embodiments of methods, terminal devices (systems), and computer program products according to embodiments of this disclosure with reference to flowchart illustrations and / or block diagrams. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0204] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0205] These computer program instructions may also be loaded onto a computer or other programmable data processing terminal equipment to cause a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable terminal equipment, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0206] While preferred embodiments of the present disclosure have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the present disclosure.

[0207] Finally, 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 terminal device 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 terminal device. 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 terminal device that includes said element.

[0208] The foregoing has provided a detailed description of an information push method, an object recommendation point description information pool generation method, an information push system, an electronic device, a storage medium, and a computer program product provided by this disclosure. Specific examples have been used to illustrate the principles and implementation methods of this disclosure. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this disclosure. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this disclosure. Therefore, the content of this specification should not be construed as a limitation of this disclosure.

Claims

1. An information push method, applied on a server side, wherein, The method includes: In response to an object operation request, obtain the object operation scenario matching the object operation request, the target object, and the user's user information; Obtain a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool; Select one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions, and use it as the target recommendation point information; Based on the target recommendation point information, object information is pushed to the user.

2. The method according to claim 1, wherein, The step of obtaining a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool includes: Obtain first preset object information of the target object, and obtain object recommendation point description information from a preset object recommendation point description information pool that matches the target object; Artificial intelligence technology is used to filter out a specified number of object recommendation point descriptions that match both the first preset and the object operation scenario from the acquired object recommendation point description information, and these are used as the specified number of object recommendation point descriptions that match the target object and the object operation scenario.

3. The method according to claim 2, wherein, The preset object recommendation point description information pool is created using the following method: From the data source corresponding to the preset object recommendation point type, obtain the source data associated with the target object respectively, and obtain the various source data associated with the target object corresponding to the preset object recommendation point type; Extract the descriptive text corresponding to the preset object recommendation point type from the source data; Using artificial intelligence technology, object recommendation point description information matching the corresponding preset object recommendation point type is generated based on the description text corresponding to the corresponding preset object recommendation point type. Based on the generated object recommendation point description information, an object recommendation point description information pool for the target object is created.

4. The method according to claim 3, wherein, The step of obtaining source data associated with the target object from the data source corresponding to the preset object recommendation point type, and obtaining various source data associated with the target object corresponding to the preset object recommendation point type, includes: Based on the second preset object information, the objects in the data source corresponding to the preset object recommendation point type are clustered to obtain several object clusters; Obtain the source data associated with each object in the object cluster where the target object is located in each data source corresponding to the preset object recommendation point type, and use it as the various source data associated with the target object corresponding to the preset object recommendation point type.

5. The method according to claim 3, wherein, The data sources corresponding to the preset object recommendation point types include one or more of the following: e-commerce platforms, social media platforms, market trend analysis platforms, search engines, holiday databases, and recommendation script template libraries.

6. The method according to claim 1, wherein, The user information includes: user profile information and / or user behavior information. The step of selecting one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information includes: The user's first user characteristics are obtained based on the user profile information, and / or the user's second user characteristics are obtained based on the user behavior information; Using artificial intelligence technology, one object recommendation point description that matches the first user feature and / or the second user feature is selected from the specified number of object recommendation point descriptions as the target recommendation point information.

7. The method according to claim 6, wherein, The user information further includes: the language of the user matching; after selecting one object recommendation point description that matches the first user feature and / or the second user feature from the specified number of object recommendation point descriptions using artificial intelligence technology as the target recommendation point information, it further includes: Based on the object operation scenario and the language, the target recommendation point information is optimized to obtain optimized target recommendation point information.

8. The method according to claim 1, wherein, The step of pushing object information to the user based on the target recommendation point information includes: Dynamically acquire the display slots that match the target recommendation point information; According to the display slot, the preset display information of the target object and the target recommendation point information are assembled to obtain the information to be displayed; The information to be displayed is shown to the user in order to push the target recommendation point information to the user.

9. An information push method, applied to a client, wherein, The term includes: In response to a page operation performed by a user on the client, an object operation request is generated based on the user's user information, the target object matched by the page operation, and the object operation scenario. The system sends the object operation request to a preset server to trigger the server to perform the following operations: In response to the object operation request, the system obtains the object operation scenario, target object, and user information that match the object operation request; it obtains a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool; it selects one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information; and it generates page display data that matches the object operation request based on the target recommendation point information. Based on the page display data, the page display content of the client is updated to push the target recommendation point information to the user.

10. A method for generating an object recommendation point description information pool, wherein, The method includes: From the data source corresponding to the preset object recommendation point type, obtain the source data associated with the target object, and obtain the source data associated with the target object and corresponding to the preset object recommendation point type; Extract the descriptive text corresponding to the preset object recommendation point type from the source data; Using artificial intelligence technology, object recommendation point description information matching the corresponding preset object recommendation point type is generated based on the description text corresponding to the corresponding preset object recommendation point type. Based on the generated object recommendation point description information, an object recommendation point description information pool for the target object is created.

11. An information push system, wherein, The system includes: a client, a server, and a pre-created pool of object recommendation point description information corresponding to the target object, wherein... The client is configured to respond to a page operation performed by a user on the client, and generate an object operation request based on the user's user information, the target object matched by the page operation, and the object operation scenario. The client is also used to send the object operation request to the server; The server is configured to respond to the object operation request by obtaining the object operation scenario, target object, and user information that match the object operation request; obtaining a specified number of object recommendation point descriptions that match the target object and the object operation scenario from the object recommendation point description information pool; selecting one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information; and generating page display data that matches the object operation request based on the target recommendation point information. The server is also used to send the page display data to the client; The client is also configured to update the page display content of the client based on the page display data, so as to push the target recommendation point information to the user.

12. An information push device, applied to a server, wherein, The device The object operation request parameter acquisition module is used to respond to an object operation request and acquire the object operation scenario, target object, and user information that match the object operation request. The first object recommendation point description information acquisition module is used to acquire a specified number of object recommendation point description information that match the target object and the object operation scenario from a preset object recommendation point description information pool. The second object recommendation point description information acquisition module is used to select one object recommendation point description information that matches the user information from the specified number of object recommendation point description information, and use it as the target recommendation point information; The information push module is used to push object information to the user based on the target recommendation point information.

13. The apparatus according to claim 12, wherein, The step of obtaining a specified number of object recommendation point descriptions that match the target object and the object operation scenario from a preset object recommendation point description information pool includes: Obtain first preset object information of the target object, and obtain object recommendation point description information from a preset object recommendation point description information pool that matches the target object; Artificial intelligence technology is used to filter out a specified number of object recommendation point descriptions that match both the first preset and the object operation scenario from the acquired object recommendation point description information, and these are used as the specified number of object recommendation point descriptions that match the target object and the object operation scenario.

14. The apparatus according to claim 13, wherein, The preset object recommendation point description information pool is created using the following method: From the data source corresponding to the preset object recommendation point type, obtain the source data associated with the target object respectively, and obtain the various source data associated with the target object corresponding to the preset object recommendation point type; Extract the descriptive text corresponding to the preset object recommendation point type from the source data; Using artificial intelligence technology, object recommendation point description information matching the corresponding preset object recommendation point type is generated based on the description text corresponding to the corresponding preset object recommendation point type. Based on the generated object recommendation point description information, an object recommendation point description information pool for the target object is created.

15. The apparatus according to claim 14, wherein, The step of obtaining source data associated with the target object from the data source corresponding to the preset object recommendation point type, and obtaining various source data associated with the target object corresponding to the preset object recommendation point type, includes: Based on the second preset object information, the objects in the data source corresponding to the preset object recommendation point type are clustered to obtain several object clusters; Obtain the source data associated with each object in the object cluster where the target object is located in each data source corresponding to the preset object recommendation point type, and use it as the various source data associated with the target object corresponding to the preset object recommendation point type.

16. The apparatus according to claim 14, wherein, The data sources corresponding to the preset object recommendation point types include one or more of the following: e-commerce platforms, social media platforms, market trend analysis platforms, search engines, holiday databases, and recommendation script template libraries.

17. The apparatus according to claim 12, wherein, The user information includes: user profile information and / or user behavior information. The step of selecting one object recommendation point description that matches the user information from the specified number of object recommendation point descriptions as the target recommendation point information includes: The user's first user characteristics are obtained based on the user profile information, and / or the user's second user characteristics are obtained based on the user behavior information; Using artificial intelligence technology, one object recommendation point description that matches the first user feature and / or the second user feature is selected from the specified number of object recommendation point descriptions as the target recommendation point information.

18. An electronic device, wherein, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-10.

19. A computer-readable storage medium, wherein, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-10.

20. A computer program product comprising a computer program / computer-executable instructions, wherein, When the computer program / computer executable instructions are executed by a processor in an electronic device, the method of any one of claims 1-10 is implemented.

Images