An information interaction method, device and system for advertisement recommendation

Abstract

The application provides an information interaction method, device and system for advertisement recommendation, so that a user can change from a passive receiver to a creator and sharer of advertisement content in the form of secondary creation of the content of the advertisement, and the enthusiasm of the user to participate in the advertisement activity is improved, so that the recommendation effect of the advertisement is improved. The method comprises the following steps: sending an advertisement and creation invitation information corresponding to the advertisement to each terminal, the creation invitation information being used to instruct the user to create an advertisement copy which is the same as the advertisement in terms of the purpose of publicity; obtaining each advertisement copy made by the user of the terminal according to the creation invitation information; screening target advertisement copies according to the content quality of each advertisement copy and recommending the target advertisement copies to the user; receiving feedback data corresponding to each target advertisement copy collected by the terminal, determining a traffic allocation ratio corresponding to each target advertisement copy according to the data, and recommending each target advertisement copy to the user of each terminal again according to the traffic allocation ratio.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an information interaction method, device, and system for advertisement recommendation. Background Art

[0002] When users use Internet products or services, the information interaction system for advertising recommendation can intelligently select and push advertising content that is most likely to arouse the user's interest.

[0003] In the information interaction system, the advertising content produced by the advertising business party is forcibly pushed by the server in the system to the terminal held by the consumer for loading and display, in order to achieve ad recommendations. During the advertising delivery cycle, the server collects feedback data such as the number of ad impressions and clicks for data analysis, and attempts to replace the materials of ads that do not meet the expected revenue conversion rate one by one, in order to achieve multiple re-production of ads and ad recommendations. From ad production to ad delivery, the server does not take into account factors such as the ad design that consumers are interested in and the demand for ad content. It simply relies on mandatory triggered feedback data as analysis and evaluation criteria for ad recommendations. Due to the incomplete feedback data, the trial and error cost of the system for ad recommendations is increased, and it may even fail to achieve the expected revenue conversion rate at the end of the ad delivery.

[0004] Therefore, improving the effectiveness of the system in recommending advertisements has become an urgent problem to be solved. Summary of the Invention

[0005] The present application provides an information interaction method, device and system for advertising recommendation, which enables users to transform from passive recipients to creators and sharers of advertising content by re-creating the content of the advertisement, thereby increasing users' enthusiasm for participating in advertising activities and thus improving the recommendation effect of the advertisement.

[0006] In a first aspect, the present application provides an information interaction method for advertisement recommendation, comprising:

[0007] Sending an advertisement and creation invitation information corresponding to the advertisement to at least one terminal, where the creation invitation information is used to instruct a user to create an advertisement copy having the same promotional purpose as the advertisement;

[0008] Obtaining at least one advertisement copy prepared by a user of at least one terminal according to the creation invitation information;

[0009] selecting at least one target advertisement copy from at least one advertisement copy based on the content quality of each advertisement copy, and recommending the target advertisement copy to a user of at least one terminal;

[0010] receiving feedback data corresponding to each target advertisement copy collected from the at least one terminal, the feedback data being used to reflect a degree of popularity of the target advertisement copy with users;

[0011] determining a traffic distribution ratio corresponding to each target advertisement copy according to each feedback data;

[0012] recommending each target advertisement copy to users of the at least one terminal again according to the traffic distribution ratio corresponding to each target advertisement copy.

[0013] In a possible design, the at least one target advertisement copy is selected from the at least one advertisement copy based on content quality of each advertisement copy, and is recommended to users of the at least one terminal, including:

[0014] obtaining initial scores of each advertisement copy in terms of various content quality indexes, the content quality indexes being used to evaluate quality of the advertisement copy;

[0015] determining a final score of each advertisement copy according to each initial score of the advertisement copy, the final score being a weighted sum of the initial scores of the advertisement copy;

[0016] sequentially ranking the at least one advertisement copy in descending order according to the final score of each advertisement copy;

[0017] determining each advertisement copy in a first quantity of advertisement copies at a front of the sequence as a target advertisement copy;

[0018] determining a traffic distribution ratio corresponding to each target advertisement copy according to the final score of each target advertisement copy;

[0019] recommending each target advertisement copy to users of the at least one terminal according to the traffic distribution ratio corresponding to each target advertisement copy.

[0020] In a possible design, the final score of each advertisement copy is determined according to each initial score of the advertisement copy, including:

[0021] determining a first weight ratio set corresponding to each first target weight factor, the first weight ratio set including a weight ratio of the first target weight factor and each first weight factor, the first weight factor being a content quality index, and the first target weight factor being one of the first weight factors corresponding to the advertisement copy;

[0022] determining a geometric mean of each weight ratio corresponding to each first target weight factor based on the first weight ratio set of each first target weight factor;

[0023] normalize the geometric mean value corresponding to each first target weight factor to obtain a first normalized weight value corresponding to each first target weight factor;

[0024] weight the respective initial score of each advertisement copy using the respective first normalized weight value, and take the sum of the weighted initial scores as the final score of the advertisement copy.

[0025] In a possible design, the feedback data includes one or more initial times of the following types:

[0026] request times, return times, exposure times, click times, loading times, purchase times, and sharing times;

[0027] The traffic distribution proportion corresponding to each target advertisement copy is determined according to each feedback data, and includes:

[0028] determining a second weight ratio value set corresponding to each second target weight factor, the second weight ratio value set including a weight ratio value of the second target weight factor and each second weight factor, the second weight factor being of a times type, and the second target weight factor being one of the second weight factors corresponding to the advertisement copy;

[0029] determining a geometric mean value of each weight ratio value corresponding to each second target weight factor based on the second weight ratio value set of the second target weight factor;

[0030] normalizing the geometric mean value corresponding to each second target weight factor to obtain a second normalized weight value corresponding to each second target weight factor;

[0031] weighting the initial time of the corresponding times type in each feedback data using the respective second normalized weight value, and taking the sum of the weighted initial times as the final time of the corresponding target advertisement copy;

[0032] The traffic distribution proportion corresponding to each target advertisement copy is determined according to the final time of each target advertisement copy.

[0033] In a possible design, the method further includes:

[0034] obtaining a word library for storing sensitive words, the word library storing and managing the sensitive words using a dictionary tree structure;

[0035] obtaining the text to be audited of the advertisement;

[0036] matching the characters of the text to be audited with the first characters of the word library one by one, the first characters corresponding to the next level nodes of the root nodes of the dictionary tree structure;

[0037] Determine the character in the text to be audited that matches the first character of the word library as a sensitive character;

[0038] Identify the to-be-matched word in the text to be audited that starts with each sensitive character, and cache the to-be-matched word in the text to be audited;

[0039] Match each cached to-be-matched word with the sensitive word of the word library;

[0040] If the text to be audited does not contain a word that matches the sensitive word of the word library, send an advertisement and a creation invitation information corresponding to the advertisement to at least one terminal.

[0041] In a second aspect, the application provides an information interaction method for advertisement recommendation, the method comprising:

[0042] Receiving and displaying an advertisement and creation invitation information corresponding to the advertisement to a user, the creation invitation information being used to instruct the user to create an advertisement copy identical to the advertisement in terms of promotional purpose;

[0043] In response to an operation of the user clicking the advertisement, displaying a login page to the user, the login page being used for the user to log in to a content creation service unit through an account, so that the user makes the advertisement copy according to the creation invitation information through the content creation service unit.

[0044] In a feasible design, the method further comprises:

[0045] Receiving and displaying at least one target advertisement copy recommended from an advertisement content service unit through a first media platform;

[0046] In response to an operation of the user clicking a sharing control corresponding to the target advertisement copy, displaying at least one second media platform that can be shared to the user, and in response to an operation of the user clicking the second media platform, sending the target advertisement copy or a network resource link of the target advertisement copy to the second media platform clicked by the user.

[0047] In a third aspect, the application provides an information interaction device for advertisement recommendation, comprising an advertisement content service unit and a content creation service unit:

[0048] The advertisement content service unit is configured to send an advertisement and creation invitation information corresponding to the advertisement to at least one terminal, the creation invitation information being used to instruct a user to create an advertisement copy identical to the advertisement in terms of promotional purpose;

[0049] The content creation service unit is configured to make at least one advertisement copy according to the creation invitation information by a user of the at least one terminal;

[0050] The advertisement content service unit is further configured to receive at least one advertisement copy from the content creation service unit;

[0051] The advertisement content service unit is further configured to select at least one target advertisement copy from the at least one advertisement copy based on content quality of each advertisement copy, and recommend the at least one target advertisement copy to a user of the at least one terminal;

[0052] The advertisement content service unit is further configured to receive feedback data corresponding to each target advertisement copy collected from the at least one terminal, the feedback data reflecting a degree of popularity of the target advertisement copy among users;

[0053] The advertisement content service unit is further configured to determine a traffic allocation ratio corresponding to each target advertisement copy according to each feedback data;

[0054] The advertisement content service unit is further configured to recommend each target advertisement copy to the user of the at least one terminal again according to the traffic allocation ratio corresponding to each target advertisement copy.

[0055] In a fourth aspect, the present application provides an information interaction device for advertisement recommendation, comprising an information transmission unit, an instruction processing unit and a display unit;

[0056] The information transmission unit is configured to receive an advertisement and creation invitation information corresponding to the advertisement, the creation invitation information being used to instruct a user to create an advertisement copy with the same promotion purpose as the advertisement;

[0057] The display unit is configured to display the advertisement and the creation invitation information;

[0058] The instruction processing unit is configured to, in response to an operation of the user clicking on the advertisement, display a login page to the user through the display unit, the login page being used for the user to log in to a content creation service unit through an account, so that the user makes the advertisement copy according to the creation invitation information through the content creation service unit;

[0059] The display unit is further configured to display a page for the user to remotely operate the content creation service unit.

[0060] In a fifth aspect, the present application provides an information interaction system for advertisement recommendation, comprising a server and a terminal, the server comprising an advertisement content service unit and a content creation service unit, and the terminal comprising an information transmission unit, a display unit and an instruction processing unit;

[0061] The advertisement content service unit is configured to send an advertisement and creation invitation information corresponding to the advertisement to at least one terminal, the creation invitation information being used to instruct a user to create an advertisement copy with the same promotion purpose as the advertisement;

[0062] The information transmission unit is configured to receive the advertisement and the creation invitation information;

[0063] The display unit is configured to display the advertisement and the creation invitation information;

[0064] The instruction processing unit is configured to, in response to the operation of the user clicking the advertisement, display a login page to the user through the display unit, the login page being used for the user to log in the content creation service unit by an account;

[0065] The display unit is configured to display a page for the user to remotely operate the content creation service unit;

[0066] The content creation service unit is configured to, for the user of the at least one terminal, create at least one advertisement copy according to the creation invitation information;

[0067] The advertisement content service unit is further configured to receive the at least one advertisement copy from the content creation service unit;

[0068] The advertisement content service unit is further configured to select at least one target advertisement copy from the at least one advertisement copy based on a content quality of each advertisement copy, and recommend the at least one target advertisement copy to the user of the at least one terminal;

[0069] The advertisement content service unit is further configured to receive feedback data corresponding to each target advertisement copy collected from the at least one terminal, the feedback data being used to reflect a degree of popularity of the target advertisement copy among the users;

[0070] The advertisement content service unit is further configured to determine a traffic allocation ratio corresponding to each target advertisement copy according to each feedback data;

[0071] The advertisement content service unit is further configured to, according to the traffic allocation ratio corresponding to each target advertisement copy, recommend each target advertisement copy to the user of the at least one terminal again.

[0072] In a sixth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the information interaction method for advertisement recommendation in the first aspect.

[0073] In a seventh aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the information interaction method for advertisement recommendation in the second aspect.

[0074] In an eighth aspect, the present application provides an electronic device, which comprises a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, and the processor implements the information interaction method for advertisement recommendation in the first aspect when executing the computer program.

[0075] In a ninth aspect, the present application provides an electronic device, which comprises a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, and the processor implements the information interaction method for advertisement recommendation in the second aspect when executing the computer program.

[0076] The information interaction method for advertisement recommendation provided by the application allows users to secondarily create advertisements by sending creation invitation information to terminals, and generates diversified advertisement copies. The method can optimize the recommendation effect of advertisements by improving the diversity of creativity and user participation, and improve the conversion rate of advertising revenue. In order to better utilize limited traffic resources for advertising promotion, it is necessary to control the number of advertisement copies recommended to users, so based on the content quality of each advertisement copy, target advertisement copies with better content quality can be screened out for recommendation. In order to optimize the effect of recommending target advertisement copies, that is, the higher the popularity of users, the more traffic cost is worth spending to recommend to more users, according to the feedback data collected by the terminal for reflecting the degree of popularity of target advertisement copies among users, the application determines the traffic allocation ratio corresponding to each target advertisement copy through hierarchical analysis. According to the traffic allocation ratio, the target advertisement copies are recommended to users, which can optimize the advertising strategy and realize effective personalized recommendation. Since the advertisement copy represents the design of the advertisement content preferred by the user, recommending the advertisement copy can increase the user's interest in browsing the advertisement content, thereby improving the click volume, sharing volume and other data, so that the media platform further improves the exposure of the advertisement copy and the original version of the advertisement with the same advertising purpose, thereby further improving the recommendation effect of the advertisement.

[0077] In summary, the application realizes efficient dissemination of advertising information, improves conversion rate, creates higher commercial value, and reduces trial and error and operating costs of the advertising platform by actively involving users (i.e. consumers) in creating advertising content. BRIEF DESCRIPTION OF DRAWINGS

[0078] In order to more clearly illustrate the technical solutions of the application, the following will briefly introduce the drawings needed in the embodiments. Obviously, for those skilled in the art, other drawings can also be obtained based on these drawings without creative labor.

[0079] Figure 1 is an example of an information interaction system structure for advertisement recommendation provided by an exemplary embodiment of the application;

[0080] Figure 2 is an example of an information interaction method for advertisement recommendation provided by an exemplary embodiment of the application;

[0081] Figure 3 is an example of a dictionary tree structure provided by an exemplary embodiment of the application;

[0082] Figure 4 is another example of an information interaction method for advertisement recommendation provided by an exemplary embodiment of the application. DETAILED DESCRIPTION

[0083] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0084] In order to enhance the participation of consumers in the process from ad content creation to ad delivery, and thus improve the effectiveness of the system in recommending ads, this application provides the following Figure 1 The information interaction system for advertising recommendation shown in FIG. Figure 1 As shown, the system includes a server 100 and a terminal 200. The server 100 includes an advertising content service unit 101 and a content creation service unit 102, while the terminal 200 includes an information transmission unit 201, an instruction processing unit 202, and a display unit 203. The terminal 200 is installed with an application for a first media platform designated by an advertising operator for advertising delivery. The following describes the functions of each component of the system:

[0085] Advertising content service unit 101 uses internet information technology to upload, manage, and publish original advertisements and the materials used to create them (images, audio). Advertising operators or advertisers can also configure advertisement delivery information and corresponding creation invitations. This information includes the delivery cycle, target audience group tags, delivery time period, delivery channels, and touchpoints. The creation invitation instructs users to create a copy of the advertisement with the same promotional purpose. "Users creating a copy of the advertisement with the same promotional purpose" can be referred to as secondary creation of the advertisement. The content of the copy and the original advertisement, such as the product or event, is identical; only the presentation format differs.

[0086] Exemplarily, publishing an advertisement includes recommending the advertisement to the user of terminal 200. Specifically, by continuously analyzing the content quality of each advertisement copy, operational feedback data, etc., the traffic ratio of each advertisement copy is determined based on the analysis results, and each advertisement copy is pushed to terminal 200 according to the traffic ratio, thereby achieving advertisement recommendation.

[0087] The feedback data includes log data such as the number of requests, returns, exposures, clicks, loads, purchases, and shares of the advertisement copy on the terminal 200 .

[0088] Exemplarily, the creation invitation information includes information indicating that the user makes secondary creation on the advertisement and requirements of the secondary creation, wherein the requirements of the secondary creation include theme content, brand logo, main background of the advertisement, text, and submission time, etc.

[0089] The content creation service unit 102 mainly uses computer information technology for editing video and audio to realize that the user of the terminal 200 makes secondary creation on the advertisement to be put.

[0090] Exemplarily, the secondary creation on the advertisement includes making an advertisement theme, writing an advertisement script, designing visual elements, selecting audio or video materials, etc. to create an eye-catching advertisement copy.

[0091] In a feasible design, the server further includes a content auditing unit 103. The content auditing unit 103 is configured to realize automatic and intelligent auditing on whether the content of the advertisement involves sensitive topics, contains misleading information, or meets legal and regulatory requirements, for example, automatically identifying and filtering malicious advertisements, junk information, bad videos, audio and text content, etc.

[0092] Exemplarily, the content auditing unit 103 also supports manual auditing to optimize the auditing results.

[0093] The information transmission unit 201 is configured to realize transmission of information. For example, receiving the advertisement issued by the advertisement content service unit 101, the creation invitation information corresponding to the advertisement, or uploading feedback data of the advertisement copy to the advertisement content service unit 101.

[0094] The instruction processing unit 202 is configured to realize that the user remotely logs in the content creation service unit 102 and remotely creates the advertisement copy on the content creation service unit 102.

[0095] The display unit 203 is configured to display the advertisement, the creation invitation information, a login page, and a page for remotely operating the content creation service unit 102.

[0096] In the application process of the system, the advertisement operator or the advertisement owner uploads the advertisement and the production materials of the advertisement through the advertisement content service unit 101, and then configures the putting period of the advertisement, the user group label of the advertisement, the putting time period, the putting channel and touch point, and the creation invitation information corresponding to the advertisement on the advertisement content service unit 101. After the advertisement passes the auditing of the content auditing unit 103, the advertisement content service unit 101 issues the advertisement and the corresponding creation invitation information to the terminal 200.

[0097] After the terminal 200 obtains the creation invitation information of the advertisement through the information transmission unit 201, the advertisement and the corresponding creation invitation information are displayed on the first media platform. After the user sees the creation invitation information on the first media platform, the user can remotely log in the content creation service unit 102 through the instruction processing unit 202, integrate existing advertising materials, recombine the script, adjust the visual design of the video, or change the overall form of the advertisement according to the requirements in the creation invitation information, to meet specific advertising needs and targets, complete secondary creation, and make a copy of the advertisement. After the copy of the advertisement passes the audit of the content audit unit 103, the content creation service unit 102 synchronizes the copy of the advertisement to the advertisement content service unit 101.

[0098] The advertisement content service unit 101 obtains the score of the copy of the advertisement in terms of content quality, and determines the traffic distribution ratio of the copy of the advertisement according to the score of the copy of the advertisement. The copy of the advertisement is pushed to the terminal 200 according to the traffic distribution ratio of each copy of the advertisement. The score factors (i.e., content quality indicators) in terms of content quality include but are not limited to: visual appeal of the advertisement, clarity of the script expression, matching degree of the target audience, innovation of the advertisement, uniqueness of the advertisement, practicality of the advertisement, and interestingness of the advertisement.

[0099] The terminal 200 collects feedback data of the copy of the advertisement operating on the first media platform, and transmits the feedback data back to the advertisement content service unit 101 through the information transmission unit 201. The advertisement content service unit 101 continuously adjusts the traffic distribution ratio of recommending each copy of the advertisement by analyzing and calculating the feedback data.

[0100] Exemplarily, the terminal 200 is installed with an application program of a second media platform (i.e., a self-media platform) (the second media platform can be a personal blog, a social media application, a video sharing platform, etc.). After the user of the terminal 200 receives the copy of the advertisement that passes the audit, the user can send the copy of the advertisement or a resource link of the copy of the advertisement to the second media platform operated by the user through a sharing or downloading function, to expand the exposure and influence of the advertisement. The terminal 200 can collect feedback data of the copy of the advertisement on the self-media platform, and send the feedback data to the advertisement content service unit 101 through the information transmission unit 201, so that the advertisement content service unit 101 tracks the effects of the sharing by analyzing the feedback data.

[0101] It can be seen that the above system realizes secondary creation of the advertisement by the user, so that the user changes from a passive receiver to a creator and sharer (which can be shared across the self-media platform) of the advertisement. In addition to providing a way for the user to express emotions and sentiments, the scheme also enables the user to actively provide preferences, interests, and recent concerns (industry, commodity, literature, and art) to the first media platform, the second media platform, and the system while making the advertisement copy. Therefore, the scheme can reduce the network, storage, computing power, and external data costs required for analyzing user behavior logs collected through the terminal. Since the advertisement copy is created by the user based on the content needs and design needs of consumers, it is beneficial to improve the promotion effect of the advertised goods among consumers and thus improve the revenue conversion rate of the advertisement.

[0102] Moreover, under the situation that user privacy protection and information supervision systems are becoming increasingly stringent, the scheme enables the first media platform and the second media platform to avoid legal risks arising from actively spying on user industries and privacy. In addition, the user shares the created advertisement as content on the self-media platform operated by the user, breaking the advertisement channel threshold between media platforms.

[0103] Based on the above information interaction system for advertisement recommendation, as shown in Figure 2 the present application provides an information interaction method for advertisement recommendation, applied to a server 100, comprising:

[0104] S310, sending an advertisement and corresponding creation invitation information to at least one terminal.

[0105] The creation invitation information is used to instruct the user to create an advertisement copy with the same promotion purpose as the advertisement. For specific content of the creation invitation information, please refer to the description in the above system embodiment, which will not be repeated here.

[0106] Before that, the advertisement operator or the advertisement owner needs to upload the prepared advertisement and the materials required for making the advertisement to the advertisement content service unit 101, and then configure the delivery information of the advertisement and the requirements for secondary creation. After the advertisement operator or the advertisement owner performs the operation of publishing the advertisement, the advertisement content service unit 101 generates the creation invitation information according to the requirements for secondary creation, and then sends the advertisement and the corresponding creation invitation information to at least one terminal.

[0107] In order to ensure the legality and compliance of the advertising content, the advertising content service unit 101 audits the content of the advertisement by the content auditing unit 103 before sending the advertisement to the terminal. Among them, the text information and audio content of the advertisement can be audited by using natural language processing technology, and the video content of the advertisement can also be audited by using image recognition technology through frame-by-frame screenshot method to improve the efficiency of auditing. Further, artificial auditing method can also be combined to improve the effect of auditing.

[0108] With the changes in society and cultural concepts, the diversification and complexity of Internet content are also increasing. This includes the continuous expansion and updating of the vocabulary range related to sensitive topics. Traditional politically sensitive vocabulary is still an important object of auditing, while emerging sensitive topics such as sex, violence, discrimination, etc. have also gradually become the focus of attention. Therefore, the content auditing unit of the system needs to be able to update and adapt to these changes in a timely manner to ensure that the review of advertising content meets the requirements of laws and regulations and platform policies, while protecting users from harmful content.

[0109] The text slicing method adopted in the present application can make the system more flexible and efficient in processing different types of sensitive vocabulary and content. Through reasonable segmentation and classification, the system can more quickly identify text slice segments that may have risks, and then perform necessary further review or processing. This method helps to improve the work efficiency of the auditing team, reduces the risk of misjudgment and oversight, and also enhances the system's ability to cope with complex content. The following describes the method for auditing the text of the advertisement, which includes the text of the advertisement and the text recognized in the audio of the advertisement:

[0110] S311, a word library for storing sensitive words is obtained, and the word library stores and manages the sensitive words by using a dictionary tree structure;

[0111] S312, the text to be audited of the advertisement is obtained;

[0112] S313, the characters of the text to be audited are matched with the first characters of the word library one by one, and the first characters correspond to the next level nodes of the root nodes of the dictionary tree structure;

[0113] S314, the characters in the text to be audited that match the first characters of the word library are determined as sensitive characters;

[0114] S315, the matching words in the text to be audited, which have each sensitive character as the first character, are identified, and the matching words in the text to be audited are cached;

[0115] S316, each matching word cached is matched with the sensitive words of the word library;

[0116] S317, if the text to be audited does not contain a word matching the sensitive word in the word library, send an advertisement and a creation invitation information corresponding to the advertisement to at least one terminal.

[0117] It should be understood that the two characters in the above examples match means that the two characters are consistent, and the two words match means that the two words are consistent.

[0118] The dictionary tree (Trie tree) is a tree structure that can efficiently store and search a set of strings, especially suitable for fast matching of large-scale sensitive words. Using the Trie tree, millions of sensitive words can be stored and managed according to the structure of the dictionary tree. For example Figure 3 As shown, the next level nodes of the root node of the dictionary tree structure correspond to the first characters a and b of the sensitive words, and e, f, and v are the tail characters of the sensitive words with a as the first character, and k is the tail character of the sensitive word with b as the first character. It can be seen that the sensitive words are [ace], [agf], [agkv], and [bk], the maximum length of the sensitive word library node is 4, and the minimum length is 2.

[0119] Taking the storage structure of the sensitive word library dictionary tree of Figure 3 , when the characters a or b appear in the advertisement text, the word recognition is performed with a or b as the first character, and the length of the recognized word to be matched is determined. Then, the text segments of 1, 2, or 3 character lengths are cached with a or b as the first character, and the character length of the cached word to be matched may be 2, 3, or 4.

[0120] The above examples will be explained in combination with the system shown in Figure 1

[0121] The content auditing unit 103 matches the cached text segments with the sensitive words in the sensitive word library by performing S311-S316. If the text to be audited does not contain a word matching the sensitive word in the word library, the content auditing unit 103 can update the status identifier of the advertisement to indicate that the advertisement passes the audit, and send the updated advertisement to the advertisement content service unit 101. The advertisement content service unit 101 sends the advertisement and the creation invitation information corresponding to the advertisement to at least one terminal after identifying the update of the status identifier of the advertisement. Each terminal 200 receives the advertisement and the creation invitation information corresponding to the advertisement through the information transmission unit 201, and displays the advertisement through the first media platform to complete the advertisement publishing.

[0122] ​Exemplarily, if the text to be audited contains words matching the sensitive words in the word library, the content auditing unit 103 marks the words in the text to be audited matching the sensitive words in the word library, updates the original text to be audited, and then sends it to the advertising content service unit 101. The advertising content service unit 101 identifies that the text to be audited is updated, and displays the marked text to be audited to the advertising business party.

[0123] The above example takes into account that the text of the advertisement has the characteristics of being short and concise. The maximum length of the node of the sensitive word library with the dictionary tree structure does not need to be long, and can meet the length requirement of the words contained in the text of the advertisement. Therefore, using the sensitive word library with a simple dictionary tree structure can segment the text of the advertisement, thereby combining the text segments of multiple lengths with character lengths of [2, 4] and storing them in the cache of the memory. In this way, it can ensure that each text segment containing sensitive words is audited and processed separately, effectively improving the processing efficiency of the advertising text content auditing system. And it can avoid marking the entire text as sensitive, ensuring detailed analysis of each text segment, thereby reducing false positives and omissions, improving the accuracy of text auditing, and thus improving the overall auditing quality.

[0124] S320, obtaining at least one advertisement copy made by a user of at least one terminal according to the creation invitation information.

[0125] Exemplarily, after obtaining the advertisement copy, the advertisement copy is audited.

[0126] For example, the system as shown in Figure 1 The above example is described with reference to the system as shown in

[0127] After the user of the terminal learns the creation invitation information through the first media platform, the user can click the advertisement to jump to the content creation service unit 102 through single sign-on. After the user logs in to the content creation service unit 102, the user uses the editing tool to create the content according to the requirements of the secondary creation and the materials of the advertisement, and saves the created content to generate a corresponding advertisement copy. The content creation service unit 102 synchronizes the advertisement copy to the advertising content service unit 101. The advertising content service unit 101 sends the advertisement copy to the content auditing unit 103. The content auditing unit 103 determines whether the advertisement copy passes the audit, updates the state identifier of the advertisement copy, and sends the updated advertisement copy to the advertising content service unit 101. For other descriptions of auditing the advertisement copy, refer to the description of auditing the advertising content in S310 above, which will not be repeated here.

[0128] S330, selecting at least one target advertisement copy from the at least one advertisement copy based on the content quality of each advertisement copy, and recommending it to the user of at least one terminal.

[0129] In one possible design, the content quality based on each ad replica is implemented by selecting at least one target ad replica from the at least one ad replica, and recommending the at least one target ad replica to users of at least one terminal in the following manner:

[0130] S331, obtaining initial scores of each ad replica in terms of respective content quality indicators for evaluating quality of the ad replica;

[0131] S332, determining a final score of each ad replica according to respective initial scores of the ad replica, the final score being a weighted sum of the initial scores of the ad replica;

[0132] S333, ranking the at least one ad replica in descending order according to the final score of each ad replica;

[0133] S334, determining each ad replica in a first number of ad replicas in the order as a target ad replica;

[0134] S335, determining a traffic distribution ratio corresponding to each target ad replica according to the final score of each target ad replica;

[0135] S336, recommending each target ad replica to users of at least one terminal according to the traffic distribution ratio corresponding to each target ad replica.

[0136] The initial score of the ad replica can be scored by a human according to the content quality indicators, or scored intelligently by using an automatic algorithm.

[0137] The first number can be set according to actual needs, for example, it can be 5.

[0138] Exemplarily, the traffic distribution ratio corresponding to each target ad replica is determined according to the final score of each target ad replica in the following manner:

[0139] Determining a sum value of the final scores of all target ad replicas;

[0140] Calculating a ratio of the final score of each target ad replica to the sum value;

[0141] Determining the ratio corresponding to each target ad replica as the traffic distribution ratio corresponding to the target ad replica.

[0142] Exemplarily, each target ad replica is recommended to users of at least one terminal according to the traffic distribution ratio corresponding to each target ad replica in the following manner:

[0143] Determining the traffic distribution ratio corresponding to each target ad replica as a ratio of a number of users corresponding to each target ad replica.

[0144] push each target advertisement copy to at least one terminal according to the proportion of the corresponding number of users.

[0145] In the above example, the advertisement content service unit 101 obtains the final score of each advertisement copy in content quality by weighting each initial score of each advertisement copy. This way obtains the score by comprehensively considering various content quality score factors, so that the score of the advertisement copy in content quality is more comprehensive and accurate. Based on this, the first number of target advertisement copies with better content quality can be accurately screened out. The better the content quality, the more traffic is worth spending to push to more users, therefore, the advertisement content service unit 101 needs to determine the corresponding traffic allocation proportion according to the final score of the target advertisement copy, realizes the ranking and recommendation of the target advertisement copy according to the content quality, and thus improves the conversion rate of the advertisement.

[0146] In a feasible design, the final score of each advertisement copy is determined according to each initial score of the advertisement copy in the following way:

[0147] determining a first weight ratio set corresponding to each first target weight factor, the first weight ratio set including a weight ratio of the first target weight factor and each first weight factor, the first weight factor being a content quality indicator, and the first target weight factor being one of the first weight factors corresponding to the advertisement copy;

[0148] determining a geometric mean value of each weight ratio corresponding to each first target weight factor based on the first weight ratio set of each first target weight factor;

[0149] normalizing the geometric mean value corresponding to each first target weight factor to obtain a first standardized weight value corresponding to each first target weight factor;

[0150] weighting the corresponding initial score of each advertisement copy using each first standardized weight value, and taking the sum of the weighted initial scores as the final score of the advertisement copy.

[0151] The following takes the first weight factors corresponding to the advertisement copy as innovation, uniqueness, interestingness, and heat as an example to illustrate the example of "determining the final score of the advertisement copy according to each initial score of the advertisement copy":

[0152] (1) determining the first weight ratio set corresponding to the first target weight factor-innovation, the first target weight factor-uniqueness, the first target weight factor-interestingness, and the first target weight factor-heat, respectively.

[0153] The values of the first weight ratio set corresponding to the innovation are shown in Table 1 as follows:

[0154] Table One

[0155]

[0156] From Table One, it can be seen that the innovation of the advertisement is not as important as the uniqueness, the ratio of the weight of the innovation to the weight of the uniqueness is 1 / 3; the innovation of the advertisement is not as important as the interestingness, the ratio of the weight of the innovation to the weight of the interestingness is 1 / 5; the innovation of the advertisement is as important as the heat, the ratio of the weight of the innovation to the weight of the heat is 1.

[0157] The respective ratios of the first weight ratio group corresponding to the uniqueness are shown in Table Two as follows:

[0158] Table Two

[0159]

[0160] From Table Two, it can be seen that the uniqueness of the advertisement is more important than the innovation, the ratio of the weight of the uniqueness to the weight of the innovation is 3; the uniqueness of the advertisement is as important as the interestingness, the ratio of the weight of the uniqueness to the weight of the interestingness is 1; the uniqueness of the advertisement is more important than the heat, the ratio of the weight of the uniqueness to the weight of the heat is 5.

[0161] The respective ratios of the first weight ratio group corresponding to the interestingness are shown in Table Three as follows:

[0162] Table Three

[0163]

[0164] From Table Three, it can be seen that the interestingness of the advertisement is more important than the innovation, the ratio of the weight of the interestingness to the weight of the innovation is 5; the interestingness of the advertisement is as important as the uniqueness, the ratio of the weight of the interestingness to the weight of the uniqueness is 1; the interestingness of the advertisement is more important than the heat, the ratio of the weight of the interestingness to the weight of the heat is 5.

[0165] The respective ratios of the first weight ratio group corresponding to the heat are shown in Table Four as follows:

[0166] Table Four

[0167]

[0168] From Table Four, it can be seen that the heat of the advertisement is as important as the innovation, the ratio of the weight of the heat to the weight of the innovation is 1; the heat of the advertisement is not as important as the uniqueness, the ratio of the weight of the heat to the weight of the uniqueness is 1 / 5; the heat of the advertisement is not as important as the interestingness, the ratio of the weight of the heat to the weight of the interestingness is 1 / 5.

[0169] In summary, the weight ratio group corresponding to the innovativeness is [1, 1 / 3, 1 / 5, 1], the weight ratio group corresponding to the distinctiveness is [3, 1, 1, 5], the weight ratio group corresponding to the interestingness is [5, 1, 1, 5], and the weight ratio group corresponding to the hotness is [1, 1 / 5, 1 / 5, 1].

[0170] It should be noted that the ratio between the weight of the first target weight factor and the weight of other first weight factors can be set according to actual needs, and the present application does not limit this. For example, as shown in Tables I to IV, when the present application sets the ratio between the weight of the first target weight factor and the weight of other first weight factors, the principle of independent comparison between two first weight factors without the influence of other first weight factors is followed. This setting method helps to ensure that the comparison between each weight factor is objective and fair, improves the objectivity and accuracy of the evaluation results, avoids the deviation of the comparison between other weight factors due to the interference of a certain weight factor, and avoids the imbalance of the entire evaluation system due to the particularity of a certain weight factor. Taking Table I as an example, the ratio between the weight of the innovativeness and the weight of the distinctiveness is 1 / 3, and the ratio between the weight of the innovativeness and the weight of the hotness is 1 / 1. If the ratio between the weight of the distinctiveness and the weight of the hotness is set to be influenced by the innovativeness, the ratio between the weight of the distinctiveness and the weight of the hotness may be incorrectly set to 3 / 1. However, according to the principle followed by the present application, the ratio between the weight of the distinctiveness and the weight of the hotness should be set independently of the innovativeness, and therefore the ratio between the weight of the distinctiveness and the weight of the hotness in Table II can be set to 5 / 1.

[0171] (2) In order to make the influence of each data point more uniform and avoid misleading situations that may occur when there are extreme values in the data, the present application uses the geometric mean method to calculate the weight ratio in each weight ratio group, as shown in the following formulas (1) to (4). Formula (1) is used to calculate the geometric mean of the weight ratio group corresponding to the innovativeness, formula (2) is used to calculate the geometric mean of the weight ratio group corresponding to the distinctiveness, formula (3) is used to calculate the geometric mean of the weight ratio group corresponding to the interestingness, and formula (4) is used to calculate the geometric mean of the weight ratio group corresponding to the hotness:

[0172] , formula (1);

[0173] , formula (2);

[0174] , formula (3);

[0175] , formula (4);

[0176] (3) The geometric mean value corresponding to each first target weight factor is normalized, so that the geometric mean value of each first weight factor is scaled to the interval [0, 1], and each first weight factor is standardized to a unified scale, so that the comparison and analysis between different features and data sets are more fair, as shown in the following formulas (5)-(8). Among them, formula (5) is used to calculate the first standardized weight value corresponding to the innovation, formula (6) is used to calculate the first standardized weight value corresponding to the uniqueness, formula (7) is used to calculate the first standardized weight value corresponding to the interestingness, and formula (8) is used to calculate the first standardized weight value corresponding to the heat:

[0177] , formula (5);

[0178] , formula (6);

[0179] , formula (7);

[0180] , formula (8);

[0181] It can be seen that the geometric mean value after normalization is the first standardized weight value of the first weight factor.

[0182] (4) The final score of the advertisement copy is obtained by weighting the corresponding initial score of the advertisement copy using each first standardized weight value, as shown in the following formula (9):

[0183] , formula (9);

[0184] Among them, is the final score of the advertisement copy, , , , are the initial scores of the advertisement copy in terms of innovation, uniqueness, interestingness and heat, , , , are the first standardized weight values corresponding to innovation, uniqueness, interestingness and heat, , , , .

[0185] The following takes the final score of the advertisement copy 1 as an example to illustrate formula (9):

[0186] ;

[0187] Similarly, the final scores of the advertisement copy 2 and the advertisement copy 3 are shown in Table 5 as follows:

[0188] Table 5

[0189]

[0190] By analogy, when the number of the indicators of the content quality corresponding to the advertisement copy is m, the first weight ratio group includes m weight ratios, and correspondingly, the geometric mean corresponding to each content quality indicator is calculated m times, and the final first standardized weight value is , , , … The final score Y is calculated according to the following formula (10):

[0191] Formula (10);

[0192] Wherein, Y is the final score of the advertisement copy, , , , … are the initial scores of each content quality indicator of the advertisement copy, , , , … are the first standardized weight values corresponding to each content quality indicator.

[0193] In the above example, the weight ratio of the first target weight factor to each first weight factor can be designed according to actual needs, for example, according to the importance of each content quality indicator of the advertisement copy in terms of popularity among consumers, the weight ratio of the first target weight factor to each first weight factor is designed. Then, the weight value corresponding to the first target weight factor can be obtained according to each weight ratio corresponding to the first target weight factor, and the weight value obtained by this method is more in line with the needs of the system to promote the advertisement popular among users.

[0194] S340, receiving feedback data corresponding to each target advertisement copy collected from at least one terminal.

[0195] Wherein, the feedback data is used to reflect the degree of popularity of the target advertisement copy among users. For the description of the feedback data, please refer to the description in the system example, which will not be repeated here.

[0196] The following will be illustrated by combining the system shown in Figure 1 .

[0197] The terminal 200 collects log data of the target advertisement copy, such as the number of requests, the number of returns, the number of exposures, the number of clicks, the number of loads, the number of purchases, the number of shares, and the like, and sends the log data to the advertisement content service unit 101 through the information transmission unit 201.

[0198] S350, according to each feedback data, determine the flow distribution proportion corresponding to each target advertisement copy.

[0199] In the case where the feedback data includes initial number of one or more types of times:

[0200] The number of requests, the number of returns, the number of exposures, the number of clicks, the number of loads, the number of purchases, the number of shares;

[0201] The advertisement content service unit 101 can determine the flow distribution proportion corresponding to each target advertisement copy according to each feedback data by the following method:

[0202] Determine the second weight ratio group corresponding to each second target weight factor, the second weight ratio group including the weight ratio of the second target weight factor and each second weight factor, the second weight factor being the type of times, and the second target weight factor being one of the second weight factors corresponding to the advertisement copy;

[0203] Based on the second weight ratio group of each second target weight factor, determine the geometric mean of each weight ratio corresponding to each second target weight factor;

[0204] Normalize the geometric mean corresponding to each second target weight factor to obtain the second normalized weight value corresponding to each second target weight factor;

[0205] Use each second normalized weight value to weight the initial number of the corresponding type of times in each feedback data, and take the sum of the weighted initial number as the final number of the corresponding target advertisement copy;

[0206] According to the final number of each target advertisement copy, determine the flow distribution proportion corresponding to each target advertisement copy.

[0207] The following takes the feedback data including the number of clicks, the number of purchases and the number of shares as an example to illustrate the above example:

[0208] (1) Determine the second weight ratio group corresponding to the second target weight factor-clicks, the second target weight factor-purchases and the second target weight factor-shares, respectively.

[0209] For example, the importance of the number of clicks is higher than the importance of the number of purchases, and the importance of the number of purchases is higher than the importance of the number of shares in reflecting the popularity of the target advertisement copy to the user, the second weight ratio corresponding to each second target weight factor is determined.

[0210] The second weight ratio corresponding to the number of clicks, the number of purchases and the number of shares is shown in Table 6 as follows:

[0211] Table 6

[0212]

[0213] The weight ratio corresponding to the number of clicks is [1, 1 / 4, 1 / 3], the weight ratio corresponding to the number of purchases is [4, 1, 3 / 2], and the weight ratio corresponding to the number of shares is [3, 2 / 3, 1].

[0214] It should be noted that the ratio between the weight of the second target weight factor and the weight of other second weight factors can be set according to actual needs, and the present application does not limit this. For example, similar to the setting example of the weight ratio of the first weight factor, when the present application sets the ratio between the weight of the second target weight factor and the weight of other second weight factors, the principle of independent comparison between two second weight factors without the influence of other second weight factors is followed.

[0215] (2) The weight ratios in each weight ratio group are calculated by using the geometric mean method, as shown in the following formulas (11) to (13). Formula (11) is used to calculate the geometric mean of the weight ratio group corresponding to the number of clicks, formula (12) is used to calculate the geometric mean of the weight ratio group corresponding to the number of purchases, and formula (13) is used to calculate the geometric mean of the weight ratio group corresponding to the number of shares:

[0216] , formula (11);

[0217] , formula (12);

[0218] , formula (13);

[0219] (3) The geometric mean value corresponding to each second target weight factor is normalized, as shown in the following formulas (14) to (16). Formula (14) is used to calculate the second standardized weight value corresponding to the number of clicks, formula (15) is used to calculate the second standardized weight value corresponding to the number of purchases, and formula (16) is used to calculate the second standardized weight value corresponding to the number of shares:

[0220] , formula (14);

[0221] , equation (15);

[0222] , equation (16);

[0223] It can be seen that the geometric mean value after normalization is the second standardized weight value of the second weight factor.

[0224] (4) The final number of the target advertisement copy is obtained by weighting the corresponding initial number of the target advertisement copy using each second standardized weight value, as shown in the following equation (17):

[0225] , equation (17);

[0226] wherein, is the final number of the target advertisement copy, , , are the values (i.e., the initial numbers) corresponding to the click number, the purchase number, and the sharing number of the target advertisement copy, respectively, , , are the second standardized weight values corresponding to the click number, the purchase number, and the sharing number, respectively, , , .

[0227] By analogy, when the number of types of numbers included in the feedback data of the target advertisement copy is , the second weight ratio group includes weight ratios, and correspondingly, the geometric mean value of each type of number is calculated times, and the finally obtained second standardized weight values are , , … , the final number is calculated as shown in the following equation (18):

[0228] , equation (18);

[0229] (5) The traffic distribution proportion corresponding to each target advertisement copy is determined according to the final number of each target advertisement copy.

[0230] Exemplarily, the traffic distribution proportion corresponding to each target advertisement copy is determined according to the final number of each target advertisement copy in the following manner:

[0231] The sum value of the final numbers of all target advertisement copies is determined;

[0232] calculating a ratio of the final number of each target advertisement copy to the sum value;

[0233] determining the ratio corresponding to each target advertisement copy as the traffic distribution ratio corresponding thereto.

[0234] In the above example, the weight ratio of the second target weight factor to each second weight factor can be designed according to actual needs, for example, according to the importance of the number of each type of target advertisement copy in reflecting the popularity of the target advertisement copy among users. Then, the weight value corresponding to the second target weight factor can be obtained according to the respective weight ratios corresponding to the second target weight factor. The weight value obtained by this method is more in line with the needs of the system to promote popular advertisements among users. On this basis, the advertisement content service unit 101 weights each initial number of each target advertisement copy using the second standardized weight value to obtain the final number of each target advertisement copy. This method obtains the final number that can comprehensively reflect the popularity of the advertisement copy among users by integrating various types of number data. Furthermore, the corresponding traffic distribution ratio can be accurately determined according to the final number of the target advertisement copy, which realizes the ranking and recommendation of target advertisement copies according to the popularity among users, thereby further improving the conversion rate of advertising revenue.

[0235] S360, according to the traffic distribution ratio corresponding to each target advertisement copy, the target advertisement copy is recommended to the user of the at least one terminal again.

[0236] Illustratively, in order to grasp the degree of popularity of the target advertisement copy among users in real time, thereby ensuring the effect of advertisement recommendation, the advertisement content service unit 101 receives the latest feedback data corresponding to each target advertisement copy collected from the at least one terminal every predetermined time. Then, according to each latest feedback data, the latest traffic distribution ratio corresponding to each target advertisement copy is determined. In order to re-recommend each target advertisement copy to the user of the at least one terminal according to the latest traffic distribution ratio corresponding to each target advertisement copy.

[0237] The information interaction method for advertisement recommendation provided in the application allows a user to secondarily create an advertisement by sending creation invitation information to a terminal, and generates diversified advertisement copies. The method can optimize the recommendation effect of the advertisement by improving the creativity diversity and user participation, and improve the revenue conversion rate of the advertisement delivery. In order to better utilize the limited traffic resources for advertisement promotion, it is necessary to control the number of advertisement copies recommended to a user, and therefore, based on the content quality of each advertisement copy, a target advertisement copy with better content quality can be screened out for recommendation. In order to optimize the effect of the recommendation of the target advertisement copy, that is, the higher the user popularity, the more traffic cost is worth spending to recommend to more users, the application determines the traffic allocation proportion corresponding to each target advertisement copy through hierarchical analysis according to the feedback data collected by the terminal for reflecting the degree of user popularity of the target advertisement copy. The target advertisement copy is recommended to a user according to the traffic allocation proportion, which can optimize the advertisement delivery strategy and realize effective personalized recommendation. Since the advertisement copy represents the design of the advertisement content preferred by the user, recommending the advertisement copy can increase the user's interest in browsing the advertisement content, thereby improving the click volume, sharing volume and other data, so that the media platform further improves the exposure of the advertisement copy and the original version of the advertisement with the same promotion purpose as the advertisement copy, thereby further improving the recommendation effect of the advertisement.

[0238] In summary, the application realizes efficient dissemination of advertisement information, improves revenue conversion rate, creates higher commercial value, and reduces trial and error and operation cost of the advertisement platform by actively involving the users (i.e. consumers) of the advertisement in creating advertisement content.

[0239] Based on the above-mentioned information interaction system for advertisement recommendation, as shown in Figure 4 the application provides an information interaction method for advertisement recommendation, applied to a terminal 200, comprising:

[0240] S410, receiving and displaying an advertisement and creation invitation information corresponding to the advertisement to a user.

[0241] The creation invitation information is used to instruct the user to create an advertisement copy with the same promotion purpose as the advertisement.

[0242] S420, in response to the user's operation of clicking the advertisement, displaying a login page to the user.

[0243] The login page is used for the user to log in to a content creation service unit through an account, so that the user makes an advertisement copy according to the creation invitation information through the content creation service unit.

[0244] The above example realizes that the user remotely logs in to the content creation service unit through the terminal, and secondarily creates the advertisement according to the creation invitation information using the content creation service unit to generate the advertisement copy.

[0245] In a feasible design, cross-media platform sharing of an advertisement copy is implemented by breaking the advertisement channel threshold between applications, and the method comprises the following steps:

[0246] S430, receiving and displaying at least one target advertisement copy recommended by the advertisement content service unit through the first media platform;

[0247] S440, in response to the operation of the user clicking the sharing control corresponding to the target advertisement copy, displaying at least one second media platform that can be shared to the user, and in response to the operation of the user clicking the second media platform, sending the target advertisement copy or the network resource link of the target advertisement copy to the second media platform clicked by the user.

[0248] Exemplarily, the terminal collects log data such as request volume, return volume, exposure volume, click volume, loading volume, purchase volume, and sharing volume of each advertisement copy on each media platform, and sends the log data to the advertisement content service unit. So that the advertisement content service unit determines the traffic distribution proportion of each advertisement copy according to the data.

[0249] Other embodiments and effects of the terminal are described in the description of the system shown in Figure 1 and the method shown in Figure 2 and will not be described here.

[0250] The application also provides an information interaction device for advertisement recommendation, as shown in Figure 1 The device comprises an advertisement content service unit and a content creation service unit:

[0251] The advertisement content service unit is configured to send an advertisement and creation invitation information corresponding to the advertisement to at least one terminal, and the creation invitation information is configured to instruct a user to create an advertisement copy with the same promotion purpose as the advertisement;

[0252] The content creation service unit is configured to make at least one advertisement copy according to the creation invitation information by a user of the at least one terminal;

[0253] The advertisement content service unit is further configured to receive at least one advertisement copy from the content creation service unit;

[0254] The advertisement content service unit is further configured to select at least one target advertisement copy from the at least one advertisement copy based on the content quality of each advertisement copy, and recommend the at least one target advertisement copy to the user of the at least one terminal;

[0255] The advertisement content service unit is further configured to receive feedback data corresponding to each target advertisement copy collected from the at least one terminal, and the feedback data is configured to reflect the degree of popularity of the target advertisement copy among users;

[0256] The advertising content service unit is further configured to determine, based on each piece of feedback data, a traffic allocation ratio corresponding to each target advertising copy;

[0257] The advertising content service unit is further configured to recommend each target advertisement copy to a user of at least one terminal again according to the traffic distribution ratio corresponding to each target advertisement copy.

[0258] In a feasible design, at least one target advertisement copy is selected from at least one advertisement copy based on the content quality of each advertisement copy and recommended to a user of at least one terminal in the following manner:

[0259] Obtaining an initial score for each advertisement copy in terms of various content quality indicators, where the content quality indicators are used to evaluate the quality of the advertisement copy;

[0260] Determine a final score for each ad copy based on the initial scores of each ad copy, where the final score is a weighted sum of the initial scores of the ad copies;

[0261] sorting at least one ad copy in descending order according to the final score of each ad copy;

[0262] Determine each advertisement copy in the first number of advertisement copies that are ranked first as a target advertisement copy;

[0263] Determine the traffic allocation ratio corresponding to each target ad copy based on the final score of each target ad copy;

[0264] Each target advertisement copy is recommended to a user of at least one terminal according to a traffic distribution ratio corresponding to each target advertisement copy.

[0265] In a feasible design, the final score of each ad copy is determined based on the initial scores of each ad copy in the following manner:

[0266] determining a first weight ratio group corresponding to each first target weight factor, the first weight ratio group including a weight ratio of the first target weight factor to each first weight factor, the first weight factor being a content quality indicator, and the first target weight factor being one of the first weight factors corresponding to the advertisement copy;

[0267] Determining, based on the first weight ratio group of each first target weight factor, a geometric mean of the weight ratios corresponding to each first target weight factor;

[0268] Normalizing the geometric mean corresponding to each first target weight factor to obtain a first standardized weight value corresponding to each first target weight factor;

[0269] The initial score of each advertisement copy is weighted by using each first standardized weight value, and the sum of the weighted initial scores is taken as the final score of the advertisement copy.

[0270] In a possible design, the feedback data includes one or more initial times of the following time types:

[0271] request times, return times, exposure times, click times, loading times, purchase times, and sharing times.

[0272] The advertisement content service unit determines the traffic distribution proportion corresponding to each target advertisement copy according to each feedback data by the following manner:

[0273] determining a second weight ratio set corresponding to each second target weight factor, the second weight ratio set including a weight ratio of the second target weight factor to each second weight factor, the second weight factor being of a time type, and the second target weight factor being one of the second weight factors corresponding to the advertisement copy;

[0274] determining a geometric mean of each weight ratio corresponding to each second target weight factor based on the second weight ratio set of each second target weight factor;

[0275] normalizing the geometric mean corresponding to each second target weight factor to obtain a second standardized weight value corresponding to each second target weight factor;

[0276] weighting the initial time of the corresponding time type in each feedback data by using each second standardized weight value, and taking the sum of the weighted initial times as the final time of the corresponding target advertisement copy;

[0277] determining the traffic distribution proportion corresponding to each target advertisement copy according to the final time of each target advertisement copy.

[0278] In a possible design, the apparatus further includes a content review unit, configured to acquire a word library for storing sensitive words, and the word library stores and manages the sensitive words by using a dictionary tree structure.

[0279] The content review unit is further configured to acquire the to-be-reviewed text of the advertisement.

[0280] The content review unit is further configured to match characters of the to-be-reviewed text with first characters of the word library one by one, the first characters corresponding to nodes at a next level of root nodes of the dictionary tree structure.

[0281] The content review unit is further configured to determine characters in the to-be-reviewed text that are matched with the first characters of the word library as sensitive characters.

[0282] The content auditing unit is further configured to identify a to-be-matched word in the to-be-audited text, which has each sensitive character as a first character, and cache the to-be-matched word in the to-be-audited text.

[0283] The content auditing unit is further configured to match each cached to-be-matched word with a sensitive word in the word library.

[0284] If the content auditing unit determines that the to-be-audited text does not contain a word matched with a sensitive word in the word library, the advertisement content service unit is configured to send an advertisement and creation invitation information corresponding to the advertisement to at least one terminal.

[0285] Other embodiments and effects of the device can be found in the description of the information interaction method for advertisement recommendation, which will not be repeated here.

[0286] The present application provides another example of an information interaction device for advertisement recommendation, as shown in the figure, which includes an information transmission unit, an instruction processing unit and a display unit. Figure 1

[0287] The information transmission unit is configured to receive an advertisement and creation invitation information corresponding to the advertisement, and the creation invitation information is configured to instruct a user to create an advertisement copy with the same promotion purpose as the advertisement.

[0288] The display unit is configured to display the advertisement and the creation invitation information.

[0289] The instruction processing unit is configured to, in response to an operation of the user clicking the advertisement, display a login page to the user through the display unit, and the login page is configured to enable the user to log in to a content creation service unit through an account, so that the user creates the advertisement copy according to the creation invitation information through the content creation service unit.

[0290] The display unit is further configured to display a page for the user to remotely operate the content creation service unit.

[0291] In a feasible design, the information transmission unit is configured to receive at least one target advertisement copy recommended from an advertisement content service unit;

[0292] The display unit is configured to display the at least one target advertisement copy through a first media platform.

[0293] The instruction processing unit is configured to, in response to an operation of the user clicking a sharing control corresponding to the target advertisement copy, display at least one second media platform that can be shared to the user through the display unit, and the instruction processing unit is further configured to, in response to an operation of the user clicking the second media platform, send the target advertisement copy or a network resource link of the target advertisement copy to the second media platform clicked by the user through the information transmission unit.

[0294] Other embodiments and effects of the device can be found in the description of the information interaction method for advertisement recommendation, which will not be repeated here.​

[0295] The application also provides an information interaction system for advertisement recommendation, as shown in the drawings, comprising a server and a terminal, the server comprises an advertisement content service unit and a content creation service unit, and the terminal comprises an information transmission unit, a display unit and an instruction processing unit: Figure 1

[0296] The advertisement content service unit is configured to send an advertisement and creation invitation information corresponding to the advertisement to at least one terminal, and the creation invitation information is configured to instruct a user to create an advertisement copy with the same promotion purpose as the advertisement;

[0297] The information transmission unit is configured to receive the advertisement and the creation invitation information;

[0298] The display unit is configured to display the advertisement and the creation invitation information;

[0299] The instruction processing unit is configured to display a login page to the user through the display unit in response to an operation of the user clicking the advertisement, and the login page is configured to enable the user to log in to the content creation service unit through an account;

[0300] The display unit is configured to display a page for the user to remotely operate the content creation service unit;

[0301] The content creation service unit is configured to enable a user of at least one terminal to create at least one advertisement copy according to the creation invitation information;

[0302] The advertisement content service unit is further configured to receive the at least one advertisement copy from the content creation service unit;

[0303] The advertisement content service unit is further configured to select at least one target advertisement copy from the at least one advertisement copy based on the content quality of each advertisement copy, and recommend the at least one target advertisement copy to the user of the at least one terminal;

[0304] The advertisement content service unit is further configured to receive feedback data corresponding to each target advertisement copy collected from the at least one terminal, and the feedback data is configured to reflect the degree of popularity of the target advertisement copy among users;

[0305] The advertisement content service unit is further configured to determine a traffic allocation ratio corresponding to each target advertisement copy according to each feedback data;

[0306] The advertisement content service unit is further configured to recommend each target advertisement copy to the user of the at least one terminal again according to the traffic allocation ratio corresponding to each target advertisement copy.

[0307] Other embodiments and effects of the system can be found in the description of the information interaction method for advertisement recommendation, which will not be repeated here.

[0308] ​The above describes the basic principles of the present application in combination with specific embodiments, but it should be noted that the advantages, benefits, effects and the like mentioned in the present application are only examples and are not limiting, and these advantages, benefits, effects and the like cannot be considered as necessary for each embodiment of the present application. In addition, the above specific details disclosed are only for the purpose of example and understanding, and are not limiting, and the above details do not limit the present application to be necessarily implemented with the above specific details.

[0309] It should be understood that although each step in the flowchart of the accompanying drawings is shown in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and they can be executed in other sequences. Moreover, at least part of the steps in the flowchart of the accompanying drawings can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence is not necessarily sequential, but can be alternately or alternately executed with at least part of other steps or sub-steps or stages of other steps.

[0310] The block diagrams of the devices, apparatuses, equipment, systems involved in the present application are only illustrative examples, and are not intended to require or imply the connection, arrangement, configuration shown in the block diagram. As those skilled in the art will recognize, these devices, apparatuses, equipment, systems can be connected, arranged, configured in any manner. Words such as "include", "contain", "have" and the like are open-ended words, meaning "including but not limited to", and can be used interchangeably. The words "or" and "and" used herein mean the word "and / or", and can be used interchangeably unless the context clearly indicates otherwise. The word "such as" used herein means the phrase "such as but not limited to", and can be used interchangeably.

[0311] It should also be noted that in the devices, equipment and methods of the present application, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions of the present application.

[0312] The above description of the disclosed aspects is provided so that any person skilled in the art can make or use the present application. Various modifications to these aspects will be apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but is intended to be consistent with the widest scope consistent with the principles and novel features disclosed herein.

[0313] The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the forms disclosed herein. Although various example aspects and embodiments have been discussed above, those of ordinary skill in the art will appreciate a variety of modifications, alternatives, permutations, additions, and sub-combinations, which fall within the scope of the application.

Claims

1. An information interaction method for advertising recommendation, characterized in that: include: Sending an advertisement and creation invitation information corresponding to the advertisement to at least one terminal, wherein the creation invitation information is used to instruct a user to create an advertisement copy having the same promotional purpose as the advertisement; Obtaining at least one advertisement copy prepared by a user of at least one of the terminals according to the creation invitation information; selecting at least one target advertisement copy from at least one advertisement copy based on the content quality of each advertisement copy, and recommending the target advertisement copy to a user of at least one terminal; receiving feedback data corresponding to each target advertisement copy collected from at least one of the terminals, wherein the feedback data is used to reflect the degree to which the target advertisement copy is popular with users; Determining, based on each of the feedback data, a traffic allocation ratio corresponding to each of the target advertisement copies; According to the traffic distribution ratio corresponding to each target advertisement copy, each target advertisement copy is recommended again to a user of at least one of the terminals.

2. The method according to claim 1, characterized in that The selecting at least one target advertisement copy from at least one advertisement copy based on the content quality of each advertisement copy and recommending the target advertisement copy to a user of at least one terminal includes: Obtaining an initial score of each of the advertisement copies in terms of various content quality indicators, wherein the content quality indicators are used to evaluate the quality of the advertisement copies; Determine a final score of each of the advertisement copies according to the initial scores of each of the advertisement copies, wherein the final score is a weighted sum of the initial scores of the advertisement copies; sorting at least one of the advertisement copies in descending order according to the final score of each of the advertisement copies; Determine each of the advertisement copies in the first number of advertisement copies that are ranked first as a target advertisement copy; Determining a traffic allocation ratio corresponding to each target advertisement copy according to the final score of each target advertisement copy; Each target advertisement copy is recommended to a user of at least one of the terminals according to the traffic distribution ratio corresponding to each target advertisement copy.

3. The method according to claim 2, characterized in that Determining a final score of each of the advertisement copies based on the initial scores of each of the advertisement copies includes: determining a first weight ratio group corresponding to each first target weight factor, the first weight ratio group including a weight ratio of the first target weight factor to each first weight factor, the first weight factor being the content quality indicator, and the first target weight factor being one of the first weight factors corresponding to the advertisement copy; Determining, based on the first weight ratio group of each first target weight factor, a geometric mean of the weight ratios corresponding to each first target weight factor; Normalizing the geometric mean corresponding to each of the first target weight factors to obtain a first standardized weight value corresponding to each of the first target weight factors; The corresponding initial score of each of the advertisement copies is weighted using each first normalized weight value, and the sum of the weighted initial scores is used as the final score of the advertisement copy.

4. The method according to any one of claims 1 to 3, characterized in that The feedback data includes the initial number of one or more of the following number types: Number of requests, returns, impressions, clicks, loads, purchases, and shares; The step of determining the traffic allocation ratio corresponding to each target advertisement copy according to each feedback data includes: determining a second weight ratio group corresponding to each second target weight factor, the second weight ratio group including a weight ratio of the second target weight factor to each second weight factor, the second weight factor being the number type, and the second target weight factor being one of the second weight factors corresponding to the advertisement copy; Determining, based on the second weight ratio group of each second target weight factor, a geometric mean of the weight ratios corresponding to each second target weight factor; Normalizing the geometric mean corresponding to each of the second target weight factors to obtain a second standardized weight value corresponding to each of the second target weight factors; weighting the initial times of the corresponding times type in each feedback data using each second normalized weight value, and taking the sum of the weighted initial times as the final times of the corresponding target advertisement copy; According to the final number of each target advertisement copy, the traffic distribution ratio corresponding to each target advertisement copy is determined.

5. The method according to any one of claims 1 to 3, characterized in that The sending of the advertisement and the creation invitation information corresponding to the advertisement to at least one terminal includes: Obtaining a vocabulary for storing sensitive words, wherein the vocabulary utilizes a dictionary tree structure to store and manage sensitive words; Get the ad's text to be reviewed; Matching the characters of the text to be reviewed with the first character of the vocabulary one by one, the first character corresponding to the node of the next level of the root node of the dictionary tree structure; Determining characters in the text to be reviewed that match the first character of the vocabulary as sensitive characters; Identifying words to be matched in the text to be reviewed that have each of the sensitive characters as their first character, and caching the words to be matched in the text to be reviewed; Match each cached word to be matched with the sensitive words in the vocabulary; If the text to be reviewed does not contain any words that match the sensitive words in the vocabulary, the advertisement and creation invitation information corresponding to the advertisement are sent to at least one terminal.

6. An information interaction method for advertising recommendation, characterized in that: The method comprises: Receiving and displaying to a user an advertisement and a creation invitation message corresponding to the advertisement, wherein the creation invitation message is used to instruct the user to create an advertisement copy having the same promotional purpose as the advertisement; In response to the user clicking on the advertisement, a login page is displayed to the user, and the login page is used for the user to log in to the content creation service unit through an account, so that the user can create an advertisement copy through the content creation service unit based on the creation invitation information.

7. The method according to claim 6, characterized in that The method further comprises: receiving and displaying at least one target advertisement copy recommended by the advertisement content service unit through a first media platform; In response to the user clicking on the sharing control corresponding to the target advertising copy, at least one second media platform that can be shared is displayed to the user; in response to the user clicking on the second media platform, the target advertising copy or the network resource link of the target advertising copy is sent to the second media platform clicked by the user.

8. An information interaction device for advertising recommendation, characterized in that: Including advertising content service unit and content creation service unit: The advertising content service unit is used to send an advertisement and creation invitation information corresponding to the advertisement to at least one terminal, wherein the creation invitation information is used to instruct the user to create an advertisement copy with the same promotional purpose as the advertisement; The content creation service unit is used for a user of at least one of the terminals to create at least one advertisement copy according to the creation invitation information; The advertising content service unit is further configured to receive at least one advertisement copy from the content creation service unit; The advertising content service unit is further configured to select at least one target advertising copy from at least one of the advertising copies based on the content quality of each of the advertising copies, and recommend the target advertising copy to a user of at least one of the terminals; The advertising content service unit is further configured to receive feedback data corresponding to each target advertisement copy collected from at least one of the terminals, wherein the feedback data is used to reflect the degree to which the target advertisement copy is popular with users; The advertising content service unit is further configured to determine, based on each piece of feedback data, a traffic allocation ratio corresponding to each target advertising copy; The advertising content service unit is further configured to recommend each target advertisement copy to a user of at least one of the terminals again according to a traffic allocation ratio corresponding to each target advertisement copy.

9. An information interaction device for advertising recommendation, characterized in that: It includes an information transmission unit, an instruction processing unit and a display unit; The information transmission unit is used to receive an advertisement and creation invitation information corresponding to the advertisement, wherein the creation invitation information is used to instruct a user to create an advertisement copy having the same promotional purpose as the advertisement; The display unit is used to display the advertisement and the creation invitation information; The instruction processing unit is configured to display a login page to the user through the display unit in response to the user clicking on the advertisement, wherein the login page is used for the user to log in to the content creation service unit through an account, so that the user can create an advertisement copy according to the creation invitation information through the content creation service unit; The display unit is also used to display a page for users to remotely operate the content creation service unit.

10. An information interaction system for advertising recommendation, characterized in that: It includes a server and a terminal. The server includes an advertising content service unit and a content creation service unit. The terminal includes an information transmission unit, a display unit, and an instruction processing unit. The advertising content service unit is used to send an advertisement and creation invitation information corresponding to the advertisement to at least one of the terminals, wherein the creation invitation information is used to instruct the user to create an advertisement copy with the same promotional purpose as the advertisement; The information transmission unit is used to receive the advertisement and the creation invitation information; The display unit is used to display the advertisement and the creation invitation information; The instruction processing unit is configured to display a login page to the user through the display unit in response to the user clicking on the advertisement, wherein the login page is used for the user to log in to the content creation service unit through an account; The display unit is used to display a page for a user to remotely operate the content creation service unit; The content creation service unit is used for a user of at least one of the terminals to create at least one advertisement copy according to the creation invitation information; The advertising content service unit is further configured to receive at least one advertisement copy from the content creation service unit; The advertising content service unit is further configured to select at least one target advertising copy from at least one of the advertising copies based on the content quality of each of the advertising copies, and recommend the target advertising copy to a user of at least one of the terminals; The advertising content service unit is further configured to receive feedback data corresponding to each target advertisement copy collected from at least one of the terminals, wherein the feedback data is used to reflect the degree to which the target advertisement copy is popular with users; The advertising content service unit is further configured to determine, based on each piece of feedback data, a traffic allocation ratio corresponding to each target advertising copy; The advertising content service unit is further configured to recommend each target advertisement copy to a user of at least one of the terminals again according to a traffic allocation ratio corresponding to each target advertisement copy.

Images