Promotion information processing method and device, and storage medium

Abstract

The application provides a processing method and device of promotion information, an electronic device and a storage medium. The method comprises the following steps: determining a target person from at least one person described in the content of an article; analyzing the content corresponding to the target person in the content of the article to obtain the character features of the target person; determining a target user to be promoted and a set of objects to be promoted which are suitable for the target user; selecting an object to be promoted related to the content of the target page corresponding to the article from the set of objects to be promoted; generating corresponding promotion information based on the selected object to be promoted and the character features of the target person; and fusing the promotion information into the content of the target page. The promotion information is used to present the promotion information in the process of presenting the target page by the terminal corresponding to the target user. Through the application, the user's favorability and click rate for the promotion information can be improved, and the promotion effect is improved.

Description

Technical Field

[0001] This application relates to the fields of artificial intelligence and internet technology, and in particular to a method, apparatus, electronic device and storage medium for processing promotional information. Background Technology

[0002] Artificial intelligence (AI) is the theory, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.

[0003] Recommendation systems are an important application branch of artificial intelligence. In related technologies, the recommendation method for advertisements in article scenarios only makes recommendations based on the user, and the content of the advertisements is pre-edited and fixed, resulting in poor recommendation performance. Summary of the Invention

[0004] This application provides a method, apparatus, electronic device, and storage medium for processing promotional information, which can improve user favorability and click-through rate of promotional information, thereby enhancing the promotional effect.

[0005] The technical solution of this application embodiment is implemented as follows:

[0006] This application provides a method for processing promotional information, including:

[0007] Identify the target character from at least one character described in the article;

[0008] The content of the article corresponding to the target person is analyzed to obtain the characteristics of the target person;

[0009] Identify the target users for the promotion of the target objects, and the set of target objects that are compatible with the target users;

[0010] From the set of objects to be promoted, select objects that are related to the content of the target page corresponding to the article;

[0011] Based on the selected target audience and their characteristics, corresponding promotional information is generated.

[0012] The promotional information is integrated into the content of the target page; wherein the promotional information is presented during the process of the target user's terminal displaying the target page.

[0013] This application provides a method for processing promotional information, including:

[0014] Receives a request from the target user to view content on the target page corresponding to the article;

[0015] In response to the viewing operation, the target page is displayed, and

[0016] When there is a target page content-related target page in the set of target users that is compatible with the target user, the promotion information corresponding to the target page is presented during the presentation of the target page.

[0017] The promotional information is generated based on the target person's personality as described in the article and the content of the article.

[0018] This application embodiment also provides a processing device for promotional information, including:

[0019] The first determination module is used to determine the target person from at least one person described in the content of the article;

[0020] The analysis module is used to analyze the content of the article that corresponds to the target person, and to obtain the characteristics of the target person.

[0021] The second determining module is used to determine the target users of the objects to be promoted, and the set of objects to be promoted that are compatible with the target users;

[0022] The selection module is used to select, from the set of objects to be promoted, objects that are related to the content of the target page corresponding to the article;

[0023] The generation module is used to generate corresponding promotional information based on the selected target and the characteristics of the target person.

[0024] The fusion module is used to fuse the promotional information into the content of the target page; wherein the promotional information is presented during the process of the target user's terminal displaying the target page.

[0025] In the above scheme, the first determining module is further used to perform semantic analysis on the content of the article to determine the target content corresponding to each of the characters;

[0026] The amount of information in the target content corresponding to each of the aforementioned individuals is statistically analyzed.

[0027] Based on the amount of information in the target content corresponding to each of the aforementioned individuals, the individuals corresponding to the target content whose information content meets the criteria for target individuals are identified as target individuals.

[0028] In the above scheme, the first determining module is further used to perform word segmentation on the content of the article to obtain multiple words corresponding to the article;

[0029] Identify the target segmentation word belonging to the noun category from the multiple segmentation words;

[0030] Count the number of words segmented for the target words corresponding to each of the aforementioned characters;

[0031] Based on the number of target words for each of the aforementioned figures, the figures corresponding to the target words whose number of words meets the criteria for target figures are identified as the target figures.

[0032] In the above scheme, the second determining module is also used to determine multiple candidate users;

[0033] Obtain the user profile of each of the candidate users;

[0034] Based on the user profile, the multiple candidate users are classified to obtain at least two user sets, and each user set is used as the target user.

[0035] In the above scheme, the device further includes:

[0036] The receiving module is used to receive viewing operations for the content of the target page corresponding to the article;

[0037] Determine the target user set to which the user corresponding to the viewing operation belongs;

[0038] Determine the target promotional information that corresponds to the target user set and the content of the target page, and

[0039] The content of the target page, which incorporates the target promotional information, is returned to the terminal of the user corresponding to the viewing operation.

[0040] In the above scheme, the analysis module is also used to extract content corresponding to the target person from the content of the article;

[0041] The content corresponding to the target person is encoded to obtain the person vector corresponding to the target person;

[0042] The person's vector is input into a machine learning model, and the machine learning model predicts the person's features from the person's vector to obtain the person's features.

[0043] In the above scheme, the analysis module is also used to extract content corresponding to the target person from the content of the article;

[0044] From at least two feature dimensions, feature extraction is performed on the content corresponding to the target person to obtain the intermediate person features corresponding to each feature dimension of the target person;

[0045] Based on the intermediate features of the target person corresponding to each of the feature dimensions, the character features of the target person are constructed.

[0046] In the above scheme, the second determining module is also used to obtain multiple objects to be promoted and the profile information corresponding to the target user;

[0047] From the plurality of objects to be promoted, at least one object to be promoted that matches the profile information is selected;

[0048] The set of objects to be promoted is constructed based on the at least one object to be promoted.

[0049] In the above scheme, the selection module is further used to perform semantic analysis on the content of the target page to obtain the page content of the target page;

[0050] Semantic recognition is performed on the text materials of each object to be promoted in the set of objects to be promoted to obtain the text material content of each object to be promoted.

[0051] The page content is matched with the text material content of each of the objects to be promoted to obtain the matching degree between the content of the target page and each of the objects to be promoted;

[0052] From the set of objects to be promoted, select objects whose matching degree reaches the matching degree threshold as objects to be promoted that are related to the content of the target page.

[0053] In the above scheme, the selection module is further used to perform word segmentation on the content of the target page to obtain multiple words corresponding to the target page;

[0054] Named entity recognition is performed on the multiple word segments to obtain the target objects contained in the content of the target page;

[0055] The target object is matched with each object to be promoted in the set of objects to be promoted to obtain the matching results;

[0056] The objects to be promoted that match the target object in the matching results are taken as objects to be promoted that are related to the content of the target page.

[0057] In the above scheme, the generation module is further used to obtain the basic information of the object to be promoted, and

[0058] Information that is compatible with the target user is determined from the basic information and used as personalized information for the target user;

[0059] Obtain the language template corresponding to the characteristics of the target person;

[0060] Based on the language template and the personalized information, corresponding promotional information is generated.

[0061] In the above solution, the fusion module is further used to obtain the fusion template corresponding to the target page. The fusion template is used to describe the layout position of the promotional information and the content of the target page in the target page after fusion.

[0062] The promotional information and the content of the target page are added to the corresponding layout positions on the target page to integrate the promotional information into the content of the target page.

[0063] In the above scheme, the device further includes:

[0064] The promotional user identification module is used to obtain the click data of the target user for each of the promotional messages when there are at least two promotional messages and each promotional message is generated based on the characteristics of different target users;

[0065] Based on the click data, the target person corresponding to at least one promotional message whose click data meets the promotion conditions is identified as the first promotional person.

[0066] Accordingly, the device further includes:

[0067] The setting module is used to set the priority of the promotion information corresponding to the first promoter to be higher than the priority of the promotion information of the second promoter; wherein, the second promoter is a target person other than the first promoter among the different target persons;

[0068] When the content of the new page includes the first promoter and the second promoter, after selecting the target to be promoted that is related to the content of the new page, the corresponding new promotion information is generated based on the characteristics of the selected target to be promoted that is related to the content of the new page and the first promoter.

[0069] The new promotional information is integrated into the content of the new page.

[0070] In the above scheme, the device further includes:

[0071] The update module is used to obtain the click data of the target user for each of the promotional messages when there are at least two promotional messages and each promotional message is generated based on the characteristics of different target users;

[0072] Based on the click data, the target person corresponding to at least one promotional message whose click data meets the promotion conditions is identified as the target promotional person.

[0073] Obtain the comments from the target user regarding the article;

[0074] Based on the comments, the language templates corresponding to each of the target promotional figures are updated. The language templates are used to integrate the characteristics of the selected target and the target figures to generate the promotional information.

[0075] This application embodiment also provides a processing device for promotional information, including:

[0076] The receiving module is used to receive viewing operations triggered by the target user for the content of the target page corresponding to the article;

[0077] The presentation module is used to respond to the viewing operation, present the target page, and when there is a target page related to the content of the target page in the set of target pages that are compatible with the target user, the promotion information corresponding to the target page is presented during the presentation of the target page.

[0078] The promotional information is generated based on the target person's personality as described in the article and the content of the article.

[0079] This application also provides an electronic device, including:

[0080] Memory, used to store executable instructions;

[0081] The processor, when executing executable instructions stored in the memory, implements the promotional information processing method provided in the embodiments of this application.

[0082] This application also provides a computer-readable storage medium storing executable instructions, which, when executed by a processor, implement the promotional information processing method provided in this application.

[0083] The embodiments of this application have the following beneficial effects:

[0084] This application analyzes the content of the target person in the article to obtain the characteristics of the target person, selects the target objects related to the content of the target page from the set of target objects that are suitable for the target users to be promoted, and then generates corresponding promotional information based on the characteristics of the target person and the target objects to be promoted, and integrates the promotional information into the content of the target page, so that the target user's terminal displays the promotional information during the process of displaying the target page.

[0085] Here, promotional information is generated based on the characteristics of the characters in the article and the target audience related to the content of the target page. This improves the matching degree between promotional information and page content, as well as their integration. At the same time, because the selected target audience is compatible with the target users, the promotional information is no longer static, realizing personalized generation of promotional information for users. This increases the attractiveness of the promotional information to users, as well as users' goodwill and click-through rate, thereby improving the promotional effect. Attached Figure Description

[0086] Figure 1 This is a schematic diagram of the architecture of the promotional information processing system 100 provided in the embodiments of this application;

[0087] Figure 2 This is a schematic diagram of the structure of an electronic device 500 for processing promotional information according to an embodiment of this application;

[0088] Figure 3 This is a flowchart illustrating the promotional information processing method provided in the embodiments of this application;

[0089] Figure 4 This is a schematic diagram illustrating the process of integrating promotional information to the target page, as provided in the embodiments of this application.

[0090] Figure 5 This is a flowchart illustrating the promotional information processing method provided in the embodiments of this application;

[0091] Figure 6 This is a schematic diagram illustrating the promotional information provided in the embodiments of this application;

[0092] Figure 7A This is a flowchart illustrating the promotional information processing method provided in the embodiments of this application;

[0093] Figure 7B This is a flowchart illustrating the promotional information processing method provided in the embodiments of this application;

[0094] Figure 8 This is a schematic diagram illustrating the presentation of promotional information in videos provided by related technologies;

[0095] Figure 9 This is a flowchart illustrating the promotional information processing method provided in the embodiments of this application;

[0096] Figure 10 This is a schematic diagram illustrating the relationships between multiple individuals described in the article provided in this application embodiment;

[0097] Figure 11 This is a schematic diagram of the structure of the promotional information processing device 555 provided in the embodiments of this application;

[0098] Figure 12 This is a schematic diagram of the structure of the promotional information processing device 600 provided in the embodiments of this application. Detailed Implementation

[0099] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0100] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0101] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0102] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0103] In the implementation of this application, the collection and processing of relevant data should strictly comply with the requirements of relevant laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0104] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0105] 1) Client: An application that runs on a terminal and provides various services, such as an instant messaging client or a video playback client.

[0106] 2) In response, used to indicate the conditions or states on which the operation performed depends. When the conditions or states on which it depends are met, one or more operations performed may be performed in real time or with a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations are performed.

[0107] 3) Product placement refers to an advertising method that integrates products or services into a work (including but not limited to dialogue, close-up shots of props, character images, scene provision, etc.).

[0108] 4) Creative in-video ads refer to a "sandwich-style advertisement" that uses the IP in the video to convey the concept or product characteristics of a specific brand during the video playback.

[0109] 5) Auction advertising refers to a new type of online advertising where users independently place and manage their ads, adjust prices to rank them, and pay according to the ad's performance.

[0110] 6) Doc2Vec, the paragraph vector method, is an unsupervised algorithm that learns fixed-length feature representations from variable-length text (e.g., sentences, paragraphs, and documents). The algorithm is trained to predict words within a document, enabling it to represent each document using a single dense vector.

[0111] 7) Random forest model is a classifier that contains multiple decision trees, and the class of its output is determined by the mode of the classes output by the individual trees.

[0112] 8) Semantic analysis refers to a method for transforming unstructured data into structured data, including Chinese word segmentation / keyword extraction. For a given sentence, it involves word segmentation, part-of-speech tagging, named entity recognition and linking, syntactic analysis, semantic role recognition, and polysemous word disambiguation.

[0113] 9) Named entity recognition, also known as proper name recognition, refers to the identification of entities with specific meanings in text, mainly including personal names, place names, organization names, proper nouns, etc. Simply put, it is the identification of the boundaries and categories of entity references in natural text.

[0114] 10) The Enneagram, also known as the Nine Personality Types, is a personality typology system. It includes activity level, regularity, range of interests, intensity of reactions, psychological qualities, distractibility, and range / duration of focus. The main types include: Perfectionist, Helper, Achiever, Individualist, Logical, Loyalist, Activist, Leader, and Peacemaker.

[0115] Based on the above explanation of the nouns and terms used in the embodiments of this application, the following describes the promotional information processing system provided in the embodiments of this application. See also Figure 1 , Figure 1 This is a schematic diagram of the architecture of the promotional information processing system 100 provided in the embodiments of this application. In order to support an exemplary application, the terminal (terminal 400-1 and terminal 400-2 are shown as examples) connects to the server 200 through the network 300. The network 300 can be a wide area network or a local area network, or a combination of the two, and data transmission is achieved using wireless or wired links.

[0116] Terminals (such as terminal 400-1 and terminal 400-2) are used to receive the viewing operation triggered by the target user for the content of the target page corresponding to the article, and in response to the viewing operation, send a request to obtain the content of the target page to server 200.

[0117] Server 200 is used to respond to a request to retrieve the content of a target page, identify a target person from at least one person described in the article content; analyze the content in the article that corresponds to the target person to obtain the target person's characteristics; determine the target users for the promotion of the target person, and a set of target objects that are suitable for the target users; select target objects related to the content of the target page corresponding to the article from the set of target objects; generate corresponding promotional information based on the selected target objects and the characteristics of the target person; integrate the promotional information into the content of the target page; and send the content of the target page integrated with the promotional information to the terminal.

[0118] Terminals (such as terminals 400-1 and 400-2) are used to present the target page on the graphical interface 410 (graphical interfaces 410-1 and 410-2 are shown as examples), and in the process of presenting the target page, to present the promotional information corresponding to the object to be promoted.

[0119] In practical applications, server 200 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. Terminals (such as terminal 400-1 and terminal 400-2) can be smartphones, tablets, laptops, desktop computers, smart speakers, smart TVs, smartwatches, etc., but are not limited to these. Terminals (such as terminal 400-1 and terminal 400-2) and server 200 can be directly or indirectly connected via wired or wireless communication, which is not limited herein.

[0120] See Figure 2 , Figure 2 This is a schematic diagram of the structure of an electronic device 500 for processing promotional information according to an embodiment of this application. In practical applications, the electronic device 500 can be... Figure 1 The server or terminal shown is exemplified by electronic device 500. Figure 1 Taking the terminal shown as an example, an electronic device implementing the promotional information processing method of this application embodiment will be described. The electronic device 500 provided in this application embodiment includes: at least one processor 510, a memory 550, at least one network interface 520, and a user interface 530. The various components in the electronic device 500 are coupled together through a bus system 540. It is understood that the bus system 540 is used to realize the connection and communication between these components. In addition to a data bus, the bus system 540 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 2 The general labeled all buses as Bus System 540.

[0121] The processor 510 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0122] User interface 530 includes one or more output devices 531 that enable the presentation of media content, including one or more speakers and / or one or more visual displays. User interface 530 also includes one or more input devices 532, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

[0123] The memory 550 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state storage, hard disk drives, optical disk drives, etc. The memory 550 may optionally include one or more storage devices physically located away from the processor 510.

[0124] The memory 550 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), and the volatile memory may be random access memory (RAM). The memory 550 described in this application embodiment is intended to include any suitable type of memory.

[0125] In some embodiments, memory 550 is capable of storing data to support various operations, examples of which include programs, modules, and data structures or subsets or supersets thereof, as illustrated below.

[0126] Operating system 551 includes system programs for handling various basic system services and performing hardware-related tasks, such as the framework layer, core library layer, driver layer, etc., for implementing various basic business functions and handling hardware-based tasks;

[0127] The network communication module 552 is used to reach other computing devices via one or more (wired or wireless) network interfaces 520, exemplary network interfaces 520 including: Bluetooth, WiFi, and Universal Serial Bus (USB), etc.

[0128] Presentation module 553 is used to enable the presentation of information (e.g., user interface for operating peripheral devices and displaying content and information) via one or more output devices 531 (e.g., display screen, speaker, etc.) associated with user interface 530.

[0129] The input processing module 554 is used to detect and translate one or more user inputs or interactions from one or more input devices 532.

[0130] In some embodiments, the promotional information processing apparatus provided in this application can be implemented in software. Figure 2 A processing device 555 for promotional information stored in memory 550 is shown. It may be software in the form of programs and plug-ins, including the following software modules: a first determination module 5551, an analysis module 5552, a second determination module 5553, a selection module 5554, a generation module 5555, and a fusion module 5556. These modules are logically related and can therefore be arbitrarily combined or further divided according to the functions they implement. The functions of each module will be described below.

[0131] In other embodiments, the promotional information processing apparatus provided in this application can be implemented using a combination of hardware and software. As an example, the promotional information processing apparatus provided in this application can be a processor in the form of a hardware decoding processor, which is programmed to execute the promotional information processing method provided in this application. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0132] Based on the above description of the promotional information processing system and electronic device provided in the embodiments of this application, the method for processing promotional information provided in the embodiments of this application is described below. In some embodiments, the method for processing promotional information provided in the embodiments of this application may be implemented by a server or a terminal alone, or by a server and a terminal working together. The following description uses a terminal implementation as an example to illustrate the method for processing promotional information provided in the embodiments of this application. See also... Figure 3 , Figure 3 This is a flowchart illustrating the promotional information processing method provided in this application embodiment. The promotional information processing method provided in this application embodiment includes:

[0133] Step 101: The terminal identifies the target person from at least one person described in the article's content.

[0134] In practical applications, terminals are equipped with client applications, such as reading clients or browser clients with reading plugins. Users can run the client on the terminal to view or browse articles. When the terminal receives a user's trigger to view an article, it presents the article content to the user through a view interface, which could be the content of a certain chapter or a certain page of the article.

[0135] In this embodiment of the application, the terminal analyzes the content of the article and determines the target character from at least one character described in the content of the article. The target character can be one or more. In practical applications, the target character is at least one protagonist described in the content of the article.

[0136] In some embodiments, the terminal may determine the target person from at least one person described in the content of the article by: performing semantic analysis on the content of the article to determine the target content corresponding to each person; calculating the information content corresponding to each person; and determining the person corresponding to the target content whose information content meets the target person condition based on the information content corresponding to each person.

[0137] Here, the terminal can acquire the content of an article, perform semantic analysis on it, and obtain the target content corresponding to each character described in the article. Then, for each character, the information content corresponding to the target content is statistically analyzed. This information content information content can be based on the number of chapters describing each character in the article, or the amount of text describing each character in the article, etc. After statistically analyzing the information content corresponding to each character, the characters whose target content information content matches the target character criteria are selected as the target characters.

[0138] In practical applications, the target person criterion can be the person corresponding to the target content with the most information. In this case, based on the statistical analysis of the amount of information in the target content corresponding to each person, the person corresponding to the target content with the most information is selected as the target person.

[0139] In practical applications, the target person criterion can also be defined as the person corresponding to target content whose information content exceeds an information content threshold. In this case, based on the statistical analysis of the information content corresponding to each person, target content whose information content exceeds the information content threshold is selected, and the person corresponding to the target content whose information content exceeds the information content threshold is designated as the target person.

[0140] In practical applications, the target person criterion can also be determined by sorting the target content in descending order of information content, and then selecting the person corresponding to the top-ranked target content. In this case, based on the information content of the target content corresponding to each person, the target content is sorted in descending order of information content, and the person corresponding to the top-ranked target content is selected as the target person.

[0141] In some embodiments, the terminal may also determine the target person from at least one person described in the content of the article by: performing word segmentation on the content of the article to obtain multiple word segments corresponding to the article; identifying the target word belonging to the noun category of person from the multiple word segments; counting the number of word segments corresponding to each person's target word; and determining the person corresponding to the target word whose word segmentation number meets the target person condition as the target person based on the number of word segments of each person's target word.

[0142] Here, the terminal can obtain the content of the article, then perform word segmentation on the content to obtain multiple word segments corresponding to the article; then perform part-of-speech recognition on each word segment to identify the target word belonging to the noun category of a person from the multiple word segments; count the number of target word segments corresponding to each person, and then, based on the number of target word segments of each person, select the person corresponding to the target word whose word segmentation number meets the target person condition as the target person.

[0143] In practical applications, the target person criterion can be the person corresponding to the target segment with the highest number of segmentations. In this case, based on the number of segmentations of each person's target segment, the person corresponding to the target segment with the highest number of segmentations is selected as the target person.

[0144] In practical applications, the target person condition can also be defined as the person corresponding to the target segment whose word count exceeds a certain threshold. In this case, based on the count of the target segment corresponding to each person, the target segment whose word count exceeds the threshold is selected, and the person corresponding to the target segment whose word count exceeds the threshold is selected as the target person.

[0145] In practical applications, the target person condition can also be determined by sorting the target words in descending order of word count, and then selecting the person corresponding to the top-ranked target word count. In this case, the target words are sorted in descending order of word count based on the number of word segments corresponding to each person, and the person corresponding to the top-ranked target word count is selected as the target person.

[0146] Step 102: Analyze the content of the article that corresponds to the target person to obtain the target person's characteristics.

[0147] After identifying the target person described in the article, extract the content within the article that corresponds to the target person. Analyze this content to determine the target person's characteristics, which may include their personality traits, behavioral characteristics, language characteristics, life experiences, and relationships with other people described in the article.

[0148] In some embodiments, the terminal can analyze the content of the article that corresponds to the target person in the following ways to obtain the target person's characteristics: extract the content corresponding to the target person from the article content; encode the content corresponding to the target person to obtain the target person's vector; input the person vector into a machine learning model, and use the machine learning model to predict the target person's characteristics to obtain the target person's characteristics.

[0149] Here, when analyzing the content corresponding to the target person, the process begins by extracting the content corresponding to the target person from the article. Then, the content corresponding to the target person is encoded to obtain the target person's vector. Specifically, the content corresponding to the target person can be input into the doc2Vec model, which generates the target person's vector. The target person's vector is then input into a pre-trained machine learning model, such as a random forest model built based on multiple decision trees. The machine learning model then predicts the target person's features, such as personality traits, from the target person's vector, thereby obtaining the target person's characteristics.

[0150] In some embodiments, the terminal may also analyze the content in the article that corresponds to the target person in the following ways to obtain the target person’s characteristics: extract the content that corresponds to the target person from the article content; extract features from the content that corresponds to the target person from at least two feature dimensions to obtain the intermediate character features of the target person in each feature dimension; and construct the character features of the target person based on the intermediate character features of the target person in each feature dimension.

[0151] Here, when analyzing the content corresponding to the target person, the first step is to extract the content corresponding to the target person from the article. Then, feature extraction is performed on the content corresponding to the target person. Specifically, feature extraction can be performed from at least two pre-set feature dimensions, such as time dimension, relationship dimension, behavioral event dimension, and personality dimension, to obtain the intermediate person features corresponding to each feature dimension of the target person. Then, the intermediate person features corresponding to each feature dimension of the target person are fused or spliced ​​according to the weights corresponding to each feature dimension to construct the person features of the target person.

[0152] Step 103: Determine the target users for the promotion of the target objects, and the set of target objects that are suitable for the target users.

[0153] Here, the target users for the promotion can be either current readers of the article or registered users of the corresponding reading client.

[0154] After identifying the target person and their characteristics associated with the article, and determining the target user, the terminal retrieves the content of the target page corresponding to the article—that is, the content of the target page that the target user is about to read (and that the terminal is about to display). Simultaneously, it also retrieves a set of potential promotional targets that are compatible with the target user.

[0155] In some embodiments, the terminal may determine a set of objects to be promoted that are compatible with the target user by: obtaining multiple objects to be promoted and profile information corresponding to the target user; selecting at least one object to be promoted that matches the profile information from the multiple objects to be promoted; and constructing a set of objects to be promoted based on at least one object to be promoted.

[0156] In practical applications, this set of objects to be promoted can be constructed by selecting at least one object from a large pool of objects that matches the target user's profile information. This large pool of objects can be pre-defined, such as cars, cosmetics, instant noodles, and plush toys. Based on the target user's profile information, at least one object matching the profile can be selected from this pool. For example, if the target user is female, then cosmetics and plush toys could both be objects matching the profile. After selecting at least one object matching the profile, a set of objects suitable for the target user is constructed based on this at least one object.

[0157] Step 104: Select objects from the set of objects to be promoted that are related to the content of the target page corresponding to the article.

[0158] After obtaining a set of promotional objects that match the target users, select promotional objects that are related to the content of the target page from the set of promotional objects. For example, if the content of the target page contains a description of instant noodles, then instant noodles can be selected from the set of promotional objects as a promotional object that is related to the content of the target page.

[0159] In some embodiments, the terminal may select objects to be promoted that are related to the content of the target page in the following manner: performing semantic analysis on the content of the target page to obtain the page content of the target page; performing semantic recognition on the text material of each object to be promoted in the set of objects to be promoted to obtain the text material content of each object to be promoted; matching the page content with the text material content of each object to be promoted to obtain the matching degree between the content of the target page and each object to be promoted; and selecting objects to be promoted that have a matching degree that reaches a matching degree threshold from the set of objects to be promoted as objects to be promoted that are related to the content of the target page.

[0160] Here, the terminal first performs semantic analysis on the content of the target page to obtain the page content. In practical applications, each target object in the set of targets to be promoted can have corresponding text materials to describe it, such as promotion points, characteristics, and basic information. Then, semantic recognition can be performed on the text materials of each target object to obtain its text material content. Next, the page content of the target page is matched with the text material content of each target object to determine the matching degree between the target page content and each target object. Finally, from the set of targets to be promoted, those with a matching degree reaching a matching threshold are selected as the targets related to the content of the target page.

[0161] In some embodiments, the terminal may also select objects to be promoted that are related to the content of the target page in the following manner: perform word segmentation on the content of the target page to obtain multiple words corresponding to the target page; perform named entity recognition on the multiple words to obtain the target objects contained in the content of the target page; match the target objects with each object to be promoted in the set of objects to be promoted to obtain matching results; and take the objects to be promoted that match the target objects in the matching results as objects to be promoted that are related to the content of the target page.

[0162] Here, the terminal can also perform word segmentation on the content of the target page to obtain multiple words corresponding to the target page; then, named entity recognition is performed on each word, specifically using a named entity recognition algorithm, to identify entities (such as place names / brands / products) contained in the content of the target page, and the identified entities are taken as the target objects corresponding to the target page; then, the target objects are matched with each object to be promoted in the set of objects to be promoted to obtain matching results; the objects to be promoted that match the target objects in the matching results are taken as objects to be promoted that are related to the content of the target page.

[0163] Step 105: Generate corresponding promotional information based on the characteristics of the selected target audience and individuals.

[0164] After determining the target audience related to the content of the target page and the characteristics of the target person in the article, the terminal generates promotional information corresponding to the target audience based on the characteristics of the target audience and the target person. This promotional information is then displayed on the target page, allowing users to view the promotional information of the target audience related to the content of the target page while viewing the content of the target page.

[0165] In some embodiments, the terminal can generate corresponding promotional information based on the selected target and the characteristics of the target person in the following ways: obtaining the basic information of the target and determining the information that matches the target user from the basic information as the personalized information of the target user; obtaining the language template corresponding to the characteristics of the target person; and generating corresponding promotional information based on the language template and the personalized information.

[0166] In practical applications, the target audience can be provided with corresponding text materials to describe its basic information, such as detailed parameters, promotional points, and user-attracting features. When generating promotional information for the target audience, the basic information can be obtained first. To further enhance the appeal of the promotional information, the terminal can extract more user-specific information from the basic information, serving as personalized information for the target user. Then, the language template corresponding to the target person's characteristics is obtained. Finally, based on this language template and personalized information, the promotional information for the target audience is generated. Here, the language template corresponding to the target person's characteristics can be pre-set or learned from a trained machine learning model, and can include semantic rules, commonly used characters, words, and sentence structures. For example, the target person's language template might include "ya," "wa," "delicious," and "want to eat," with exclamatory sentences being common. If the personalized information for the target user is "crispy noodles, fragrant and crispy," then the promotional information generated based on the target person's language template and personalized information would be "Wow, these crispy noodles are fragrant, crispy, and delicious! I want to eat more!"

[0167] In practical applications, the target person can be one or multiple. When there is only one target person, promotional information is generated based on the selected target and the characteristics of that single target person. For example, if the personalized information of the target user corresponding to the target is "crispy noodles, fragrant and crunchy", then the promotional information generated based on the target person's language template and personalized information would be "Wow, these crispy noodles are so delicious, I want to eat more!"

[0168] When there are multiple target individuals, corresponding promotional information can be generated based on the characteristics of each target individual and the selected target audience. For example, it can be presented through dialogue: Target individual A: "Wow, this crispy noodle is so delicious, I want more!" Target individual B: "Okay, then come with me to XXX to place an order." Alternatively, promotional information can be generated based on the relationships between each target individual, their characteristics, and the selected target audience. For example, Zhao Wu's relationships could include Zhao San, Zhao Er, Li Er, Sun Er, and Yu Yi. When generating promotional information based on the target individual Zhao Wu, it can be generated together with Zhao Wu's related individuals. For example, the promotional information generated based on Zhao Wu could be "Come find me (referring to Zhao Wu) with Sun Er for a trip to XX Mountain."

[0169] Step 106: Integrate promotional information into the content of the target page.

[0170] This promotional information is used to display promotional information during the process of presenting the target page on the target user's corresponding terminal.

[0171] In some embodiments, the terminal can integrate promotional information into the content of the target page in the following way: obtain the integration template corresponding to the target page, which describes the layout position of the integrated promotional information and the content of the target page in the target page; add the promotional information and the content of the target page to the corresponding layout positions in the target page respectively, so as to integrate the promotional information into the content of the target page.

[0172] In practical applications, a pre-set integration template for the target page can be used. This template describes the layout and position of the integrated promotional information and the target page's content within the target page. See [link to template]. Figure 4 , Figure 4 This is a schematic diagram illustrating the process of integrating promotional information into the content of a target page, as provided in an embodiment of this application. Here, in the integration template corresponding to the target page, position area ① on the target page represents the layout position of the target page content, and position area ② on the target page represents the layout position of the promotional information. When integrating promotional information into the content of the target page, the promotional information and the target page content are added to their respective layout positions on the target page; that is, the target page content is added to position area ①, and the promotional information is added to position area ②, thus obtaining the content of the target page integrated with the promotional information.

[0173] In some embodiments, when there are at least two promotional messages and each promotional message is generated based on the characteristics of different target individuals, the terminal can obtain the click data of the target user for each promotional message; based on the click data, the target individual corresponding to at least one promotional message whose click data meets the promotion conditions is determined as the first promotional individual;

[0174] Based on this, the terminal can set the priority of the promotion information corresponding to the first promoter to be higher than the priority of the promotion information of the second promoter; wherein, the second promoter is a target person other than the first promoter among different target persons; when the content of the new page includes the first promoter and the second promoter, after selecting the target to be promoted related to the content of the new page, the corresponding new promotion information is generated based on the characteristics of the selected target to be promoted related to the content of the new page and the first promoter; the new promotion information is then integrated into the content of the new page.

[0175] In practical applications, target users may encounter multiple promotional messages while reading an article. These promotional messages may be generated based on the characteristics of different target individuals. Furthermore, target users may have varying degrees of liking for each individual while reading the article; therefore, the attractiveness of promotional messages generated based on different target individuals will also differ. Thus, the terminal can also acquire click data from the target user regarding each promotional message during the reading process. Based on this click data, it can determine which promotional messages meet the promotion criteria (e.g., click data reaching a click threshold), and then identify the target individual corresponding to the promotional message that meets the criteria as the primary promotional individual.

[0176] For the target user, the promotional effect of the first promoter is higher than that of other target promoters. Therefore, the terminal can set the priority of the promotional information corresponding to the first promoter higher than that of the promotional information of the second promoter, where the second promoter is the target person other than the first promoter. In the subsequent promotion process, if it is determined that the content of the new page that the target user wants to read contains both the first promoter and the second promoter, since the promotional information corresponding to the first promoter has a higher priority than that of the second promoter, new promotional information can be generated based on the characteristics of the first promoter and the recommended objects related to the content of the new page. This new promotional information is then integrated into the content of the new page and displayed to the target user through the terminal.

[0177] In some embodiments, when there are at least two promotional messages and each promotional message is generated based on the characteristics of different target individuals, the terminal can obtain the click data of the target user for each promotional message; based on the click data, determine the target individual corresponding to at least one promotional message whose click data meets the promotion conditions as the target promotion individual; obtain the comment content of the target user on the article; based on the comment content, update the language template corresponding to each target promotion individual, which is used to integrate the characteristics of the selected target and the target individual to generate promotional messages.

[0178] In practical applications, target users may encounter multiple promotional messages while reading an article. These messages may be generated based on the characteristics of different target individuals. Since target users may have varying degrees of liking for each other during the reading process, the attractiveness of promotional messages generated based on different target individuals will also differ. Therefore, the terminal can also acquire click data from the target user regarding each promotional message during the reading process. Based on this click data, it can identify promotional messages that meet promotional conditions (e.g., click data reaching a click threshold), and then determine the target individual corresponding to these promotional messages as the target promoter. For the target user, the promotional effect of this target promoter is higher than that of other target individuals. Therefore, during the target user's subsequent reading, the frequency of promotional messages generated based on this target promoter can be increased, i.e., more promotional messages can be used to promote the target product to the target user using this target promoter.

[0179] Furthermore, to enhance the promotional effectiveness of the target promoter, their voice template can be optimized and updated to make the generated promotional information more appealing to target users. Specifically, comments from target users on articles can be obtained. These comments may express their opinions and hopes regarding the target promoter, related storylines, or analyses of the target promoter's personality. Therefore, the target promoter's language template can be updated based on these comments to obtain an updated language template. In subsequent recommendations to target users, the updated language template of the target promoter and the promoted individual can be used to generate promotional information.

[0180] By applying the above embodiments of this application, this application analyzes the content of the target person in the article to obtain the characteristics of the target person, selects the target object related to the content of the target page from the set of target objects that are compatible with the target users to be promoted, and generates corresponding promotional information based on the characteristics of the target person and the target object. The promotional information is then integrated into the content of the target page, so that the target user's terminal displays the promotional information during the presentation of the target page.

[0181] Here, promotional information is generated based on the characteristics of the characters in the article and the target audience related to the content of the target page. This improves the matching degree between promotional information and page content, as well as their integration. At the same time, because the selected target audience is compatible with the target users, the promotional information is no longer static, realizing personalized generation of promotional information for users. This increases the attractiveness of the promotional information to users, as well as users' goodwill and click-through rate, thereby improving the promotional effect.

[0182] The following describes the method for processing promotional information provided in the embodiments of this application. In some embodiments, the method for processing promotional information provided in the embodiments of this application may be implemented by a server or a terminal alone, or by a server and a terminal working together. The following describes the method for processing promotional information provided in the embodiments of this application using a terminal implementation as an example. See also Figure 5 , Figure 5 This is a flowchart illustrating the promotional information processing method provided in this application embodiment. The promotional information processing method provided in this application embodiment includes:

[0183] Step 201: The terminal receives a request from the target user to view the content of the target page corresponding to the article.

[0184] Step 202: In response to the viewing operation, the target page is displayed, and when there is a target page related to the content of the target page in the set of target pages that is compatible with the target user, the promotion information corresponding to the target page is displayed during the display of the target page.

[0185] The promotional information is generated based on the target audience and the personality of the target person described in the article, where the target person is at least one of at least one of the characters described in the article.

[0186] The terminal receives a viewing operation triggered by a target user for the content of a target page corresponding to an article. In response to this viewing operation, it determines whether there is a target object related to the content of the target page in the set of target objects suitable for the target user. If so, the target page containing the content of the target page is displayed, and during the display of the target page, the promotional information corresponding to the target object is also displayed.

[0187] See Figure 6 , Figure 6 This is a schematic diagram illustrating the presentation of promotional information provided in this application embodiment. Here, the promotional information of the target user "King of Glory XX" is presented in the bottom and middle areas of the target page using methods such as the target user's avatar and virtual image. Figure 6The sub-image (1) uses the avatar of the target person "Lan X Yu" to display the promotional information of the recommended object "Wang Zhe XX" in the bottom area of ​​the target page: "Come to Wang Zhe XX and grow together in the game." Figure 6 The sub-image (2) in the target page presents the promotional information of the recommended object "King XX" in the bottom area of ​​the target page through the avatar of the target person "Lan X Yu"; "I have eaten all the eggshells, click on the ad to help me grow up"; Figure 6 The sub-image (3) in the target page presents the promotional information of the target "King XX" in the lower middle part of the target page through the virtual image of the target person "Lan X Xiao". "Raising XX is not easy, I rely on you to click on the ads to make a living."

[0188] By applying the above embodiments of this application, this application analyzes the content of the target person in the article to obtain the characteristics of the target person, selects the target object related to the content of the target page from the set of target objects that are compatible with the target users to be promoted, and generates corresponding promotional information based on the characteristics of the target person and the target object. The promotional information is then integrated into the content of the target page, so that the target user's terminal displays the promotional information during the presentation of the target page.

[0189] Here, promotional information is generated based on the characteristics of the characters in the article and the target audience related to the content of the target page. This improves the matching degree between promotional information and page content, as well as their integration. At the same time, because the selected target audience is compatible with the target users, the promotional information is no longer static, realizing personalized generation of promotional information for users. This increases the attractiveness of the promotional information to users, as well as users' goodwill and click-through rate, thereby improving the promotional effect.

[0190] The following continues to describe the method for processing promotional information provided in the embodiments of this application. The method for processing promotional information provided in the embodiments of this application can be implemented collaboratively by a terminal and a server. See also... Figure 7A , Figure 7A This is a flowchart illustrating the promotional information processing method provided in this application embodiment. The promotional information processing method provided in this application embodiment includes:

[0191] Step 301: In response to the target user's request to view the content of the target page corresponding to the article, the terminal sends a request to the server to retrieve the content of the target page.

[0192] Step 302: The server responds to the request to retrieve the content of the target page and identifies the target person from at least one person described in the content of the article.

[0193] Step 303: Extract content corresponding to the target person from the article and encode it to obtain the target person's vector; input the target person's vector into the machine learning model, and use the machine learning model to predict the target person's features to obtain the target person's features.

[0194] Step 304: Obtain the content of the target page corresponding to the article, as well as the set of target users to be promoted.

[0195] Here, the set of objects to be promoted is constructed by selecting at least one object from a large pool of objects that matches the target user's profile information. This large pool of objects can be pre-defined, such as cars, cosmetics, instant noodles, and plush toys. Based on the target user's profile information, at least one object matching the profile information can be selected from the large pool of objects. For example, if the target user is female, then cosmetics and plush toys could both be objects matching the profile information. After selecting at least one object matching the profile information, a set of objects suitable for the target user is constructed based on this at least one object.

[0196] Step 305: Perform word segmentation on the content of the target page, and perform named entity recognition on the obtained multiple word segments to obtain the target objects contained in the content of the target page; match the target objects with each object to be promoted in the set of objects to be promoted to obtain the matching results, and take the objects to be promoted that match the target objects in the matching results as objects to be promoted that are related to the content of the target page.

[0197] Step 306: Generate corresponding promotional information based on the characteristics of the selected target audience and individuals.

[0198] Here, the target audience can have corresponding text materials set to describe its basic information, such as detailed parameters, promotional points, and user-attracting features. When generating promotional information for the target audience, the basic information can be obtained first. To further enhance the appeal of the promotional information to the target users, the terminal can also extract information tailored to the target users from the basic information, serving as personalized information for each user. Then, the language template corresponding to the target person's characteristics is obtained. Finally, based on this language template and the personalized information, the promotional information for the target audience is generated. Here, the language template corresponding to the target person's characteristics can be pre-set or learned from a trained machine learning model, and can include semantic rules, commonly used characters, words, and sentence structures.

[0199] For example, the target person's language template includes "ya", "wa", "delicious", and "want to eat". The commonly used sentence structure is an exclamation. When the personalized information of the target user corresponding to the object to be recommended is "crispy noodles, fragrant and crispy", then the promotional information generated based on the target person's language template and personalized information is "Wow, this crispy noodle is fragrant and crispy and delicious. I want to eat it again!"

[0200] Step 307: Integrate the promotional information into the content of the target page, and send the content of the target page with integrated promotional information to the terminal.

[0201] Step 308: The terminal displays the target page, and during the process of displaying the target page, it displays the promotional information corresponding to the target to be promoted.

[0202] Here, the terminal displays the content of the target page and promotional information.

[0203] By applying the above embodiments of this application, this application analyzes the content of the target person in the article to obtain the characteristics of the target person, selects the target object related to the content of the target page from the set of target objects that are compatible with the target users to be promoted, and generates corresponding promotional information based on the characteristics of the target person and the target object. The promotional information is then integrated into the content of the target page, so that the target user's terminal displays the promotional information during the presentation of the target page.

[0204] Here, promotional information is generated based on the characteristics of the characters in the article and the target audience related to the content of the target page. This improves the matching degree between promotional information and page content, as well as their integration. At the same time, because the selected target audience is compatible with the target users, the promotional information is no longer static, realizing personalized generation of promotional information for users. This increases the attractiveness of the promotional information to users, as well as users' goodwill and click-through rate, thereby improving the promotional effect.

[0205] The following continues to describe the method for processing promotional information provided in the embodiments of this application. See also Figure 7B , Figure 7B This is a flowchart illustrating the promotional information processing method provided in this application embodiment. The promotional information processing method provided in this application embodiment includes:

[0206] Step 501: The server identifies the target person from at least one person described in the article's content.

[0207] In this embodiment of the application, promotional information for the target page of the corresponding article can be generated in advance for each target user, so that when the user views the content of the target page of the article, the corresponding promotional information can be found directly. Specifically, the server first determines the target character from at least one character described in the article content. The target character can be one or more. In practical applications, the target character is at least one protagonist described in the article content.

[0208] Step 502: Analyze the content of the article that corresponds to the target person to obtain the target person's characteristics.

[0209] Here, after identifying the target person described in the article, the content within the article that corresponds to the target person is retrieved. This content is then analyzed to determine the target person's characteristics, which may include their personality traits, behavioral patterns, language characteristics, past experiences, and relationships with other individuals described in the article.

[0210] Step 503: Determine the target users for the promotion of the target objects, and the set of target objects that are suitable for the target users.

[0211] In some embodiments, step 503 can be implemented by steps 5031 to 5034, including: step 5031: determining multiple candidate users; step 5032: obtaining user profiles for each candidate user; step 5033: classifying multiple candidate users based on user profiles to obtain at least two user sets, and using each user set as target users; step 5034: determining a set of objects to be promoted that are compatible with the target users.

[0212] Here, the candidate user can be a registered user of the client where the aforementioned article is located (such as a novel reading client). In practical applications, user profiles of each candidate user can be obtained, such as basic user information (user ID, age, gender, etc.), user reading interests, and user preferences for article characters. Then, based on the user profiles, the identified candidate users are categorized into multiple user sets, such as a set of female users, a set of users aged 18-25, a set of users who like science fiction novels, etc. Each user set is then used as the target user group, and a set of promotional items suitable for that target user group is determined. For example, for the set of female users, the set of promotional items suitable for them could include cosmetics, plush toys, small accessories, etc.

[0213] Step 504: Select objects from the set of objects to be promoted that are related to the content of the target page corresponding to the article.

[0214] Here, after obtaining a set of promotional objects suitable for the target users, the promotional objects related to the content of the target page are selected from this set. For example, if the target page content contains a description of instant noodles, then instant noodles can be selected as a promotional object related to the target page content from the promotional object set. This can be achieved by semantic analysis and matching of the target page content and the textual content of each promotional object in the promotional object set; alternatively, named entity recognition can be performed on the target page content to obtain the included entities (such as place names / brands / products), and the identified entities are matched with each promotional object in the promotional object set to determine the promotional objects related to the content of the target page corresponding to the article.

[0215] Step 505: Generate corresponding promotional information based on the characteristics of the selected target audience and individuals.

[0216] Here, after determining the target audience related to the content of the target page and the characteristics of the target person in the article, the server generates promotional information corresponding to the target audience based on the characteristics of the target audience and the target person. This promotional information is then displayed on the target page, allowing users to view the promotional information of the target audience related to the content of the target page. Specifically, the server can obtain basic information about the target audience and determine information suitable for the target user from this basic information, serving as personalized information for the corresponding target user; obtain language templates corresponding to the characteristics of the target person; and generate corresponding promotional information based on the language templates and personalized information.

[0217] In practical applications, the target person can be one or multiple. When there is only one target person, promotional information is generated based on the selected target and the characteristics of that single target person. For example, if the personalized information of the target user corresponding to the target is "crispy noodles, fragrant and crunchy", then the promotional information generated based on the target person's language template and personalized information would be "Wow, these crispy noodles are so delicious, I want to eat more!"

[0218] When there are multiple target individuals, corresponding promotional information can be generated based on the characteristics of each target individual and the selected target audience. For example, it can be presented through dialogue: Target individual A: "Wow, this crispy noodle is so delicious, I want more!" Target individual B: "Okay, then come with me to XXX to place an order." Alternatively, promotional information can be generated based on the relationships between each target individual, their characteristics, and the selected target audience. For example, Zhao Wu's relationships could include Zhao San, Zhao Er, Li Er, Sun Er, and Yu Yi. When generating promotional information based on the target individual Zhao Wu, it can be generated together with Zhao Wu's related individuals. For example, the promotional information generated based on Zhao Wu could be "Come find me (referring to Zhao Wu) with Sun Er for a trip to XX Mountain."

[0219] Step 506: Integrate promotional information into the content of the target page.

[0220] This promotional information is used to display promotional information during the process of presenting the target page on the target user's corresponding terminal.

[0221] In some embodiments, after generating promotional information for the target page of the article corresponding to each target user, the server can send the promotional information to the terminal in the following ways: receiving a viewing operation for the content of the target page corresponding to the article; determining the target user set to which the user corresponding to the viewing operation belongs; determining the target promotional information corresponding to the target user set and the content of the target page; and returning the content of the target page incorporating the target promotional information to the terminal of the user corresponding to the viewing operation.

[0222] In this embodiment, promotional information for the target page of the corresponding article can be generated in advance for each target user. The target user can be each registered user of the client where the article is located (e.g., a novel reading client), or at least two user sets obtained by classifying multiple registered users. In practice, when the number of registered users exceeds a user threshold, the registered users can first be classified according to their respective interests to obtain at least two user sets, which can then be used as target users, thereby reducing the hardware resources consumed in generating promotional information. When the number of registered users is less than the user threshold, each registered user can be used as a target user; classification is also possible, and this embodiment does not impose any restrictions.

[0223] After generating promotional information for the target page of the article corresponding to each target user, if a view operation is received for the content of the target page corresponding to the article, the system responds to the view operation by first determining the user set to which the user to which the view operation belongs, and using this user set as the target user set; then, from the pre-generated promotional information for the target page corresponding to each target user, the system searches for and determines the target promotional information corresponding to the target user set and the content of the target page; and then returns the content of the target page, which incorporates the target promotional information, to the terminal of the user corresponding to the view operation, so that the user corresponding to the view operation sees the corresponding promotional information while viewing the content of the target page.

[0224] By applying the above embodiments of this application, this application analyzes the content of the target person in the article to obtain the characteristics of the target person, selects the target object related to the content of the target page from the set of target objects that are compatible with the target users to be promoted, and generates corresponding promotional information based on the characteristics of the target person and the target object. The promotional information is then integrated into the content of the target page, so that the target user's terminal displays the promotional information during the presentation of the target page.

[0225] Here, promotional information is generated based on the characteristics of the characters in the article and the target audience related to the content of the target page. This improves the matching degree between promotional information and page content, as well as their integration. At the same time, because the selected target audience is compatible with the target users, the promotional information is no longer static, realizing personalized generation of promotional information for users. This increases the attractiveness of the promotional information to users, as well as users' goodwill and click-through rate, thereby improving the promotional effect.

[0226] The following describes an exemplary application of the embodiments of this application in a real-world application scenario.

[0227] Currently, advertising recommendations within novels are only made to the individual user, and advertisers choose to embed ads based on readership. However, most online novels rarely meet advertisers' budgets. Furthermore, embedded advertising has limited brand applicability, especially in text-based media like novels, often only suitable for well-known brands, making it inaccessible to paid advertising. Traditional methods involve advertisers signing contracts with authors to embed fixed product names and highlight a character's product usage habits within the text. However, this approach fails to provide personalized recommendations to individual readers.

[0228] Among related technologies, "Video in" is applicable to the embedding of dynamic video advertisements. The principle is to analyze the video content and, through technologies such as realistic props and dynamic textures, automatically fill and composite advertisements related to the video content into the empty spaces within the video content. See [link to related technology]. Figure 8 , Figure 8 This is a schematic diagram illustrating the presentation of promotional information in videos provided by related technologies. When items such as toys and crayfish appear in the video, corresponding advertisements are displayed in the video using dynamic stickers. However, the content of the advertisements remains fixed, pre-set by the operators, and cannot be personalized to different viewers.

[0229] Therefore, the relevant technologies either analyze the characters and script structure of the novel's text information, and the analysis results are not combined with advertising recommendations; or they use artificial intelligence technology to analyze video scenes, identify gaps that can be inserted, and then synthesize advertising materials. Such insertion technologies have not been used in novel scenes.

[0230] Based on this, embodiments of this application provide a method for processing promotional information to at least solve the aforementioned problems, which will be described in detail below. The method for processing promotional information provided in embodiments of this application can be implemented collaboratively by a terminal and a server. See also Figure 9 , Figure 9 This is a flowchart illustrating the method for processing promotional information provided in this application embodiment. The method for processing promotional information provided in this application embodiment includes:

[0231] Step 401: The terminal displays the text content of the novel viewed by the target user.

[0232] Step 402: The server segments the text of the novel to identify the target characters.

[0233] Here, the server performs word segmentation on the text content of the novel being viewed by the target user, obtaining multiple word segments corresponding to the novel. Then, it counts the number of target word segments with the part-of-speech (NOS) among these multiple segments, sorts the target segments in descending order of their number, and identifies the characters with the highest number of segments (e.g., 4) as the target characters. Specifically, a custom feature dictionary can be loaded using a word segmentation tool to improve segmentation performance; the posseg tool is used to count the number of all NOS segments in the text content, sorting them in descending order to obtain the target characters, such as A1 / A2 / B1 / B2, etc.

[0234] Step 403: Generate the target person's vector, and use a machine learning model to predict the target person's features from the vector.

[0235] Here, the server extracts the text descriptions of the target characters from the novel to obtain an analysis sample for each target character, which is the text content of the target character in the novel. When analyzing the personality traits of the target character, the accuracy of the personality trait analysis is based on the amount of text information. Therefore, it is necessary to determine whether the amount of text content of the target character meets the personality trait analysis conditions (such as a text information amount threshold). Then, the personality traits of the target characters whose text information amount meets the personality trait analysis conditions are analyzed.

[0236] Personality analysis is a complex field within psychology, encompassing both static and dynamic structures. This application utilizes personality analysis to predict what a target person might say, intelligently generating promotional information for the target audience in a tone consistent with their personality traits. Therefore, it predicts and outputs the target person's personality characteristics based on well-established psychological theories (i.e., the Enneagram).

[0237] Specifically, the text content of the target person is input into the doc2Vec model, which generates a person vector for the target person. Then, the person vector is input into a machine learning model, such as a random forest model built based on multiple decision trees. The machine learning model then predicts the person features of the target person, such as personality traits, to obtain the person features of the target person. For example, the personality traits of the target person A1 / A2 / B1 / B2 are x1 / x2 / x3 / x4 respectively.

[0238] Here, the target person's characteristics can include their personality traits, behavioral characteristics, language characteristics, life experiences, and relationships with other people described in the article. For example, using the target person's characteristics as relational features, see [link to relevant documentation]. Figure 10 , Figure 10 This is a schematic diagram illustrating the relationships between multiple characters described in the article provided in this application embodiment. Here, the target characters are Zhao Wu, Zhao San, and Zhao Er, and the connecting lines represent the relationships between the characters. For example, Zhao Wu's relationships could include Zhao San, Zhao Er, Li Er, Sun Er, and Yu Yi. When generating promotional information based on the target character Zhao Wu, it can be generated together with Zhao Wu's related characters. For example, the promotional information generated based on Zhao Wu could be "Come find me (referring to Zhao Wu) with Sun Er for a trip to XX Mountain."

[0239] Step 404: Segment the text content of the target chapter of the novel to obtain multiple segments of the target chapter, and perform named entity recognition on the multiple segments to obtain the entities contained in the target chapter.

[0240] Step 405: Determine whether it is necessary to generate promotional information corresponding to the entity. If yes, proceed to step 406; otherwise, proceed to step 407.

[0241] Here, for the entities identified in the target chapter, it is determined whether to display them or whether there is suitable ad inventory to display them, based on the context of the target chapter and the existing ad delivery logic.

[0242] Specifically, the text content of the target chapter of the novel being viewed by the target user is segmented into multiple words. Then, named entity recognition is performed on these multiple words to identify the entities contained in the target chapter, such as place names, brands, and products. Simultaneously, sentiment analysis is performed on the text content to identify the positive or negative sentiment of these entities (such as place names, brands, and products). If the sentiment is negative (e.g., plane crash, diarrhea), ads for that category are not recommended; if the sentiment is positive, ads can be targeted at the entities contained in the target chapter, such as cars and instant noodles.

[0243] Furthermore, based on the advertising placement logic, the entities to be recommended are determined from the identified entities. Here, the advertising placement logic can be the target audience selected by the advertiser during the placement process, such as car promotions only being displayed to men in first-tier cities. According to the advertising placement logic, entities corresponding to the target users are selected from the entities identified in step 404 as the entities to be recommended. For example, if the target users are men in first-tier cities, then the entity "car" can be selected as the entity to be recommended to the target users. At this point, promotional information corresponding to the entity (i.e., the entity to be recommended) needs to be generated.

[0244] Here, the server performs target promotion for each chapter of the novel. That is, it identifies the target audience related to each chapter. Therefore, when a user wants to read a target chapter, the target chapter is analyzed to determine the target audience related to the target chapter.

[0245] Step 406: Based on the target person's characteristics and the recommended individuals, generate promotional information and return it to the terminal.

[0246] Here, promotional information for the target audience is intelligently generated based on the text material and the target person's characteristics (such as personality traits x1 / x2 / x3 / x4). In practical applications, when the promotional information is displayed on the terminal, it can be introduced by a "spokesperson" (i.e., the target person). Since novels with high traffic undergo manual review, character images of the target person can also be generated to be displayed alongside the promotional information, improving the user experience.

[0247] Step 407: Do not generate promotional information for the objects to be recommended.

[0248] Step 408: The terminal displays promotional information at the end of the target chapter of the novel.

[0249] Here, users can scroll down or turn pages to view the end of the target chapter of the novel, at which point the device displays promotional information for the novel to be recommended. In practical applications, see... Figure 6 Based on visual intensity, the following images sequentially illustrate various combinations, including using the target person's portrait, using the target person's full body, and even using a scene.

[0250] Step 409: When a click action is received from a target user on the promotional information, send the click data to the server.

[0251] Step 410: Identify the target audience of the promotional information whose click data meets the promotion conditions as the target promotional audience, so as to optimize the language template of the target promotional audience.

[0252] Here, language templates are used to integrate the characteristics of the selected target audience and individuals to generate promotional information.

[0253] Here, based on click data of promotional messages from different target users, high-performing promotional figures are identified, and these figures are prioritized for generating promotional messages for target users. Additionally, the text of the promotional messages can be optimized by incorporating comment content.

[0254] By applying the above embodiments of this application, the method of recommending advertisements based on text analysis and content understanding and utilizing scenes / characters in novels not only improves the matching between advertisements and content, but is also applicable to the access of bidding advertisements, thereby increasing users' favorability and click-through rate of advertisements in novel scenes.

[0255] The following describes the promotional information processing apparatus 555 provided in the embodiments of this application. In some embodiments, the promotional information processing apparatus may be implemented as a software module. See also Figure 11 , Figure 11 This is a schematic diagram of the structure of the promotional information processing device 555 provided in this application embodiment. The promotional information processing device 555 provided in this application embodiment includes:

[0256] The first determining module 5551 is used to determine the target person from at least one person described in the content of the article;

[0257] Analysis module 5552 is used to analyze the content of the article that corresponds to the target person, and obtain the characteristics of the target person;

[0258] The second determining module 5553 is used to determine the target users of the objects to be promoted, and the set of objects to be promoted that are compatible with the target users;

[0259] The selection module 5554 is used to select, from the set of objects to be promoted, objects that are related to the content of the target page corresponding to the article;

[0260] The generation module 5555 is used to generate corresponding promotional information based on the selected target object and the characteristics of the target person.

[0261] The fusion module 5556 is used to fuse the promotional information into the content of the target page; wherein the promotional information is presented during the process of the target user's terminal displaying the target page.

[0262] In some embodiments, the first determining module 5551 is further configured to perform semantic analysis on the content of the article to determine the target content corresponding to each of the figures;

[0263] The amount of information in the target content corresponding to each of the aforementioned individuals is statistically analyzed.

[0264] Based on the amount of information in the target content corresponding to each of the aforementioned individuals, the individuals corresponding to the target content whose information content meets the criteria for target individuals are identified as target individuals.

[0265] In some embodiments, the first determining module 5551 is further configured to perform word segmentation on the content of the article to obtain multiple word segments corresponding to the article;

[0266] Identify the target segmentation word belonging to the noun category from the multiple segmentation words;

[0267] Count the number of words segmented for the target words corresponding to each of the aforementioned characters;

[0268] Based on the number of target words for each of the aforementioned figures, the figures corresponding to the target words whose number of words meets the criteria for target figures are identified as the target figures.

[0269] In some embodiments, the second determining module 5553 is further configured to determine a plurality of candidate users;

[0270] Obtain the user profile of each of the candidate users;

[0271] Based on the user profile, the multiple candidate users are classified to obtain at least two user sets, and each user set is used as the target user.

[0272] In some embodiments, the apparatus further includes:

[0273] The receiving module is used to receive viewing operations for the content of the target page corresponding to the article;

[0274] Determine the target user set to which the user corresponding to the viewing operation belongs;

[0275] Determine the target promotional information that corresponds to the target user set and the content of the target page, and

[0276] The content of the target page, which incorporates the target promotional information, is returned to the terminal of the user corresponding to the viewing operation.

[0277] In some embodiments, the analysis module 5552 is further configured to extract content corresponding to the target person from the content of the article;

[0278] The content corresponding to the target person is encoded to obtain the person vector corresponding to the target person;

[0279] The person's vector is input into a machine learning model, and the machine learning model predicts the person's features from the person's vector to obtain the person's features.

[0280] In some embodiments, the analysis module 5552 is further configured to extract content corresponding to the target person from the content of the article;

[0281] From at least two feature dimensions, feature extraction is performed on the content corresponding to the target person to obtain the intermediate person features corresponding to each feature dimension of the target person;

[0282] Based on the intermediate features of the target person corresponding to each of the feature dimensions, the character features of the target person are constructed.

[0283] In some embodiments, the second determining module 5553 is further configured to obtain multiple objects to be promoted and the profile information corresponding to the target user;

[0284] From the plurality of objects to be promoted, at least one object to be promoted that matches the profile information is selected;

[0285] The set of objects to be promoted is constructed based on the at least one object to be promoted.

[0286] In some embodiments, the selection module 5554 is further configured to perform semantic analysis on the content of the target page to obtain the page content of the target page;

[0287] Semantic recognition is performed on the text materials of each object to be promoted in the set of objects to be promoted to obtain the text material content of each object to be promoted.

[0288] The page content is matched with the text material content of each of the objects to be promoted to obtain the matching degree between the content of the target page and each of the objects to be promoted;

[0289] From the set of objects to be promoted, select objects whose matching degree reaches the matching degree threshold as objects to be promoted that are related to the content of the target page.

[0290] In some embodiments, the selection module 5554 is further configured to perform word segmentation on the content of the target page to obtain multiple word segments corresponding to the target page;

[0291] Named entity recognition is performed on the multiple word segments to obtain the target objects contained in the content of the target page;

[0292] The target object is matched with each object to be promoted in the set of objects to be promoted to obtain the matching results;

[0293] The objects to be promoted that match the target object in the matching results are taken as objects to be promoted that are related to the content of the target page.

[0294] In some embodiments, the generation module 5555 is further configured to obtain basic information of the object to be promoted, and determine information that is compatible with the target user from the basic information, as personalized information corresponding to the target user;

[0295] Obtain the language template corresponding to the characteristics of the target person;

[0296] Based on the language template and the personalized information, corresponding promotional information is generated.

[0297] In some embodiments, the fusion module 5556 is further configured to obtain a fusion template corresponding to the target page, wherein the fusion template is used to describe the fused promotional information and the layout position of the content of the target page in the target page;

[0298] The promotional information and the content of the target page are added to the corresponding layout positions on the target page to integrate the promotional information into the content of the target page.

[0299] In some embodiments, the apparatus further includes:

[0300] The promotional user identification module is used to obtain the click data of the target user for each of the promotional messages when there are at least two promotional messages and each promotional message is generated based on the characteristics of different target users;

[0301] Based on the click data, the target person corresponding to at least one promotional message whose click data meets the promotion conditions is identified as the first promotional person.

[0302] Accordingly, the device further includes:

[0303] The setting module is used to set the priority of the promotion information corresponding to the first promoter to be higher than the priority of the promotion information of the second promoter; wherein, the second promoter is a target person other than the first promoter among the different target persons;

[0304] When the content of the new page includes the first promoter and the second promoter, after selecting the target to be promoted that is related to the content of the new page, the corresponding new promotion information is generated based on the characteristics of the selected target to be promoted that is related to the content of the new page and the first promoter.

[0305] The new promotional information is integrated into the content of the new page.

[0306] In some embodiments, the apparatus further includes:

[0307] The update module is used to obtain the click data of the target user for each of the promotional messages when there are at least two promotional messages and each promotional message is generated based on the characteristics of different target users;

[0308] Based on the click data, the target person corresponding to at least one promotional message whose click data meets the promotion conditions is identified as the target promotional person.

[0309] Obtain the comments from the target user regarding the article;

[0310] Based on the comments, the language templates corresponding to each of the target promotional figures are updated. The language templates are used to integrate the characteristics of the selected target and the target figures to generate the promotional information.

[0311] By applying the above embodiments of this application, this application analyzes the content of the target person in the article to obtain the characteristics of the target person, selects the target object related to the content of the target page from the set of target objects that are compatible with the target users to be promoted, and generates corresponding promotional information based on the characteristics of the target person and the target object. The promotional information is then integrated into the content of the target page, so that the target user's terminal displays the promotional information during the presentation of the target page.

[0312] Here, promotional information is generated based on the characteristics of the characters in the article and the target audience related to the content of the target page. This improves the matching degree between promotional information and page content, as well as their integration. At the same time, because the selected target audience is compatible with the target users, the promotional information is no longer static, realizing personalized generation of promotional information for users. This increases the attractiveness of the promotional information to users, as well as users' goodwill and click-through rate, thereby improving the promotional effect.

[0313] The following describes the promotional information processing apparatus 600 provided in the embodiments of this application. In some embodiments, the promotional information processing apparatus may be implemented using software modules. See also Figure 12 , Figure 12 This is a schematic diagram of the structure of the promotional information processing device 600 provided in this application embodiment. The promotional information processing device 600 provided in this application embodiment includes:

[0314] The receiving module 610 is used to receive the viewing operation triggered by the target user for the content of the target page corresponding to the article;

[0315] The presentation module 620 is used to respond to the viewing operation, present the target page, and when there is a target page related to the content of the target page in the set of target pages that are compatible with the target user, the presentation module 620 presents the promotion information corresponding to the target page during the presentation of the target page.

[0316] The promotional information is generated based on the target person's personality as described in the article and the content of the article.

[0317] By applying the above embodiments of this application, this application analyzes the content of the target person in the article to obtain the characteristics of the target person, selects the target object related to the content of the target page from the set of target objects that are compatible with the target users to be promoted, and generates corresponding promotional information based on the characteristics of the target person and the target object. The promotional information is then integrated into the content of the target page, so that the target user's terminal displays the promotional information during the presentation of the target page.

[0318] Here, promotional information is generated based on the characteristics of the characters in the article and the target audience related to the content of the target page. This improves the matching degree between promotional information and page content, as well as their integration. At the same time, because the selected target audience is compatible with the target users, the promotional information is no longer static, realizing personalized generation of promotional information for users. This increases the attractiveness of the promotional information to users, as well as users' goodwill and click-through rate, thereby improving the promotional effect.

[0319] This application embodiment also provides an electronic device, the electronic device comprising:

[0320] Memory, used to store executable instructions;

[0321] The processor, when executing executable instructions stored in the memory, implements the promotional information processing method provided in the embodiments of this application.

[0322] This application also provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the promotional information processing method provided in this application.

[0323] This application also provides a computer-readable storage medium storing executable instructions, which, when executed by a processor, implement the promotional information processing method provided in this application.

[0324] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a variety of devices including one or any combination of the above-mentioned memories.

[0325] In some embodiments, executable instructions may take the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0326] As an example, executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple collaborating files (e.g., a file that stores one or more modules, subroutines, or code sections).

[0327] As an example, executable instructions can be deployed to execute on a single computing device, or on multiple computing devices located in one location, or on multiple computing devices distributed across multiple locations and interconnected via a communication network.

[0328] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, and improvements made within the spirit and scope of this application are included within the scope of protection of this application.

Claims

1. A method for processing promotional information, characterized in that, The method includes: Identify the target character from at least one character described in the article; The content of the article corresponding to the target person is analyzed to obtain the character characteristics of the target person. The character characteristics include at least one or more of the following: personality characteristics, behavioral characteristics, language characteristics, or relationship characteristics with other people described in the article. Identify the target users for the promotion of the target objects, and identify the set of target objects that are compatible with the target users; From the set of objects to be promoted, select objects that are related to the content of the target page corresponding to the article; Information that matches the target user is determined from the basic information of the target object to be promoted, and this information is used as the personalized information corresponding to the target user. Based on the language template corresponding to the target person's characteristics and the personalized information, corresponding promotional information is generated. The language template includes semantic rules, words, phrases, and sentence structures commonly used by the target person. Obtain the fusion template corresponding to the target page. The fusion template is used to describe the layout position of the promotional information and the content of the target page in the target page after fusion. The promotional information and the content of the target page are added to the corresponding layout positions on the target page to integrate the promotional information into the content of the target page; wherein, the promotional information is presented during the process of the target user's terminal displaying the target page; Based on the comments of the target users on the article, the language templates corresponding to each target promoter are updated. The target promoter is a person who meets the promotion conditions. The language template is used to integrate the characteristics of the selected target to be promoted and the target person to generate the promotion information.

2. The method as described in claim 1, characterized in that, The process of identifying the target person from at least one person described in the article's content includes: Semantic analysis is performed on the content of the article to determine the target content corresponding to each of the aforementioned figures; The amount of information in the target content corresponding to each of the aforementioned individuals is statistically analyzed. Based on the amount of information in the target content corresponding to each of the aforementioned individuals, the individuals corresponding to the target content whose information content meets the criteria for target individuals are identified as target individuals.

3. The method as described in claim 1, characterized in that, The process of determining the target users for promotion includes: Identify multiple candidate users; Obtain the user profile of each of the candidate users; Based on the user profile, the multiple candidate users are classified to obtain at least two user sets, and each user set is used as the target user.

4. The method as described in claim 3, characterized in that, The method further includes: Received a request to view the content of the target page corresponding to the article; Determine the target user set to which the user corresponding to the viewing operation belongs; Determine the target promotional information that corresponds to the target user set and the content of the target page, and The content of the target page, which incorporates the target promotional information, is returned to the terminal of the user corresponding to the viewing operation.

5. The method as described in claim 1, characterized in that, The analysis of the content in the article corresponding to the target person to obtain the target person's characteristics includes: Extract content corresponding to the target person from the content of the article; The content corresponding to the target person is encoded to obtain the person vector corresponding to the target person; The person's vector is input into a machine learning model, and the machine learning model predicts the person's features from the person's vector to obtain the person's features.

6. The method as described in claim 1, characterized in that, The determination of the set of target users suitable for promotion includes: Obtain multiple target users to be promoted, as well as their profile information; From the plurality of objects to be promoted, at least one object to be promoted that matches the profile information is selected; The set of objects to be promoted is constructed based on the at least one object to be promoted.

7. The method as described in claim 1, characterized in that, The step of selecting promotional objects from the set of objects to be promoted that are related to the content of the target page corresponding to the article includes: Semantic analysis is performed on the content of the target page to obtain the page content of the target page; Semantic recognition is performed on the text materials of each object to be promoted in the set of objects to be promoted to obtain the text material content of each object to be promoted. The page content is matched with the text material content of each of the objects to be promoted to obtain the matching degree between the content of the target page and each of the objects to be promoted; From the set of objects to be promoted, select objects whose matching degree reaches the matching degree threshold as objects to be promoted that are related to the content of the target page.

8. The method as described in claim 1, characterized in that, The step of selecting promotional objects from the set of objects to be promoted that are related to the content of the target page corresponding to the article includes: The content of the target page is segmented into words to obtain multiple words corresponding to the target page. Named entity recognition is performed on the multiple word segments to obtain the target objects contained in the content of the target page; The target object is matched with each object to be promoted in the set of objects to be promoted to obtain the matching results; The objects to be promoted that match the target object in the matching results are taken as objects to be promoted that are related to the content of the target page.

9. The method as described in claim 1, characterized in that, The method further includes: When there are at least two promotional messages, and each promotional message is generated based on the characteristics of a different target person, the click data of the target user for each promotional message is obtained; Based on the click data, the target person corresponding to at least one promotional message whose click data meets the promotion conditions is identified as the first promotional person. Accordingly, the method further includes: The promotional information corresponding to the first promoter is given a higher priority than the promotional information of the second promoter; wherein, the second promoter is a target person other than the first promoter among the different target persons; When the content of the new page includes the first promoter and the second promoter, after selecting the target to be promoted that is related to the content of the new page, the corresponding new promotion information is generated based on the characteristics of the selected target to be promoted that is related to the content of the new page and the first promoter. The new promotional information is integrated into the content of the new page.

10. The method as described in claim 1, characterized in that, The method further includes: When there are at least two promotional messages, and each promotional message is generated based on the characteristics of a different target person, the click data of the target user for each promotional message is obtained; Based on the click data, the target person corresponding to at least one promotional message whose click data meets the promotion conditions is identified as the target promotional person. Obtain the comments from the target user regarding the article.

11. A method for processing promotional information, characterized in that, The method includes: Receives a request from the target user to view content on the target page corresponding to the article; In response to the viewing operation, the target page is displayed, and When there is a target page content-related target page in the set of target users that is compatible with the target user, the promotion information corresponding to the target page is presented during the presentation of the target page. The promotional information is incorporated into the content of the target page using the method described in any one of claims 1 to 10.

12. A device for processing promotional information, characterized in that, The device includes: The first determination module is used to determine the target person from at least one person described in the content of the article; The analysis module is used to analyze the content of the article that corresponds to the target person to obtain the characteristics of the target person. The characteristics include one or more of the following: personality traits, behavioral traits, language traits, or relationship traits with other people described in the article. The second determining module is used to determine the target users of the objects to be promoted, and the set of objects to be promoted that are compatible with the target users; The selection module is used to select, from the set of objects to be promoted, objects that are related to the content of the target page corresponding to the article; The generation module is used to determine information that matches the target user from the basic information of the object to be promoted, as the personalized information corresponding to the target user; Based on the language template corresponding to the target person's characteristics and the personalized information, corresponding promotional information is generated. The language template includes semantic rules, words, phrases, and sentence structures commonly used by the target person. The fusion module is used to obtain a fusion template corresponding to the target page. The fusion template describes the layout position of the fused promotional information and the content of the target page on the target page. The promotional information and the content of the target page are added to the corresponding layout positions on the target page to fuse the promotional information into the content of the target page. The promotional information is used to be presented during the process of the target user's terminal displaying the target page. The update module is used to update the language templates corresponding to each target promoter based on the comments of the target users on the article. The target promoter is a target person who meets the promotion conditions. The language template is used to integrate the characteristics of the selected target to be promoted and the target person to generate the promotion information.

13. The apparatus as claimed in claim 12, characterized in that, The device further includes: The promotional user identification module is used to obtain click data of the target user for each of the promotional information when there are at least two promotional information and each promotional information is generated based on the characteristics of different target users; and based on the click data, determine the target user corresponding to at least one promotional information whose click data meets the promotion conditions as the first promotional user. The settings module is used to set the priority of the promotion information corresponding to the first promoter to be higher than the priority of the promotion information of the second promoter.

14. The apparatus as claimed in claim 12, characterized in that, The device further includes: The update module is used to, when there are at least two promotional messages and each promotional message is generated based on the characteristics of different target individuals, acquire the click data of the target user for each promotional message; based on the click data, determine the target individual corresponding to at least one promotional message whose click data meets the promotion conditions as the target promotional individual; and acquire the comment content of the target user for the article.

15. An electronic device, characterized in that, include: Memory, used to store executable instructions; A processor, when executing executable instructions stored in the memory, implements the method for processing promotional information as described in any one of claims 1 to 11.

16. A computer-readable storage medium, characterized in that, The device stores executable instructions, which, when executed, are used to implement the method for processing promotional information as described in any one of claims 1 to 11.

17. A computer program product comprising: A computer-executable instruction, characterized in that, when executed by a processor, the computer-executable instruction implements the method described in any one of claims 1 to 11.

Images