Method and system for automatically generating and publishing all-media content

Abstract

The invention relates to a method for automatically generating and publishing all-media content, which belongs to the field of electric digital data processing, and comprises the following steps of: intelligently predicting the playing or opening type of a main image / audio / animation / webpage of a target residential user in a current time segment by adopting a content type prediction model; and automatically generating all-media content corresponding to the current time segment for the target residential user at a far-end all-media content server based on the intelligent prediction result. The invention further relates to an all-media content automatic generating and publishing system. According to the invention, the technical problem in the prior art that all-media content meeting watching requirements of users at different time periods cannot be automatically generated for each intelligent home network user is solved; the content type prediction model can be adopted to intelligently predict the playing or opening type of the main image / audio / animation / webpage of each intelligent home network user in the current time segment, and then the matched all-media content is generated, so that the technical problem is solved.

Description

Technical Field

[0001] The present invention relates to the field of electrical digital data processing, and in particular to a method and system for automatically generating and publishing all-media content. Background Art

[0002] All-media is a way of information dissemination that integrates various media presentation means such as audio, video, animation, and web pages. It is disseminated through different media forms such as movies, publications, newspapers, magazines, and websites, using the integrated radio and television network and the Internet, and finally enabling various terminals to receive information in a fused manner, meeting the needs of anyone to obtain the required information at any time, at any place, and through any terminal. The characteristic of all-media lies in the maximum integration of its media information flow means. From the perspective of the dissemination carrier tools, it covers various forms such as newspapers, magazines, radio, television, audio-visual, movies, publications, the Internet, telecommunications, and satellite communications. How to recommend all-media data that matches the time-sharing viewing habits of different users at different time segments for different users is the key to improving the generation speed and efficiency of all-media content.

[0003] Exemplarily, Chinese Patent Publication No. CN107943864A proposes a secure and controllable intelligent recommendation system in a multimedia content environment. The system filters pornographic, terrorist, and violent business content by combining rule-defined review, machine learning review, and manual inspection review to ensure that the subsequent recommended content is not illegal and does not violate morality and social ethics; at the same time, it comprehensively uses methods such as hot content recommendation, personalized recommendation based on user preferences, relevant recommendation based on user similarity and content similarity, and combines manual intervention to recommend content to users in real time. Through the combination of multiple recommendation algorithms, it meets the requirements of users in different times and different scenarios for the satisfaction, immediacy, variability, novelty, surprise, etc. of multimedia content consumption.

[0004] Exemplarily, Chinese Patent Publication No. CN117828463A proposes an AI-assisted method for automatically generating and publishing all-media content. The method includes: collecting hot data; obtaining a first heat eigenvalue according to the click-through rate and comment volume data in the hot data, obtaining a true evaluation eigenvalue of the hot data according to the historical data of the hot publisher, and obtaining the heat value of the hot data according to the first heat eigenvalue and the true evaluation eigenvalue of the hot data; constructing a hot data evaluation model according to the change of the heat value and the click volume of the hot data, and obtaining an evaluation eigenvalue of the necessity of publishing the hot data; judging the necessity of publishing the hot data according to the evaluation eigenvalue of the necessity of information publishing, and then automatically generating and publishing content through AI technology to complete the automatic generation and publishing of all-media content. The present invention aims to solve the problem of inaccurate acquisition of hot data.

[0005] However, the above technical solutions are either a rough all-media recommendation generation mechanism that requires manual intervention or an all-media recommendation generation mechanism that completely relies on hot data. Neither of them deeply analyzes the different demand types of various media such as audio, video, animation, and web pages for different users at different times. In particular, they do not automatically recommend, generate, and publish all-media content for different household users at different times with the micro-unit of all-media content on-demand and viewing, i.e., household users, resulting in a low level of intelligence and insufficient accuracy in the automatic recommendation, generation, and publishing mechanism of all-media content in the prior art. Summary of the Invention

[0006] To solve the technical problems in the prior art, the present invention provides an all-media content automatic generation and publishing method and system. For the users of the smart home network arranged and managed by the same wireless router, a content type prediction model with a customized structure is used to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the smart home network user in the current time segment based on various user-related information of the smart home network user and the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the smart home network user in each historical time segment before the current time segment. And at the all-media content server at the remote end, all-media content corresponding to the current time segment is automatically generated for the smart home network user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the smart home network user in the current time segment and published to the smart home network user, thereby improving the type recommendation accuracy of various different all-media content for each micro-unit of all-media content viewing, i.e., each household user, and meeting the all-media content viewing needs of each household user at different times.

[0007] According to the first aspect of the present invention, an all-media content automatic generation and publishing method is provided, and the method includes:

[0008] Analyze the number of permanent residents, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user as various user-related information of the target residential user;

[0009] Parse the data of each playback content type corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user to output the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment;

[0010] Perform multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model;

[0011] Use the content type prediction model to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment based on the duration of each time segment, various user-related information of the target residential user, the first historical playback content, second historical playback content, and third historical playback content corresponding to the target residential user in the current time segment;

[0012] At the remote all-media content server, automatically generate all-media content corresponding to the current time segment for the target residential user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment and publish it to the target residential user.

[0013] According to the second aspect of the present invention, an all-media content automatic generation and publishing system is provided. The system includes a memory and multiple processors. The multiple processors are located at the wireless router end of the target residential user and the remote all-media content server. The memory stores a computer program, and the computer program is configured to be executed by the multiple processors to complete the following steps:

[0014] Analyze the permanent population number, number of smart TVs, number of smart speakers, and network bandwidth of the target residential user as various user-related information of the target residential user;

[0015] At the wireless router end of the target residential user, analyze the playback content type data corresponding to each historical time segment before the current time segment for the mobile phone / smart TV / smart speaker of the target residential user and output it as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment;

[0016] Perform multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model;

[0017] Use the content type prediction model to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment based on the duration of each time segment, various user-related information of the target residential user, the first historical playback content, second historical playback content, and third historical playback content corresponding to the target residential user in the current time segment;

[0018] At the all-media content server at the remote end, based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time period, all-media content corresponding to the current time period is automatically generated for the target residential user and published to the target residential user.

[0019] According to the third aspect of the present invention, an all-media content automatic generation and publishing system is provided. The system includes:

[0020] A first parsing mechanism for parsing the permanent population number, number of smart TVs, number of smart speakers, and network bandwidth of the target residential user to serve as various user-related information of the target residential user;

[0021] A second parsing mechanism for parsing, at the wireless router end of the target residential user, the data of each playback content type corresponding to each historical time period before the current time period of the mobile phone / smart TV / smart speaker of the target residential user to output the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time period;

[0022] A step-by-step establishment mechanism for performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model;

[0023] A type prediction mechanism, connected to the first parsing mechanism, the second parsing mechanism, and the step-by-step establishment mechanism respectively, for using the content type prediction model to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time period according to the duration of each time period, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time period;

[0024] An automatic generation mechanism for automatically generating, at the all-media content server at the remote end, all-media content corresponding to the current time period for the target residential user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time period and publishing it to the target residential user.

[0025] Compared with the prior art, the present invention has at least the following five prominent substantive features:

[0026] First: For each residential user, an artificial intelligence model is used to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the residential user within the current time segment, which is a future time segment. Based on the intelligent prediction results, the all-media content of the residential user within the current time segment is automatically generated and published to the residential user. The recommended video playback type, recommended audio playback type, recommended animation playback type, and recommended web page opening type in the automatically generated all-media content are respectively the main video playback type, main audio playback type, main animation playback type, and main web page opening type within the current time segment. Thus, different all-media content matching different time periods is dynamically and automatically generated for each residential user and published to the residential user, improving the intelligent level of all-media content generation and the recommendation accuracy rate;

[0027] Second: A content type prediction model with a customized structure is used to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment. The content type prediction model is a Hoffit neural network after multiple learning operations, and the number of learning operations performed on the Hoffit neural network is positively correlated with the permanent population number of the target residential user. Thus, content type prediction models with different structures are constructed for different residential users, ensuring the effectiveness and stability of the intelligent prediction results;

[0028] Third: Multiple selected basic information is used to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment. The multiple basic information includes the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user within the current time segment. The full and comprehensive screening of the above multiple basic information further ensures the effectiveness and stability of the intelligent prediction results;

[0029] Fourth: Specifically, at the wireless router of the target residential user, the playback content type data corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user is parsed as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user within the current time segment. Among them, the single playback content type data corresponding to each time segment is the main video playback type, main audio playback type, main animation playback type, and main web page opening type within the time segment;

[0030] Fifth: In each learning operation performed on the Hoffit neural network, the main video playback types, main audio playback types, main animation playback types, and main web page opening types of known target residential users within a certain historical time period are used as the output content of the Hoffit neural network, and the duration of each time period, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user within the certain historical time period are used as the input content of the Hoffit neural network, thus completing the current learning operation and ensuring the learning effect of each learning operation performed on the Hoffit neural network. Description of the Drawings

[0031] The following will describe the embodiments of the present invention in conjunction with the drawings, where:

[0032] Figure 1 It is a technical flow chart of the all-media content automatic generation and publishing method and system according to the present invention.

[0033] Figure 2 It is a step flow chart of the all-media content automatic generation and publishing method shown in Embodiment 1 of the present invention.

[0034] Figure 3 It is a step flow chart of the all-media content automatic generation and publishing method shown in Embodiment 2 of the present invention.

[0035] Figure 4 It is a step flow chart of the all-media content automatic generation and publishing method shown in Embodiment 3 of the present invention.

[0036] Figure 5 It is a step flow chart of the all-media content automatic generation and publishing method shown in Embodiment 4 of the present invention.

[0037] Figure 6 It is a structural schematic diagram of the all-media content automatic generation and publishing system shown in Embodiment 5 of the present invention.

[0038] Figure 7 It is a structural schematic diagram of the all-media content automatic generation and publishing system shown in Embodiment 6 of the present invention. Detailed Embodiments

[0039] As Figure 1 shown, a technical flow chart of the all-media content automatic generation and publishing method and system according to the present invention is given.

[0040] The present invention performs intelligent prediction, automatic generation, and active publishing of all-media content that meets the future time-segment viewing habits of each intelligent home network user. The all-media content includes various different media contents such as images, audio, animations, and web pages. By intelligently predicting the different playback / open types of these various different media contents, the automatic generation of the all-media content for the user's future time segments is completed.

[0041] In Figure 1 it, the specific technical process of the present invention is as follows:

[0042] Technical process A: For the target residential users who are intelligent home network users arranged and managed by the same wireless router, a content type prediction model with a customized structure is designed to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential users in the current time segment, which is a kind of future time segment, so as to provide reliable basic data for the automatic generation of the all-media content in the current time segment;

[0043] Specifically, the structural customization of the content type prediction model is mainly manifested in the following aspects:

[0044] First, the content type prediction model is a Hopfield neural network after multiple learning operations;

[0045] Second, the number of learning operations performed on the Hopfield neural network is positively correlated with the number of permanent residents of the target residential user, so as to construct content type prediction models with different structures for different intelligent home network users;

[0046] Third, in each learning operation performed on the Hopfield neural network, the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the known target residential user in a certain historical time segment are used as the output content of the Hopfield neural network, and the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the certain historical time segment are used as the input content of the Hopfield neural network to complete this learning operation, thus ensuring the learning effect of each learning operation performed on the Hopfield neural network;

[0047] In this way, content type prediction models with different structures are constructed for different intelligent home network users, thus ensuring the effectiveness and stability of the intelligent prediction results;

[0048] Technical Process B: For intelligent prediction of the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment which is a future time segment divided by time, a number of basic information items are screened specifically;

[0049] Specifically, the number of basic information items includes the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment. The full and comprehensive screening of the above-mentioned number of basic information items further ensures the effectiveness and stability of the intelligent prediction results;

[0050] More specifically, at the wireless router end of the target residential user, the data of the playback content types corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user are parsed as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment. Among them, the single-piece playback content type data corresponding to each time segment is the main video playback type, main audio playback type, main animation playback type, and main web page opening type within the time segment;

[0051] In this way, the full and comprehensive screening of the above-mentioned number of basic information items further ensures the effectiveness and stability of the intelligent prediction results;

[0052] Technical Process C: Use Technical Process A to customize a content type prediction model with a structural design for the target residential user. Based on the number of basic information items screened specifically by Technical Process B, intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment which is a future time segment divided by time;

[0053] Technical Process D: At the remote all-media content server, based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user intelligently predicted by Technical Process C in the current time segment, automatically generate the all-media content corresponding to the current time segment for the target residential user and publish it to the target residential user;

[0054] Specifically, the recommended video playback type, recommended audio playback type, recommended animation playback type, and recommended web page opening type in the automatically generated all-media content are respectively the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment;

[0055] Technical process E: At the all-media content server at the remote end, the all-media content corresponding to the current time segment is automatically generated for the target residential user and sent to the target residential user;

[0056] It can be seen that the present invention improves the accuracy of type recommendation of various different all-media contents for each micro unit of all-media content viewing, that is, each household user, and meets the all-media content viewing needs of each household user at different time periods.

[0057] The key points of the present invention are: the automatic generation and active release of all-media content preferred by the smart home network user in the current time segment based on the intelligent prediction results of the main video playback type, main audio playback type, main animation playback type, and main web page opening type of each smart home network user in the current time segment, the customization of content type prediction models with different structures for different smart home network users, and the comprehensive and sufficient screening of a number of basic information for intelligent prediction of various media types.

[0058] Next, the all-media content automatic generation and release method and system of the present invention will be specifically described by way of embodiments.

[0059] Embodiment 1

[0060] Figure 2 It is a step flow chart of the all-media content automatic generation and release method shown in Embodiment 1 according to the present invention.

[0061] As Figure 2 shown, the all-media content automatic generation and release method includes the following specific steps:

[0062] Step S101: Analyze the permanent population number, number of smart TVs, number of smart speakers, and network bandwidth of the target residential user to use as various user-related information of the target residential user;

[0063] Exemplarily, analyzing the permanent population number, number of smart TVs, number of smart speakers, and network bandwidth of the target residential user to use as various user-related information of the target residential user includes: multiple value analysis components can be selected to respectively analyze the permanent population number, number of smart TVs, number of smart speakers, and network bandwidth of the target residential user;

[0064] Step S102: Analyze the data of each playback content type corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user to output the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment;

[0065] Specifically, parsing the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user to output the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment includes: the current time segment is from 11:45 am to 12:00 pm, and the current time segment starts from the current moment, that is, 11:45 am, so it can be regarded as a future time segment;

[0066] Step S103: Perform multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model;

[0067] Exemplarily, performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model includes: a numerical simulation mechanism can be selected to complete the simulation and testing of the model construction process of performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model;

[0068] Step S104: Use the content type prediction model to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment according to the duration of each time segment, various user association information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment;

[0069] Step S105: Automatically generate the all-media content corresponding to the current time segment for the target residential user at the remote all-media content server based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment and publish it to the target residential user;

[0070] Exemplarily, automatically generating the all-media content corresponding to the current time segment for the target residential user at the remote all-media content server based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment and publishing it to the target residential user includes: using a frequency division duplex communication link to automatically generate the all-media content corresponding to the current time segment for the target residential user and publish it to the target residential user;

[0071] Among them, the single-piece play content type data corresponding to each time segment is the main video play type, main audio play type, main animation play type, and main web page opening type within the time segment;

[0072] Among them, at the all-media content server at the remote end, based on the main video play type, main audio play type, main animation play type, and main web page opening type of the target residential user within the current time segment, automatically generating the all-media content corresponding to the current time segment for the target residential user and publishing it to the target residential user includes: the recommended video play type, recommended audio play type, recommended animation play type, and recommended web page opening type in the automatically generated all-media content are respectively the main video play type, main audio play type, main animation play type, and main web page opening type of the target residential user within the current time segment;

[0073] Among them, parsing the play content type data corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user to output the first historical play content / second historical play content / third historical play content corresponding to the current time segment of the target residential user includes: each historical time segment before the current time segment and the current time segment form a complete time interval on the time axis, and the duration of each time segment is equal to the set duration;

[0074] Exemplarily, each historical time segment before the current time segment and the current time segment form a complete time interval on the time axis, and the duration of each time segment is equal to the set duration includes: the duration of each time segment is 15 minutes;

[0075] Among them, performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model includes: the number of learning operations performed on the Hoffit neural network is positively correlated with the permanent population of the target residential user;

[0076] Exemplarily, the number of learning operations performed on the Hoffit neural network is positively correlated with the permanent population of the target residential user includes: when the permanent population of the target residential user is 1 person, the number of learning operations performed on the Hoffit neural network is 200 times; when the permanent population of the target residential user is 2 people, the number of learning operations performed on the Hoffit neural network is 300 times; when the permanent population of the target residential user is 3 people, the number of learning operations performed on the Hoffit neural network is 400 times; when the permanent population of the target residential user is 4 people, the number of learning operations performed on the Hoffit neural network is 500 times, and so on;

[0077] And among them, performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model further includes: in each learning operation performed on the Hoffit neural network, using the main video playback type, main audio playback type, main animation playback type, and main web page opening type of a known target residential user within a certain historical time period as the output content of the Hoffit neural network, and using the duration of each time period, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user within the certain historical time period as the input content of the Hoffit neural network to complete the current learning operation.

[0078] Embodiment 2

[0079] Figure 3 It is a flowchart of the steps of the all-media content automatic generation and publishing method shown in Embodiment 2 of the present invention.

[0080] As Figure 3 shown, different from the embodiment in Figure 2 after automatically generating the all-media content corresponding to the current time period for the target residential user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user at the all-media content server at the remote end and publishing it to the target residential user, that is, after step S105, the method further includes:

[0081] Step S106: Recording the all-media content corresponding to the current time period automatically generated for the target residential user at the all-media content server at the remote end;

[0082] Exemplarily, recording the all-media content corresponding to the current time period automatically generated for the target residential user at the all-media content server at the remote end includes: the all-media content server at the remote end is a big data service network element, a blockchain service network element, or a cloud computing service network element.

[0083] Embodiment 3

[0084] Figure 4 It is a flowchart of the steps of the all-media content automatic generation and publishing method shown in Embodiment 3 of the present invention.

[0085] As Figure 4 shown, different from the embodiment in Figure 2Different from the embodiments in the embodiment, after using the content type prediction model to intelligently predict the main image playback type, main audio playback type, main animation playback type and main web page opening type of the target residential user in the current time segment according to the duration of each time segment, various user association information of the target residential user, the first historical playback content, the second historical playback content and the third historical playback content corresponding to the target residential user in the current time segment, after step S104, the method further includes:

[0086] Step S107: Displaying the main image playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment on the wireless router of the target residential user;

[0087] For example, displaying the main image playback type, main audio playback type, main animation playback type and main web page opening type of the target residential user in the current time segment on the wireless router end of the target residential user includes: using a liquid crystal display screen embedded in the front panel of the wireless router end of the target residential user to display the main image playback type, main audio playback type, main animation playback type and main web page opening type of the target residential user in the current time segment.

[0088] Example 4

[0089] Figure 5 This is a flowchart of the steps of a method for automatically generating and publishing all-media content according to Embodiment 4 of the present invention.

[0090] like Figure 5 As shown, Figure 2 Different from the embodiment in, after performing multiple learning operations on the Hoffet neural network to obtain the Hoffet neural network after the multiple learning operations and outputting it as the content type prediction model, that is, after step S103, the method further includes:

[0091] Step S108: Complete the model storage of the content type prediction model by storing various model parameters of the content type prediction model

[0092] For example, completing the model storage of the content type prediction model by storing various model parameters of the content type prediction model includes: selecting to use a TF storage chip or an MMC storage chip to complete the model storage of the content type prediction model by storing various model parameters of the content type prediction model.

[0093] Next, various method embodiments of the present invention are described in detail.

[0094] In the method for automatically generating and publishing all-media content according to various method embodiments of the present invention:

[0095] Using a content type prediction model, based on the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment, to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment, including: performing numerical normalization processing on the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment respectively, and then inputting them into the content type prediction model;

[0096] Exemplarily, performing numerical normalization processing on the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment respectively, and then inputting them into the content type prediction model includes: the numerical normalization processing is octal numerical conversion processing;

[0097] Among them, using a content type prediction model, based on the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment, to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment further includes: executing the content type prediction model to obtain the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment output by the content type prediction model.

[0098] And in the all-media content automatic generation and publishing method according to each method embodiment of the present invention:

[0099] Parse the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and output it as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment. It further includes: when there are multiple mobile phones of the target residential user, for each historical time segment, place more than one video type played by each mobile phone in the historical time segment into the video type set of the mobile phone of the target residential user in the historical time segment, and use the video type with the most occurrences in the video type set of the mobile phone of the target residential user in the historical time segment as the main video playback type of the mobile phone of the target residential user in the historical time segment;

[0100] Among them, parsing the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting it as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment further includes: when there are multiple smart TVs of the target residential user, for each historical time segment, place more than one animation type played by each smart TV in the historical time segment into the animation type set of the smart TV of the target residential user in the historical time segment, and use the animation type with the most occurrences in the animation type set of the smart TV of the target residential user in the historical time segment as the main animation playback type of the smart TV of the target residential user in the historical time segment;

[0101] Among them, parsing the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting it as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment further includes: when there are multiple smart TVs of the target residential user, for each historical time segment, place more than one web page type opened by each smart TV in the historical time segment into the web page type set of the smart TV of the target residential user in the historical time segment, and use the web page type with the most occurrences in the web page type set of the smart TV of the target residential user in the historical time segment as the main web page opening type of the smart TV of the target residential user in the historical time segment;

[0102] Among them, parsing the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment, further includes: when there are multiple smart speakers for the target residential user, for each historical time segment, placing more than one audio type played by each smart speaker in the historical time segment into the audio type set of the smart speaker of the target residential user in the historical time segment, and taking the audio type with the most occurrences in the audio type set of the smart speaker of the target residential user in the historical time segment as the main audio playback type of the smart speaker of the target residential user in the historical time segment;

[0103] Among them, parsing the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment, further includes: when there are multiple mobile phones for the target residential user, for each historical time segment, placing more than one audio type played by each mobile phone in the historical time segment into the audio type set of the mobile phone of the target residential user in the historical time segment, and taking the audio type with the most occurrences in the audio type set of the mobile phone of the target residential user in the historical time segment as the main audio playback type of the mobile phone of the target residential user in the historical time segment;

[0104] Among them, parsing the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment, further includes: when there are multiple mobile phones for the target residential user, for each historical time segment, placing more than one web page type opened by each mobile phone in the historical time segment into the web page type set of the mobile phone of the target residential user in the historical time segment, and taking the web page type with the most occurrences in the web page type set of the mobile phone of the target residential user in the historical time segment as the main web page opening type of the mobile phone of the target residential user in the historical time segment;

[0105] Among them, parsing the data of each type of played content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the data as the first historical played content / second historical played content / third historical played content corresponding to the target residential user in the current time segment. It further includes: when there are multiple mobile phones of the target residential user, for each historical time segment, placing more than one type of animation played by each mobile phone in the historical time segment into the animation type set of the mobile phone of the target residential user in the historical time segment, and taking the animation type with the most occurrences in the animation type set of the mobile phone of the target residential user in the historical time segment as the main animation played type of the mobile phone of the target residential user in the historical time segment;

[0106] Among them, parsing the data of each type of played content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the data as the first historical played content / second historical played content / third historical played content corresponding to the target residential user in the current time segment. It further includes: when there are multiple smart TVs of the target residential user, for each historical time segment, placing more than one type of video played by each smart TV in the historical time segment into the video type set of the smart TV of the target residential user in the historical time segment, and taking the video type with the most occurrences in the video type set of the smart TV of the target residential user in the historical time segment as the main video played type of the smart TV of the target residential user in the historical time segment;

[0107] And among them, parsing the data of each type of played content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the data as the first historical played content / second historical played content / third historical played content corresponding to the target residential user in the current time segment. It further includes: when there are multiple smart TVs of the target residential user, for each historical time segment, placing more than one type of audio played by each smart TV in the historical time segment into the audio type set of the smart TV of the target residential user in the historical time segment, and taking the audio type with the most occurrences in the audio type set of the smart TV of the target residential user in the historical time segment as the main audio played type of the smart TV of the target residential user in the historical time segment.

[0108] Embodiment 5

[0109] Figure 6 FIG. 5 is a schematic structural diagram of an all-media content automatic generation and publishing system according to Embodiment 5 of the present invention.

[0110] As shown Figure 6 in the figure, the all-media content automatic generation and publishing system includes a memory and multiple processors. The multiple processors are respectively located at the wireless router end of the target residential user and the all-media content server at the remote end. The memory stores a computer program, and the computer program is configured to be executed by the multiple processors to complete the following steps:

[0111] Step S101: Analyze the permanent population quantity, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user to use them as various user-related information of the target residential user;

[0112] Exemplarily, analyzing the permanent population quantity, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user to use them as various user-related information of the target residential user includes: Multiple numerical analysis components can be selected to respectively analyze the permanent population quantity, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user;

[0113] Step S102: At the wireless router end of the target residential user, analyze the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user to output the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user at the current time segment;

[0114] Specifically, analyzing the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user to output the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user at the current time segment includes: The current time segment is from 11:45 am to 12:00 pm. The current time segment starts from the current moment, that is, 11:45 am, so it can be regarded as a kind of future time segment;

[0115] Step S103: Perform multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model;

[0116] Exemplarily, performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model includes: A numerical simulation mechanism can be selected to complete the simulation and testing of the model construction process of performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model;

[0117] Step S104: Use the content type prediction model to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment based on the duration of each time segment, various user association information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment;

[0118] Step S105: At the remote omnimedia content server, automatically generate the omnimedia content corresponding to the current time segment for the target residential user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment, and publish it to the target residential user;

[0119] Exemplarily, at the remote omnimedia content server, automatically generating the omnimedia content corresponding to the current time segment for the target residential user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment, and publishing it to the target residential user includes: using a frequency division duplex communication link to automatically generate the omnimedia content corresponding to the current time segment for the target residential user and publish it to the target residential user;

[0120] Among them, the single-piece playback content type data corresponding to each time segment is the main video playback type, main audio playback type, main animation playback type, and main web page opening type within the said time segment;

[0121] Among them, at the remote omnimedia content server, automatically generating the omnimedia content corresponding to the current time segment for the target residential user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment, and publishing it to the target residential user includes: the recommended video playback type, recommended audio playback type, recommended animation playback type, and recommended web page opening type in the automatically generated omnimedia content are respectively the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment;

[0122] Among them, parsing the respective pieces of playback content type data corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router of the target residential user to output as the first historical playback content / second historical playback content / third historical playback content corresponding to the current time segment of the target residential user includes: each historical time segment before the current time segment and the current time segment form a complete time interval on the time axis, and the duration of each time segment is equal to the set duration;

[0123] Exemplarily, each historical time segment before the current time segment and the current time segment form a complete time interval on the time axis, and the duration of each time segment is equal to the set duration, including: the duration of each time segment is 15 minutes;

[0124] Among them, performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model includes: the number of learning operations performed on the Hoffit neural network is positively correlated with the number of permanent residents of the target residential user;

[0125] Exemplarily, the number of learning operations performed on the Hoffit neural network being positively correlated with the number of permanent residents of the target residential user includes: when the number of permanent residents of the target residential user is 1, the number of learning operations performed on the Hoffit neural network is 200 times; when the number of permanent residents of the target residential user is 2, the number of learning operations performed on the Hoffit neural network is 300 times; when the number of permanent residents of the target residential user is 3, the number of learning operations performed on the Hoffit neural network is 400 times; when the number of permanent residents of the target residential user is 4, the number of learning operations performed on the Hoffit neural network is 500 times, and so on;

[0126] And among them, performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model further includes: in each learning operation performed on the Hoffit neural network, taking the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the known target residential user in a certain historical time segment as the output content of the Hoffit neural network, and taking the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the certain historical time segment as the input content of the Hoffit neural network to complete the current learning operation;

[0127] Such as Figure 6 shown, exemplarily, N processors are given, and the N processors are respectively located at the wireless router end of the target residential user and the all-media content server at the remote end, where N is a natural number greater than or equal to 1.

[0128] Embodiment 6

[0129] Figure 7 It is a schematic structural diagram of an all-media content automatic generation and publishing system shown according to Embodiment 6 of the present invention.

[0130] Such as Figure 7As shown in the figure, the all-media content automatic generation and publishing system includes the following components:

[0131] An information input mechanism, which is used to analyze the permanent population quantity, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user, so as to serve as various user-related information of the target residential user;

[0132] Exemplarily, analyzing the permanent population quantity, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user to serve as various user-related information of the target residential user includes: Multiple numerical analysis components can be selected to respectively analyze the permanent population quantity, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user;

[0133] A formula collection mechanism, which is used to analyze the data of each play content type corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, so as to output the first historical play content / second historical play content / third historical play content corresponding to the target residential user at the current time segment;

[0134] Specifically, analyzing the data of each play content type corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user to output the first historical play content / second historical play content / third historical play content corresponding to the target residential user at the current time segment includes: The current time segment is from 11:45 am to 12:00 pm, and the current time segment starts from the current moment, that is, 11:45 am, so it can be regarded as a future time segment;

[0135] A multiple training mechanism, which is used to perform multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model;

[0136] Exemplarily, performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model includes: A numerical simulation mechanism can be selected to complete the simulation and testing of the model construction process of performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as a content type prediction model;

[0137] An intelligent prediction mechanism, which is respectively connected to the information input mechanism, the formula collection mechanism, and the multiple training mechanism, is used to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment according to the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment by using a content type prediction model;

[0138] A recommendation processing mechanism, which is connected to the intelligent prediction mechanism, is used to automatically generate all-media content corresponding to the current time segment for the target residential user at the all-media content server at the remote end based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment and publish it to the target residential user;

[0139] Exemplarily, automatically generating all-media content corresponding to the current time segment for the target residential user at the all-media content server at the remote end based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment and publishing it to the target residential user includes: using a frequency-division duplex communication link to automatically generate all-media content corresponding to the current time segment for the target residential user and publish it to the target residential user;

[0140] Wherein, the single-piece playback content type data corresponding to each time segment is the main video playback type, main audio playback type, main animation playback type, and main web page opening type within the time segment;

[0141] Wherein, automatically generating all-media content corresponding to the current time segment for the target residential user at the all-media content server at the remote end based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment and publishing it to the target residential user includes: the recommended video playback type, recommended audio playback type, recommended animation playback type, and recommended web page opening type in the automatically generated all-media content are respectively the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment;

[0142] The wireless router of the target residential user analyzes the playback content type data corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user to output the first historical playback content / second historical playback content / third historical playback content corresponding to the current time segment of the target residential user, including: each historical time segment before the current time segment and the current time segment form a complete time interval on the time axis, and the duration of each time segment is equal to the set duration;

[0143] For example, each historical time segment before the current time segment and the current time segment form a complete time interval on the time axis, and the duration of each time segment is equal to the set duration, including: each time segment lasts for 15 minutes;

[0144] Wherein, performing multiple learning operations on the Hoffitt neural network to obtain the Hoffitt neural network after the multiple learning operations and outputting the Hoffitt neural network as the content type prediction model includes: the number of learning operations performed on the Hoffitt neural network is positively correlated with the number of permanent residents of the target residential users;

[0145] For example, the number of learning operations performed on the Hoffitt neural network is positively correlated with the number of permanent residents of the target residential user, including: when the number of permanent residents of the target residential user is 1, the number of learning operations performed on the Hoffitt neural network is 200 times, when the number of permanent residents of the target residential user is 2, the number of learning operations performed on the Hoffitt neural network is 300 times, when the number of permanent residents of the target residential user is 3, the number of learning operations performed on the Hoffitt neural network is 400 times, when the number of permanent residents of the target residential user is 4, the number of learning operations performed on the Hoffitt neural network is 500 times, and so on;

[0146] And wherein, performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model also includes: in each learning operation performed on the Hoffit neural network, using the main image playback type, main audio playback type, main animation playback type and main web page opening type of the known target residential user in a certain historical time segment as the output content of the Hoffit neural network, and using the duration of each time segment, various user-related information of the target residential user, and the first historical playback content, second historical playback content and third historical playback content corresponding to the target residential user in the said certain historical time segment as the input content of the Hoffit neural network to complete this learning operation.

[0147] In addition, the present invention may also cite the following technical contents to further demonstrate the outstanding substantial progress of the present invention:

[0148] Among them, parsing the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user to output the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user at the current time segment further includes: the number of time segments of each historical time segment before the current time segment is positively correlated with the network bandwidth of the target residential user;

[0149] And among them, the number of time segments of each historical time segment before the current time segment being positively correlated with the network bandwidth of the target residential user includes: using a numerical mapping formula to represent the numerical mapping relationship between the number of time segments of each historical time segment before the current time segment and the network bandwidth of the target residential user;

[0150] Exemplarily, using a numerical mapping formula to represent the numerical mapping relationship between the number of time segments of each historical time segment before the current time segment and the network bandwidth of the target residential user includes: in the numerical mapping formula, the network bandwidth of the target residential user is the input value of the numerical mapping formula, and the number of time segments of each historical time segment before the current time segment corresponding to the network bandwidth of the target residential user is the output value of the numerical mapping formula.

[0151] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limitations on the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. An automatic generation and publishing method for all-media content, characterized in that, The method includes: Analyzing the number of permanent residents, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user as various user association information of the target residential user; Parsing, at the wireless router of the target residential user, the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user, and outputting the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user at the current time segment; Performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model; Using the content type prediction model to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment according to the duration of each time segment, the various user association information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user at the current time segment; At the remote all-media content server, automatically generating all-media content corresponding to the current time segment for the target residential user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment and publishing it to the target residential user.

2. The all-media content automatic generation and publishing method according to claim 1, wherein: The data of a single type of playback content corresponding to each time segment is the main video playback type, main audio playback type, main animation playback type, and main web page opening type within the time segment; Among them, automatically generating all-media content corresponding to the current time segment for the target residential user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user at the remote all-media content server includes: the recommended video playback type, recommended audio playback type, recommended animation playback type, and recommended web page opening type in the automatically generated all-media content are respectively the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time segment; Among them, parsing, at the wireless router of the target residential user, the data of each type of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user, and outputting the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user at the current time segment includes: each historical time segment before the current time segment and the current time segment form a complete time interval on the time axis, and the duration of each time segment is equal to the set duration.

3. The all-media content automatic generation and publishing method according to claim 2, wherein: ​ Performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model includes: the number of learning operations performed on the Hoffit neural network is positively correlated with the permanent population of the target residential user; Among them, performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model further includes: in each learning operation performed on the Hoffit neural network, taking the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the known target residential user within a certain historical time period as the output content of the Hoffit neural network, taking the duration of each time period, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user within the certain historical time period as the input content of the Hoffit neural network, and completing the current learning operation.

4. The all-media content automatic generation and publishing method according to claim 3, wherein After automatically generating the all-media content corresponding to the current time period for the target residential user based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user at the all-media content server at the remote end and publishing it to the target residential user, the method further includes: Recording the all-media content corresponding to the current time period automatically generated for the target residential user at the all-media content server at the remote end.

5. The method for automatically generating and publishing all-media content according to claim 3, characterized in that, After using the content type prediction model to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time period according to the duration of each time period, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user within the current time period, the method further includes: Displaying the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user within the current time period at the wireless router of the target residential user.

6. The all-media content automatic generation and release method according to claim 3, characterized in that After performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model, the method further includes: Completing the model storage of the content type prediction model by storing various model parameters of the content type prediction model.

7. The all-media content automatic generation and publishing method according to any one of claims 3-6, characterized in that: Using a content type prediction model, based on the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment, to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment, including: performing numerical normalization processing on the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment respectively, and then inputting them into the content type prediction model; Using a content type prediction model, based on the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment, to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment further includes: executing the content type prediction model to obtain the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment output by the content type prediction model.

8. The all-media content automatic generation and publishing method according to any one of claims 3-6, characterized in that: Parsing, at the wireless router end of the target residential user, the data of each playback content type corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment output further includes: when there are multiple mobile phones of the target residential user, for each historical time segment, placing all video types played by each mobile phone in the historical time segment in the video type set of the mobile phone of the target residential user in the historical time segment, and taking the video type with the most occurrences in the video type set of the mobile phone of the target residential user in the historical time segment as the main video playback type of the mobile phone of the target residential user in the historical time segment; Among them, parsing the data of each type of played content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the data as the first historical played content / second historical played content / third historical played content corresponding to the target residential user in the current time segment. It further includes: when there are multiple smart TVs for the target residential user, for each historical time segment, placing more than one type of animation played by each smart TV in the historical time segment into the set of animation types of the smart TVs of the target residential user in the historical time segment, and taking the animation type with the most occurrences in the set of animation types of the smart TVs of the target residential user in the historical time segment as the main animation playing type of the smart TVs of the target residential user in the historical time segment; Among them, parsing the data of each type of played content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the data as the first historical played content / second historical played content / third historical played content corresponding to the target residential user in the current time segment. It further includes: when there are multiple smart TVs for the target residential user, for each historical time segment, placing more than one type of web page opened by each smart TV in the historical time segment into the set of web page types of the smart TVs of the target residential user in the historical time segment, and taking the web page type with the most occurrences in the set of web page types of the smart TVs of the target residential user in the historical time segment as the main web page opening type of the smart TVs of the target residential user in the historical time segment; Among them, parsing the data of each type of played content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the data as the first historical played content / second historical played content / third historical played content corresponding to the target residential user in the current time segment. It further includes: when there are multiple smart speakers for the target residential user, for each historical time segment, placing more than one type of audio played by each smart speaker in the historical time segment into the set of audio types of the smart speakers of the target residential user in the historical time segment, and taking the audio type with the most occurrences in the set of audio types of the smart speakers of the target residential user in the historical time segment as the main audio playing type of the smart speakers of the target residential user in the historical time segment; Among them, parsing the data of each type of playback content corresponding to each historical time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user before the current time segment, and outputting the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment, further includes: when there are multiple mobile phones of the target residential user, for each historical time segment, placing one or more audio types played by each mobile phone in the historical time segment into the audio type set of the mobile phone of the target residential user in the historical time segment, and taking the audio type with the most occurrences in the audio type set of the mobile phone of the target residential user in the historical time segment as the main audio playback type of the mobile phone of the target residential user in the historical time segment; Among them, parsing the data of each type of playback content corresponding to each historical time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user before the current time segment, and outputting the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment, further includes: when there are multiple mobile phones of the target residential user, for each historical time segment, placing one or more web page types opened by each mobile phone in the historical time segment into the web page type set of the mobile phone of the target residential user in the historical time segment, and taking the web page type with the most occurrences in the web page type set of the mobile phone of the target residential user in the historical time segment as the main web page opening type of the mobile phone of the target residential user in the historical time segment; Among them, parsing the data of each type of playback content corresponding to each historical time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user before the current time segment, and outputting the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment, further includes: when there are multiple mobile phones of the target residential user, for each historical time segment, placing one or more animation types played by each mobile phone in the historical time segment into the animation type set of the mobile phone of the target residential user in the historical time segment, and taking the animation type with the most occurrences in the animation type set of the mobile phone of the target residential user in the historical time segment as the main animation playback type of the mobile phone of the target residential user in the historical time segment; Among them, parsing the data of various types of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the data as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment, further includes: when there are multiple smart TVs for the target residential user, for each historical time segment, placing more than one video type played by each smart TV in the historical time segment into the video type set of the smart TV of the target residential user in the historical time segment, and taking the video type with the most occurrences in the video type set of the smart TV of the target residential user in the historical time segment as the main video playback type of the smart TV of the target residential user in the historical time segment; Among them, parsing the data of various types of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the data as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment, further includes: when there are multiple smart TVs for the target residential user, for each historical time segment, placing more than one audio type played by each smart TV in the historical time segment into the audio type set of the smart TV of the target residential user in the historical time segment, and taking the audio type with the most occurrences in the audio type set of the smart TV of the target residential user in the historical time segment as the main audio playback type of the smart TV of the target residential user in the historical time segment.

9. An all-media content automatic generation and publishing system, characterized in that, The system includes a memory and multiple processors, the multiple processors are located at the wireless router end of the target residential user and the all-media content server at the remote end, the memory stores a computer program, and the computer program is configured to be executed by the multiple processors to complete the following steps: Parsing the permanent population number, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user as various user-related information of the target residential user; Parsing the data of various types of playback content corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and outputting the data as the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment; Performing multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and outputting it as a content type prediction model; The content type prediction model is used to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment according to the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment; At the all-media content server at the remote end, based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment, the all-media content corresponding to the current time segment is automatically generated for the target residential user and published to the target residential user.

10. An all-media content automatic generation and publishing system, characterized in that, The system includes: The first parsing mechanism is used to parse the permanent population number, the number of smart TVs, the number of smart speakers, and the network bandwidth of the target residential user as various user-related information of the target residential user; The second parsing mechanism is used to parse the data of each playback content type corresponding to each historical time segment before the current time segment of the mobile phone / smart TV / smart speaker of the target residential user at the wireless router end of the target residential user, and output the first historical playback content / second historical playback content / third historical playback content corresponding to the target residential user in the current time segment; The hierarchical establishment mechanism is used to perform multiple learning operations on the Hoffit neural network to obtain the Hoffit neural network after multiple learning operations and output it as the content type prediction model; The type prediction mechanism is respectively connected to the first parsing mechanism, the second parsing mechanism, and the hierarchical establishment mechanism, and is used to use the content type prediction model to intelligently predict the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment according to the duration of each time segment, various user-related information of the target residential user, the first historical playback content, the second historical playback content, and the third historical playback content corresponding to the target residential user in the current time segment; The automatic generation mechanism is used to automatically generate the all-media content corresponding to the current time segment for the target residential user at the all-media content server at the remote end based on the main video playback type, main audio playback type, main animation playback type, and main web page opening type of the target residential user in the current time segment and publish it to the target residential user.

Images