CN114866845A - Information detection method and system based on short video release - Google Patents

Information detection method and system based on short video release Download PDF

Info

Publication number
CN114866845A
CN114866845A CN202210781713.0A CN202210781713A CN114866845A CN 114866845 A CN114866845 A CN 114866845A CN 202210781713 A CN202210781713 A CN 202210781713A CN 114866845 A CN114866845 A CN 114866845A
Authority
CN
China
Prior art keywords
preference
interaction
release
distribution
short video
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202210781713.0A
Other languages
Chinese (zh)
Other versions
CN114866845B (en
Inventor
杨爽
谢匡华
朱福青
谢匡亮
何春
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Changsha Meida Network Technology Co ltd
Original Assignee
Changsha Meida Network Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Changsha Meida Network Technology Co ltd filed Critical Changsha Meida Network Technology Co ltd
Priority to CN202210781713.0A priority Critical patent/CN114866845B/en
Publication of CN114866845A publication Critical patent/CN114866845A/en
Application granted granted Critical
Publication of CN114866845B publication Critical patent/CN114866845B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
    • H04N21/44213—Monitoring of end-user related data
    • H04N21/44222—Analytics of user selections, e.g. selection of programs or purchase activity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections

Abstract

The application provides an information detection method and system based on short video distribution, which can enable a first sample short video distribution event to realize feature learning of distribution interaction preference data on the basis of not acquiring the distribution interaction preference data from a page interaction analysis process, and fully perform model updating based on distribution attribute data distributed by the first sample short video distribution event and distribution interaction preference data responded by the distribution interaction process, so that the model updating reliability is improved, and a preference analysis deployment model generated based on parameter layer information after iterative optimization can accurately determine distribution preference relevance among different short video distribution events, so that the preference relevance analysis precision is improved. And after the model analysis feedback information is obtained, the first sample short video release event and the page interaction analysis process are respectively subjected to parameter layer information optimization, and the real-time data information of the short video release event and the release interaction process is effectively utilized for model training.

Description

Information detection method and system based on short video release
Technical Field
The application relates to the technical field of information detection, in particular to an information detection method and system based on short video publishing.
Background
Short videos are short videos, which are a mode for transmitting internet content, and are videos transmitted on new internet media within a short time (e.g., within 5 minutes); with the popularization of mobile terminals and the increasing speed of networks, short and fast mass flow transmission contents are gradually favored by various large platforms, fans and capital. In the related art, personalized short videos are pushed for users by analyzing short video interest behavior data of the users, however, in a short video distribution dimension, distribution preference relevance analysis aiming at different users is lacked, so that effective data content reference cannot be provided for the users in the short video distribution process.
Disclosure of Invention
In order to overcome at least the above disadvantages in the prior art, the present application aims to provide an information detection method and system based on short video distribution.
In a first aspect, the present application provides an information detection method based on short video distribution, which is applied to an information detection system based on short video distribution, and the method includes:
analyzing a first distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event based on a distribution preference analysis model; the first sample short video release event and the second sample short video release event both exist in a target short video service page, and the target short video service page is associated with a page interaction analysis process;
obtaining a second publishing preference correlation between the first sample short video publishing event and the second sample short video publishing event from the page interaction analysis process, wherein the second publishing preference correlation is output by the page interaction analysis process based on a publishing interaction analysis model analysis;
generating corresponding model analysis feedback information based on the first publishing preference relevance and the second publishing preference relevance, and iteratively optimizing parameter layer information of the publishing preference analysis model based on the model analysis feedback information; the model analysis feedback information is also configured to the page interaction analysis process to iteratively optimize the parameter layer information of the release interaction analysis model; generating a preference analysis deployment model based on parameter layer information after iterative optimization of a release preference analysis model and parameter layer information after iterative optimization of a release interaction analysis model, wherein the preference analysis deployment model is configured to analyze release preference correlation among different short video release events;
analyzing the release preference correlation between a first target short video release event and a second target short video release event which are loaded randomly based on the preference analysis deployment model, and determining the attraction degree between a first target user corresponding to the first target short video release event and a second target user corresponding to the second target short video release event based on the release preference correlation so as to obtain the attraction degree between each first target user and other second target users;
and based on the attraction degree between each first target user and each other second target user, sharing data content to each other second target user when the first target user initiates a short video publishing event next time.
In a second aspect, the present application further provides an information detection system based on short video distribution, where the information detection system based on short video distribution includes a processor and a machine-readable storage medium, where a computer program is stored in the machine-readable storage medium, and the computer program is loaded and executed in conjunction with the processor to implement the short video distribution-based information detection method of the first aspect.
By adopting the technical scheme of any aspect, in the application, after determining the first distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event based on the distribution preference analysis model and the distribution attribute data distributed by the first sample short video distribution event, the first distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event can be obtained from the page interaction analysis process, so that the feature learning of the distribution interaction preference data is realized on the basis that the distribution interaction preference data is not obtained from the page interaction analysis process by the first sample short video distribution event. After the first publishing preference relevance and the second publishing preference relevance are obtained, the first sample short video publishing event can generate corresponding model analysis feedback information based on the obtained first publishing preference relevance and the second publishing preference relevance, so that parameter layer information of a publishing preference analysis model can be processed according to the model analysis feedback information, and parameter layer information of the publishing interaction analysis model can be optimized by a page interaction analysis process, so that the first sample short video publishing event can be fully subjected to model updating based on publishing attribute data, event data and publishing interaction preference data responded by the publishing interaction process, the model updating reliability is improved, and the publishing preference relevance among different short video publishing events can be accurately determined based on a preference analysis deployment model generated based on the parameter layer information after iterative optimization, and the preference relevance analysis precision is improved. And after the model analysis feedback information is obtained, the first sample short video release event and the page interaction analysis process are respectively subjected to parameter layer information optimization, and the real-time data information of the short video release event and the release interaction process is effectively utilized for model training.
Drawings
In order to more clearly explain the technical solutions of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and it will be apparent to those skilled in the art that other related drawings can be obtained by combining these drawings without inventive effort.
Fig. 1 is a schematic flowchart of an information detection method based on short video distribution according to an embodiment of the present application.
Fig. 2 is a schematic block diagram of a structure of an information detection system based on short video distribution, which is provided in an embodiment of the present application and is used for implementing the short video distribution-based information detection method.
Detailed Description
The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a particular application and its requirements. It will be apparent to those of ordinary skill in the art that various changes can be made to the disclosed embodiments and that the general principles defined in this application can be applied to other embodiments and applications without departing from the principles and scope of the application. Thus, the present application is not limited to the described embodiments, but should be accorded the widest scope consistent with the claims.
The terminology used in the description presented herein is for the purpose of describing particular example embodiments only and is not intended to limit the scope of the present application. As used herein, the singular forms "a", "an" and "the" may include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, components, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, and/or groups thereof.
These and other features and characteristics of the present application, as well as the methods of operation and functions of the related elements of structure and the combination of parts and economies of manufacture, will become more apparent upon consideration of the following description of the drawings, all of which form a part of this specification. It is to be understood, however, that the drawings are designed solely for the purposes of illustration and description and are not intended as a definition of the limits of the application. It should be understood that the drawings are not to scale.
Flow charts are used herein to illustrate operations performed by systems incorporating some embodiments of the present application. It should be understood that the operations in the flow diagrams may be performed out of order. Rather, various steps may be processed on a reverse order basis or simultaneously. Further, one or more other operations may be added to the flowchart. One or more operations may also be deleted from the flowchart.
The present application is described in detail below with reference to fig. 1 and 2 of the drawings, and the specific operation method in the embodiments of the present application can also be applied to the embodiments of the apparatus or the embodiments of the system; as shown in fig. 1, fig. 1 is a schematic flowchart of an information detection method based on short video distribution according to an embodiment of the present application, where the information detection method based on short video distribution includes: step S101, analyzing a first distribution preference correlation between a first sample short video distribution event and a second sample short video distribution event based on a distribution preference analysis model; the first sample short video release event and the second sample short video release event both exist in a target short video service page, and the target short video service page is associated with a page interaction analysis process.
The page interaction analysis process is mainly used for processing the issuing interaction preference data responded by the issuing interaction process.
In some exemplary design ideas, before updating the weight information of the publishing preference analysis model and the weight information of the publishing interaction analysis model respectively, first, in a loading stage, loading data is input into a first sample short video publishing event and a page interaction analysis process respectively, in some exemplary design ideas, the loading data input into the first sample short video publishing event is publishing attribute data of the first sample short video publishing event, and the publishing attribute data mainly comprises two parts, namely publishing element information responded by the first sample short video publishing event and publishing configuration data of the first sample short video publishing event. In addition to the publishing attribute data published by the first sample short video publishing event, publishing interaction preference data for publishing an interaction process response is also included. Based on the distribution attribute data distributed by the first sample short video distribution event, the first sample short video distribution event can iteratively update the weight information of the distribution preference analysis model based on the distribution attribute data, and based on the distribution interaction preference data responded by the distribution interaction process, the iterative update of the weight information of the distribution interaction analysis model based on the distribution interaction preference data can be realized.
In some exemplary design considerations, the release interaction analysis model and the release preference analysis model may be collectively referred to as a federal learning model (i.e., a preference analysis deployment model).
In some exemplary design ideas, based on the publishing attribute data obtained by performing the request response on the first sample short video publishing event, the first sample short video publishing event may trigger a data loading process of a publishing preference analysis model to perform iterative update on parameter layer information of the publishing preference analysis model in the following, and in the data loading process of the publishing preference analysis model, the first sample short video publishing event may determine a first publishing preference correlation between the first sample short video publishing event and a second sample short video publishing event, and then may iteratively update the parameter layer information of the publishing preference analysis model based on the first publishing preference correlation. For example, when the first sample short video distribution event analyzes the first distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event based on the distribution preference analysis model, after the distribution attribute data distributed by the first sample short video distribution event is acquired, the parameter layer information of the distribution preference analysis model is initialized, and the initial parameter layer information of the distribution preference analysis model is determined.
In some exemplary design considerations, the publishing interaction preference data of the publishing interaction process response determines a second publishing preference association between the first sample short video publishing event and the second sample short video publishing event in the data recording process of the publishing interaction analysis model, and then, based on the determination of the second publishing preference association, the first sample short video publishing event may obtain the second publishing preference association, so that the federated learning model is iteratively updated based on the first publishing preference association and the second publishing preference association. The data recording process of the issuing preference analysis model and the data recording process of the issuing interaction analysis model have no necessary processing sequence relation.
Step S102, obtaining a second release preference relevance between the first sample short video release event and the second sample short video release event from the page interaction analysis process, wherein the second release preference relevance is analyzed and output by the page interaction analysis process based on the release interaction analysis model.
The page interaction analysis process is an analysis model for a publishing interaction process, and in some exemplary design ideas, the number of the publishing interaction processes configured for a target short video service page is one or more, and then the page interaction analysis process analyzes second publishing preference relevance of publishing interaction preference data based on response of each publishing interaction process, so that the second publishing preference relevance between a first sample short video publishing event and a second sample short video publishing event is finally determined based on an analysis result of the second publishing preference relevance corresponding to each roadside publishing interaction process. For example, when determining the second distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event, the page interaction analysis process first initializes the parameter layer information of the distribution interaction analysis model, and determines the starting parameter layer information of the distribution interaction analysis model, where the starting parameter layer information of the distribution interaction analysis model can be represented by HE, 0 and satisfies HE, 0= [1, …, 1] T, and then the page interaction analysis process can determine the second distribution preference correlation based on the starting parameter layer information HE, 0 and the corresponding distribution interaction preference data DS, where the second distribution preference correlation is provided. In some exemplary design approaches, on the basis that the target short video service page includes multiple (assumed to be n) publishing interaction processes, the initial parameter layer information of the publishing interaction analysis model may be represented as HE, i, 0, and HE, i, 0= [1, …, 1] T, (i, …, n), in other words, the publishing interaction preference data is obtained by one or more publishing interaction processes in response to the request, and may be specifically represented as DS, (i, …, n), and the number of initial parameter layer information of the publishing interaction analysis model is the same as the number of publishing interaction processes, and any initial parameter layer information of the publishing interaction analysis model represents the influence weight of the publishing interaction preference data corresponding to the publishing interaction process.
When the number of issuing interactive processes contained in the target short video service page is n, and when the page interaction analysis process determines the second issuing preference relevance based on the initial parameter layer information and the responding issuing interactive preference data, the issuing interactive preference data responded by any issuing interactive process and the corresponding initial parameter layer information can be obtained first, and then the issuing interactive preference data responded by the corresponding issuing interactive processes can be fused based on the corresponding initial parameter layer information to determine the second issuing preference relevance.
After the page interaction analysis process determines the second release preference correlation between the first sample short video release event and the second sample short video release event based on the release interaction preference data responded by the n release interaction processes configured in the target short video service page, the page interaction analysis process can also determine the initial calculation parameter value of the corresponding training cost calculation layer of the release interaction analysis model.
In the page interaction analysis process, based on the issued interaction preference data responded by n issued interaction processes in the target short video service page and the initial parameter layer information HE, 0 of the issued interaction analysis model, obtaining the second issued preference correlation between the first sample short video issued event and the second sample short video issued event and the initial calculation parameter value LR of the issued interaction analysis model corresponding to the training cost calculation layer, after 0, the second issued preference correlation and the initial calculation parameter value LR and 0 of the issued interaction analysis model corresponding to the training cost calculation layer can be coded and uploaded to the online service platform, so that the first sample short video issued event can be used in the model training process of the online service platform, the coded data can be obtained from the online service platform, and the initial calculation parameter value LR of the second issued preference correlation and the issued interaction analysis model corresponding to the training cost calculation layer can be realized, and (5) acquiring 0.
In some exemplary design ideas, when the page interaction analysis process uploads the initial calculation parameter values LR, 0 of the training cost calculation layer corresponding to the second publishing preference correlation and publishing interaction analysis model to the online service platform, the initial calculation parameter values LR, 0 of the training cost calculation layer corresponding to the second publishing preference correlation and publishing interaction analysis model may be uploaded separately or the page interaction analysis process may upload the initial calculation parameter values LR, 0 of the training cost calculation layer corresponding to the second publishing preference correlation and publishing interaction analysis model in a joint coding manner, some exemplary design ideas are not limited, and the first sample short video publishing event is not limited when obtaining the second publishing preference correlation between the first sample short video publishing event and the second sample short video publishing event from the page interaction analysis process, the encoded data can be obtained from the page interaction analysis process through the online service platform, the encoded data can be decoded, and the second release preference relevance can be obtained from the decoded data. In some exemplary design ideas, based on processing of encoded data, an initial calculation parameter value of the issuing interactive analysis model corresponding to the training cost calculation layer may also be obtained.
After the first sample short video distribution event obtains the first distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event, which are analyzed and output by the distribution preference analysis model, and the second distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event, which are analyzed and output by the distribution interaction analysis model, the model analysis feedback information for the federal learning model can be generated based on the first distribution preference correlation and the second distribution preference correlation, and further, the parameter layer information can be updated and optimized based on the model analysis feedback information, that is, step S103 is executed in turn.
Step S103, generating corresponding model analysis feedback information based on the first release preference relevance and the second release preference relevance, and updating parameter layer information of the release preference analysis model based on the model analysis feedback information; the model analysis feedback information is also configured in the page interaction analysis process to update the parameter layer information of the issuing interaction analysis model.
And step S104, generating a preference analysis deployment model based on the parameter layer information after the iterative optimization of the release preference analysis model and the parameter layer information after the iterative optimization of the release interaction analysis model, wherein the preference analysis deployment model is configured to analyze the release preference correlation among different short video release events.
In some exemplary design ideas, the first sample short video distribution event may implement, at a parameter layer information updating stage, model analysis feedback information generated based on the first distribution preference relevance and the second distribution preference relevance to update the parameter layer information, where when the first sample short video distribution event generates the model analysis feedback information based on the first distribution preference relevance and the second distribution preference relevance, the target distribution preference relevance may be obtained first, and in some exemplary design ideas, the target distribution preference relevance refers to: and determining model analysis feedback information based on the first distribution preference relevance, the second distribution preference relevance and the target distribution preference relevance after the first sample short video distribution event obtains the target distribution preference relevance.
After model analysis feedback information is obtained from a first sample short video release event, an iterative optimization process of a federal learning model is carried out based on the model analysis feedback information, namely, the process of respectively optimizing parameter layer information of a release preference analysis model and the process of optimizing the parameter layer information of a release interaction analysis model based on the model analysis feedback information. In some exemplary design ideas, the parameter layer information after the iterative optimization of the release preference analysis model and the parameter layer information after the iterative optimization of the release interaction analysis model are determined based on the iterative optimization of at least one training phase, and the model analysis feedback information is used for updating the parameter layer information of the release preference analysis model in the current training phase and is configured in the page interaction analysis process to update the parameter layer information of the release interaction analysis model; when the first sample short video publishing event updates and publishes the parameter layer information of the preference analysis model in the first round based on the model analysis feedback information and enables the page interaction analysis process to update and publish the parameter layer information of the interaction analysis model, the initial parameter layer information of the preference analysis model can be updated and published based on the model analysis feedback information after the initial parameter layer information Wc of the preference analysis model is obtained and 0; and transmitting the model analysis feedback information to the page interaction analysis process so that the page interaction analysis process updates and releases initial parameter layer information of the interaction analysis model based on the model analysis feedback information.
In some exemplary design considerations, when the first sample short video publishing event updates the initial parameter layer information of the publishing preference analysis model based on the model analysis feedback information, the feedback information may be analyzed based on the model to obtain an optimized vector value when optimizing the initial parameter layer information Wc, 0 of the publishing preference analysis model. In a specific implementation, the first sample short video publishing event may first obtain a current global training cost parameter and a corresponding first optimization vector function when optimizing initial parameter layer information of a publishing preference analysis model, then may perform back propagation on the first optimization vector function based on the current global training cost parameter, and determine an optimization vector value when optimizing the initial parameter layer information of the publishing preference analysis model according to model analysis feedback information. In some exemplary design ideas, based on the federal learning model, since the federal learning model includes a release preference analysis model and a release interaction analysis model, then, a global training cost parameter corresponding to the federal learning model can be constructed and obtained based on a training cost calculation layer of the release preference analysis model, a training cost calculation layer of the release interaction analysis model, and an interaction training cost calculation layer between the release preference analysis model and the release interaction analysis model.
Based on the training cost calculation layer, the current global training cost parameter obtained by the first sample short video distribution event after the first iteration optimization can be represented by L0, the current global training cost parameter can be obtained by the training cost calculation layer of the first sample short video distribution event based on the distribution preference analysis model, and according to the initial parameter layer information of the distribution preference analysis model and the distribution attribute data of the first sample short video distribution event, the current training cost parameter of the distribution preference analysis model is determined, the current training cost parameter of the distribution interaction analysis model is obtained from the page interaction analysis process, and based on the second distribution preference correlation obtained from the page interaction analysis process, the initial parameter layer information of the distribution preference analysis model and the distribution attribute data of the first sample short video distribution event, the current collaborative cost parameter between the distribution preference analysis model and the distribution interaction analysis model is determined, and determining a current global training cost parameter based on the current training cost parameter of the release preference analysis model, the current training cost parameter of the release interaction analysis model and the current collaborative training cost parameter.
Based on the current global training cost parameter obtained by the first sample short video publishing event, because the training cost calculation layer is subjected to derivation processing, a corresponding optimized vector function can be obtained, then, based on the optimized vector function D obtained by the derivation processing performed by the global training cost calculation layer L, the first sample short video publishing event can obtain an optimized vector value D0 for the initial parameter layer information of the publishing preference analysis model, and then, the first sample short video publishing event can be based on the optimized vector value of the initial parameter layer information of the publishing preference analysis model and optimize the initial parameter layer information of the publishing preference analysis model based on the first training direction parameter. In some exemplary design ideas, an optimized vector function for the initial parameter layer information of the release preference analysis model, which is obtained by performing derivation processing on the global training cost parameter L, may be represented.
Based on the optimization vector function aiming at the initial parameter layer information of the distribution preference analysis model and the distribution attribute data distributed by the first sample short video distribution event, the optimization vector value corresponding to the initial parameter layer information of the distribution preference analysis model can be obtained. The optimization vector value corresponding to the initial parameter layer information of the distribution preference analysis model, which is obtained by the first sample short video distribution event, may also be referred to as an initial optimization vector value of the distribution preference analysis model, and then the computer device may further obtain the training direction parameter, so that the parameter layer information of the distribution preference analysis model may be optimized based on an optimization vector descent method. In some exemplary design ideas, when the first sample short video publishing event is optimized based on an optimization vector descent method, the parameter layer information of the publishing preference analysis model may be optimized based on the training direction parameters obtained by the above-mentioned sum.
After iterative optimization is carried out on initial parameter layer information of a release preference analysis model, the first sample short video release event can obtain intermediate parameter layer information of the release preference analysis model determined after the initial parameter layer information of the release preference analysis model is updated, and model analysis feedback information is optimized based on the intermediate parameter layer information; and then, the parameter layer information and the model analysis feedback information of the release preference analysis model can be iteratively updated until the parameter layer information after the release preference analysis model is iteratively optimized is obtained.
In some exemplary design concepts, the publishing interaction analysis model includes an interaction preference analysis unit configured to be deployed in any one publishing interaction process, the number of publishing interaction processes is one or more, then, when the number of the publishing interaction processes is plural, the page interaction analysis process updates the initial parameter layer information of the publishing interaction analysis model based on the model analysis feedback information in such a manner that, first, the page interaction analysis process will obtain the model analysis feedback information D0 determined by the first sample short video distribution event from the online service platform, and then, the page interaction analysis process can determine a reference optimization vector value when an interaction preference analysis unit in any issuing interaction process optimizes initial parameter layer information of the issuing interaction analysis model based on issuing interaction preference data and model analysis feedback information responded by any issuing interaction process. After obtaining the reference optimization vector value of the corresponding interaction preference analysis unit, each issuing interaction process can sum the reference optimization vector values corresponding to each issuing interaction process to determine the optimization vector value of the initial parameter layer information for the issuing interaction analysis model.
After the page interaction analysis process obtains the optimized vector value of the initial parameter layer information for the issuing interaction analysis model, the optimized vector value of the initial parameter layer information for the issuing interaction analysis model can be based on the optimized vector value of the initial parameter layer information for the issuing interaction analysis model, and iterative optimization is performed on the initial parameter layer information for the issuing interaction analysis model based on a second training direction parameter, wherein the second training direction parameter can be the same as the first training direction parameter or different from the first training direction parameter, and the second training direction parameter can also be represented when the second training direction parameter is the same as the first training direction parameter. In some exemplary design ideas, the page interaction analysis process performs an iterative optimization process on initial parameter layer information of the issuing interaction analysis model based on the second training direction parameter, and also performs an optimization process on parameter layer information of the issuing interaction analysis model based on an optimization vector descent method.
In addition, when the parameter layer information of the issuing interactive analysis model is optimized based on the optimization vector (gradient) descent algorithm, after the parameter layer information corresponding to the issuing interactive analysis model after iterative optimization is obtained in the page interactive analysis process by referring to issuing interactive preference data responded by the corresponding issuing interactive process, the parameter layer information corresponding to the issuing interactive analysis model after iterative optimization can be sent to a first sample short video issuing event through a cloud, and after the first sample short video issuing event obtains the parameter layer information corresponding to the issuing interactive analysis model after iterative optimization, the preference analysis deployment model can be generated by combining the parameter layer information corresponding to the issuing preference analysis model after iterative optimization, namely step S104 can be executed.
In this embodiment, after determining the first distribution preference association between the first sample short video distribution event and the second sample short video distribution event based on the distribution attribute data distributed by the first sample short video distribution event and based on the distribution preference analysis model, the first sample short video distribution event may further obtain the second distribution preference association between the first sample short video distribution event and the second sample short video distribution event from the page interaction analysis process, so that the first sample short video distribution event realizes feature learning on the distribution interaction preference data on the basis that the distribution interaction preference data is not obtained from the page interaction analysis process. After the first publishing preference relevance and the second publishing preference relevance are obtained, the first sample short video publishing event can generate corresponding model analysis feedback information based on the obtained first publishing preference relevance and the second publishing preference relevance, so that parameter layer information of a publishing preference analysis model can be processed according to the model analysis feedback information, and parameter layer information of the publishing interaction analysis model can be optimized by a page interaction analysis process, so that the first sample short video publishing event can be fully subjected to model updating based on publishing attribute data, event data and publishing interaction preference data responded by the publishing interaction process, the model updating reliability is improved, and the publishing preference relevance among different short video publishing events can be accurately determined based on a preference analysis deployment model generated based on the parameter layer information after iterative optimization, and the preference relevance analysis precision is improved. And after the model analysis feedback information is obtained, the first sample short video release event and the page interaction analysis process are respectively subjected to parameter layer information optimization, and the real-time data information of the short video release event and the release interaction process is effectively utilized for model training.
Step S105, analyzing the distribution preference correlation between the arbitrarily-loaded first target short video distribution event and the second target short video distribution event based on the preference analysis deployment model, and determining the attraction degree between a first target user corresponding to the first target short video distribution event and a second target user corresponding to the second target short video distribution event based on the distribution preference correlation so as to obtain the attraction degree between each first target user and each other second target user.
For example, a mapping relationship between the relevance interval and the attraction degree corresponding to different relevance of the distribution preference may be set in advance, and the attraction degree between the first target user corresponding to the first target short video distribution event and the second target user corresponding to the second target short video distribution event may be determined based on the mapping relationship and the relevance of the distribution preference, so that the attraction degree between each first target user and each of the other second target users may be obtained. Wherein the attractiveness may be used to indicate the likelihood of a matching user representation between two users.
And step S106, based on the attraction degree between each first target user and each other second target user, sharing data content to each other second target user when the first target user initiates a short video release event next time.
For example, when the attraction degree between the first target user and any one of the other second target users is greater than the preset attraction degree, the data content can be shared with each of the other second target users when the first target user initiates a short video distribution event next time.
For example, the specific implementation manner of sharing the data content to the other second target users when the first target user initiates the short video distribution event next time may be implemented by the following steps.
Step S1061, obtaining user activity path data corresponding to the next time when the first target user initiates a short video release event;
step S1062, acquiring each publishing intention knowledge point of the short video publishing event based on the user activity path data;
step S1063, acquiring a plurality of online short video publishing element libraries, where one online short video publishing element library corresponds to one publishing intention knowledge point, and each online short video publishing element library includes a plurality of online short video publishing elements with different parameters.
Step S1064, executing the following steps for each release intention knowledge point: and acquiring one online short video publishing element in the corresponding online short video publishing element library based on the publishing intention knowledge point as a candidate short video publishing element, and constructing a content push network corresponding to the user activity path data based on each candidate short video publishing element so as to form a target content push network. The target content push network comprises candidate short video publishing elements which are connected through different video element logic relations.
Step S1065, sharing data content to the other second target users based on the target content push network.
Based on the steps, the online short video publishing elements basically matched with the publishing intention knowledge points corresponding to the user activity path data are obtained through the online short video publishing element library to serve as candidate short video publishing elements, and the influence of factors such as noise and the like caused by directly pushing the content by using the user activity path data is overcome.
For example, in some possible embodiments, the obtaining of the respective publishing intention knowledge points of the short video publishing event based on the user activity path data includes: acquiring a pre-trained release intention mining model; and loading the user activity path data into the pre-trained publishing intention mining model so as to obtain publishing intention fields output by the pre-trained publishing intention mining model, wherein one publishing intention field represents one online short video publishing element.
For example, in some possible embodiments, the release intent field includes a shared release intent knowledge point; when the release intention field is a shared release intention knowledge point, the acquiring each release intention knowledge point of the short video release event further comprises: acquiring data in user activity path data corresponding to the shared publishing intention knowledge point, wherein the data is called shared data; obtaining the shared service node information of the shared data; acquiring a preset service node database, wherein the service node database comprises a plurality of preset service node data and service node intention knowledge points corresponding to each service node data; analyzing whether shared service node information of the shared data and each preset service node data meet a first target requirement, wherein the first target requirement corresponds to a release intention knowledge point, and if so, acquiring the service node intention knowledge point corresponding to the preset service node data meeting the first target requirement as the release intention knowledge point of the shared release intention knowledge point.
For example, in some possible embodiments, the obtaining, as the candidate short video publishing element, one online short video publishing element in the corresponding online short video publishing element library based on the publishing intent knowledge point includes: calculating matching parameter values of the release intention knowledge points and each online short video release element in the online short video release element library corresponding to the release intention knowledge points respectively; and analyzing whether one matching parameter value exceeds a preset matching parameter value in the acquired matching parameter values, if so, acquiring an online short video publishing element corresponding to the matching parameter value exceeding the preset matching parameter value as a candidate short video publishing element of the publishing intention knowledge point.
For example, in some possible embodiments, the parameters of the online short video publishing element include video tagging parameters, and each online short video publishing element has a unique video tagging parameter in each online short video publishing element library; the obtaining of one online short video publishing element in the corresponding online short video publishing element library based on the publishing intention knowledge point as a candidate short video publishing element includes: acquiring a pre-trained short video publishing element mining model; loading the publishing intention knowledge points to the pre-trained short video publishing element mining model so as to obtain model mining information output by the pre-trained short video publishing element mining model, wherein the model mining information represents video tag parameters of the publishing intention knowledge points; and acquiring online short video publishing elements corresponding to the unique video tag parameters which are the same as the video tag parameters of the publishing intention knowledge points as candidate short video publishing elements.
For example, in some possible embodiments, when there is no online short video publishing element corresponding to the unique video tag parameter that is the same as the video tag parameter of the publishing intention knowledge point, the publishing intention knowledge point of the online short video publishing element that is not corresponding to the unique video tag parameter that is the same as the video tag parameter of the publishing intention knowledge point is called a fuzzy publishing intention knowledge point, and the obtaining, based on the publishing intention knowledge point, one online short video publishing element in the corresponding online short video publishing element library as the candidate short video publishing element further includes: decomposing fuzzy release intention knowledge points to obtain each release intention knowledge component; decomposing each online short video publishing element in an online short video publishing element library corresponding to the publishing intention knowledge point, thereby obtaining a preset publishing intention knowledge component of each online short video publishing element; performing the following steps for each of the publishing intent knowledge components: calculating matching parameter values of the release intention knowledge components and the preset release intention knowledge components; analyzing whether one matching parameter value is larger than a preset matching parameter value or not, and if so, acquiring a preset release intention knowledge component of which the matching parameter value is larger than the preset matching parameter value; and splicing the acquired preset publishing intention knowledge components to form candidate short video publishing elements of the fuzzy publishing intention knowledge points.
Fig. 2 schematically illustrates a short video distribution-based information detection system 100 that may be used to implement various embodiments described in the present application.
For one embodiment, fig. 2 shows a short video publication-based information detection system 100, the short video publication-based information detection system 100 having one or more processors 102, a control module 104 coupled to at least one of the processor(s) 102, a memory 106 coupled to the control module 104, an NVM/storage device 108 coupled to the control module 104, one or more input/output devices 110 coupled to the control module 104, and a network interface 112 coupled to the control module 104. Is a non-volatile memory storage device.
The processor 102 may include one or more single-core or multi-core processors, and the processor 102 may include any combination of general-purpose or special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In some embodiments, the short video distribution-based information detection system 100 can be a server device such as a gateway described in the embodiments of the present application.
In some embodiments, the short video distribution-based information detection system 100 may include one or more computer-readable media (e.g., memory 106 or NVM/storage 108) having instructions 114 and one or more processors 102 configured to execute the instructions 114 in conjunction with the one or more computer-readable media to implement modules to perform the actions described in this embodiment.
For one embodiment, control module 104 may include any suitable interface controller to provide any suitable interface to at least one of the processor(s) 102 and/or any suitable device or component in communication with control module 104.
Control module 104 may include a memory controller module to provide an interface to memory 106. The memory controller module may be a hardware module, a software module, and/or a firmware module.
The memory 106 may be used to load and store data and/or instructions 114, for example, for the short video distribution based information detection system 100. For one embodiment, memory 106 may comprise any suitable volatile memory, such as suitable DRAM. In some embodiments, the memory 106 may comprise a double data rate type four synchronous dynamic random access memory (DDR4 SDRAM).
For one embodiment, control module 104 may include one or more input/output controllers to provide an interface to NVM/storage 108 and one or more input/output devices 110.
For example, NVM/storage 108 may be used to store data and/or instructions 114. NVM/storage 108 may include any suitable non-volatile memory (e.g., flash memory) and/or may include any suitable non-volatile storage device(s) (e.g., one or more Hard Disk Drives (HDDs), one or more Compact Disc (CD) drives, and/or one or more Digital Versatile Disc (DVD) drives).
The NVM/storage 108 may include storage resources that are physically part of the device on which the short video distribution-based information detection system 100 is installed, or it may be accessible by the device and need not be part of the device. For example, NVM/storage 108 may be accessible via input/output device(s) 110 over a network.
The input/output device(s) 110 may provide an interface for the short video publication-based information detection system 100 to communicate with any other suitable device, and the input/output device 110 may include a communication component, a pinyin component, a sensor component, and so forth. The network interface 112 may provide an interface for the short video publication-based information detection system 100 to communicate in accordance with one or more networks, and the short video publication-based information detection system 100 may wirelessly communicate with one or more components of a wireless network in accordance with any of one or more wireless network standards and/or protocols, such as access to a communication standard-based wireless network, e.g., WiFi, 2G, 3G, 4G, 5G, etc., or a combination thereof.
For one embodiment, at least one of the processor(s) 102 may be packaged together with logic for one or more controllers (e.g., memory controller modules) of the control module 104. For one embodiment, at least one of the processor(s) 102 may be packaged together with logic for one or more controller(s) of the control module 104 to form a System In Package (SiP). For one embodiment, at least one of the processor(s) 102 may be integrated on the same die with logic for one or more controller(s) of the control module 104. For one embodiment, at least one of the processor(s) 102 may be integrated on the same die with logic of one or more controllers of the control module 104 to form a system on a chip (SoC).
In various embodiments, the short video distribution based information detection system 100 may be, but is not limited to: a server, a desktop computing device, or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet, a netbook, etc.), among other terminal devices. In various embodiments, the short video publication-based information detection system 100 may have more or fewer components and/or different architectures. For example, in some embodiments, the short video publication-based information detection system 100 includes one or more cameras, a keyboard, a Liquid Crystal Display (LCD) screen (including touch screen displays), a non-volatile memory port, a plurality of antennas, a graphics chip, an Application Specific Integrated Circuit (ASIC), and a speaker.
An embodiment of the present application provides an electronic device, including: one or more processors; and one or more machine readable media having instructions stored thereon that, when executed by the one or more processors, cause the electronic device to perform a data processing method as described in one or more of the present applications.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The embodiments in the present specification are all described in a progressive manner, and each embodiment focuses on differences from other embodiments, and portions that are the same and similar between the embodiments may be referred to each other.
Embodiments of the present application are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and the basis of a flow and/or block of the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including the preferred embodiment and all changes and modifications that fall within the true scope of the embodiments of the present application.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
The present application is described in detail above, and the principles and embodiments of the present application are described herein by using specific examples, which are only used to help understand the method and the core idea of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims (10)

1. An information detection method based on short video distribution is characterized in that the method is applied to an information detection system based on short video distribution, and comprises the following steps:
analyzing a first distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event based on a distribution preference analysis model; the first sample short video release event and the second sample short video release event both exist in a target short video service page, and the target short video service page is associated with a page interaction analysis process;
obtaining a second publishing preference correlation between the first sample short video publishing event and the second sample short video publishing event from the page interaction analysis process, wherein the second publishing preference correlation is output by the page interaction analysis process based on a publishing interaction analysis model analysis;
generating corresponding model analysis feedback information based on the first publishing preference relevance and the second publishing preference relevance, and iteratively optimizing parameter layer information of the publishing preference analysis model based on the model analysis feedback information; the model analysis feedback information is also configured to the page interaction analysis process to iteratively optimize the parameter layer information of the release interaction analysis model;
generating a preference analysis deployment model based on parameter layer information after iterative optimization of a release preference analysis model and parameter layer information after iterative optimization of a release interaction analysis model, wherein the preference analysis deployment model is configured to analyze release preference correlation among different short video release events;
analyzing the release preference correlation between a first target short video release event and a second target short video release event which are loaded randomly based on the preference analysis deployment model, and determining the attraction degree between a first target user corresponding to the first target short video release event and a second target user corresponding to the second target short video release event based on the release preference correlation so as to obtain the attraction degree between each first target user and other second target users;
and based on the attraction degree between each first target user and each other second target user, sharing data content to each other second target user when the first target user initiates a short video publishing event next time.
2. The short video distribution-based information detection method according to claim 1, wherein the parameter layer information after iterative optimization of the distribution preference analysis model and the parameter layer information after iterative optimization of the distribution interaction analysis model are determined after iterative optimization based on at least one training phase, and the model analysis feedback information is used for iterative optimization of the parameter layer information of the distribution preference analysis model in a current training phase and is configured to the parameter layer information of the distribution interaction analysis model in the page interaction analysis process; the step of iteratively optimizing the parameter layer information of the release preference analysis model in a first round based on the model analysis feedback information and iteratively optimizing the parameter layer information of the release interaction analysis model by the page interaction analysis process specifically includes:
acquiring initial parameter layer information of the release preference analysis model, and iteratively optimizing the initial parameter layer information of the release preference analysis model based on the model analysis feedback information;
transmitting the model analysis feedback information to the page interaction analysis process so that the page interaction analysis process iteratively optimizes initial parameter layer information of the issuing interaction analysis model based on the model analysis feedback information;
the step of iteratively optimizing the initial parameter layer information of the release preference analysis model based on the model analysis feedback information specifically includes:
based on the model analysis feedback information, obtaining an optimization vector value when the initial parameter layer information of the release preference analysis model is optimized currently;
optimizing the initial parameter layer information of the release preference analysis model based on an optimized vector value of the initial parameter layer information of the release preference analysis model and based on a first training direction parameter, wherein the first training direction parameter is used for expressing a step length of the optimized vector value moving to a training cost parameter optimal solution;
the step of obtaining an optimized vector value when optimizing the initial parameter layer information of the release preference analysis model at present based on the model analysis feedback information specifically includes:
acquiring a current global training cost parameter and a corresponding first optimization vector function when optimizing initial parameter layer information of the release preference analysis model;
performing back propagation on the first optimization vector function based on the current global training cost parameter, and determining an optimization vector value when the initial parameter layer information of the release preference analysis model is optimized currently according to the model analysis feedback information;
the step of obtaining the current global training cost parameter specifically includes:
determining a current training cost parameter of the release preference analysis model according to the initial parameter layer information of the release preference analysis model and the release attribute data released by the first sample short video release event based on a training cost calculation layer of the release preference analysis model;
acquiring a current training cost parameter of the issuing interactive analysis model from the page interactive analysis process;
determining a current collaborative training cost parameter between the release preference analysis model and the release interaction analysis model based on a second release preference relevance obtained from the page interaction analysis process, initial parameter layer information of the release preference analysis model, and release attribute data released by the first sample short video release event;
and determining a current global training cost parameter based on the current training cost parameter of the release preference analysis model, the current training cost parameter of the release interaction analysis model and the current collaborative training cost parameter.
3. The short video distribution-based information detection method according to claim 2, wherein the method further comprises:
obtaining intermediate parameter layer information of the release preference analysis model determined after iterative optimization of the initial parameter layer information of the release preference analysis model, and optimizing the model analysis feedback information based on the intermediate parameter layer information;
and iteratively updating the parameter layer information and the model analysis feedback information of the release preference analysis model until the parameter layer information after the iterative optimization of the release preference analysis model is obtained.
4. The short video distribution-based information detection method according to claim 2, wherein the distribution interaction analysis model includes an interaction preference analysis unit configured in any distribution interaction process, and when the number of distribution interaction processes is multiple, the page interaction analysis process iteratively optimizes initial parameter layer information of the distribution interaction analysis model based on the model analysis feedback information, specifically including:
the page interaction analysis process determines a reference optimization vector value when an interaction preference analysis unit in any one issuing interaction process optimizes initial parameter layer information of the issuing interaction analysis model based on issuing interaction preference data responded by any one issuing interaction process and the model analysis feedback information;
the page interaction analysis process sums the reference optimization vector values corresponding to each release interaction process, and determines the optimization vector values aiming at the initial parameter layer information of the release interaction analysis model;
and performing iterative optimization on the initial parameter layer information of the issuing interactive analysis model based on the optimized vector value of the initial parameter layer information of the issuing interactive analysis model and based on a second training direction parameter.
5. The short video distribution-based information detection method according to claim 1, wherein the step of obtaining a second distribution preference association between the first sample short video distribution event and the second sample short video distribution event from the page interaction analysis process specifically includes:
acquiring coded data from the page interaction analysis process, wherein the coded data is determined by the page interaction analysis process after determining second release preference relevance based on the responded release interaction preference data and the release interaction analysis model and encoding the second release preference relevance and the calculation parameter value of the training cost calculation layer corresponding to the release interaction analysis model after issuing the calculation parameter value of the training cost calculation layer corresponding to the interaction analysis model;
the encoded data is decoded and a second publication preference association is obtained from the decoded data.
6. The short video distribution-based information detection method according to claim 5, wherein the step of determining the second distribution preference relevance and the calculation parameter value of the distribution interaction analysis model corresponding to the training cost calculation layer specifically comprises:
the page interaction analysis process initializes the parameter layer information of the issuing interaction analysis model and determines the initial parameter layer information of the issuing interaction analysis model;
and the page interaction analysis process determines the relevance of the second release preference based on the initial parameter layer information of the release interaction analysis model and the responsive release interaction preference data, and the initial calculation parameter value of the corresponding training cost calculation layer of the release interaction analysis model.
7. The short video distribution-based information detection method according to claim 6, wherein the distribution interaction preference data is obtained by one or more distribution interaction processes performing request response, the number of initial parameter layer information of the distribution interaction analysis model is the same as the number of distribution interaction processes, and any initial parameter layer information of the distribution interaction analysis model represents an influence weight of distribution interaction preference data corresponding to the distribution interaction process response;
when the issuing interaction preference data is obtained by a plurality of issuing interaction process responses, the step of determining the second issuing preference relevance by the page interaction analysis process based on the initial parameter layer information of the issuing interaction analysis model and the issuing interaction preference data of the responses specifically includes:
the page interaction analysis process acquires the issuing interaction preference data responded by any issuing interaction process and corresponding initial parameter layer information;
and the page interaction analysis process fuses the issuing interaction preference data responded by the corresponding issuing interaction process based on the corresponding initial parameter layer information, and determines the relevance of the second issuing preference.
8. The short video distribution-based information detection method according to claim 1, wherein the step of analyzing the first distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event based on the distribution preference analysis model specifically comprises:
acquiring publishing attribute data published by the first sample short video publishing event, initializing parameter layer information of the publishing preference analysis model, and determining initial parameter layer information of the publishing preference analysis model;
and performing distribution preference correlation analysis on the distribution attribute data based on the initial parameter layer information of the distribution preference analysis model, and determining the first distribution preference correlation between the first sample short video distribution event and the second sample short video distribution event.
9. The method according to claim 1, wherein the step of generating model analysis feedback information based on the first distribution preference relevance and the second distribution preference relevance specifically includes:
obtaining target release preference relevance, wherein the target release preference relevance refers to the following steps: the issuing preference relevance is jointly determined by combining the issuing attribute data and the issuing preference analysis model and the issuing interactive preference data and the issuing interactive analysis model;
determining model analysis feedback information based on the first publishing preference relevance, the second publishing preference relevance and the target publishing preference relevance.
10. An information detection system based on short video distribution, characterized in that the information detection system based on short video distribution comprises a processor and a machine-readable storage medium, wherein the machine-readable storage medium stores machine-executable instructions, and the machine-executable instructions are loaded and executed by the processor to implement the information detection method based on short video distribution according to any one of claims 1 to 9.
CN202210781713.0A 2022-07-05 2022-07-05 Information detection method and system based on short video release Active CN114866845B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210781713.0A CN114866845B (en) 2022-07-05 2022-07-05 Information detection method and system based on short video release

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210781713.0A CN114866845B (en) 2022-07-05 2022-07-05 Information detection method and system based on short video release

Publications (2)

Publication Number Publication Date
CN114866845A true CN114866845A (en) 2022-08-05
CN114866845B CN114866845B (en) 2022-09-23

Family

ID=82625588

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210781713.0A Active CN114866845B (en) 2022-07-05 2022-07-05 Information detection method and system based on short video release

Country Status (1)

Country Link
CN (1) CN114866845B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160041998A1 (en) * 2014-08-05 2016-02-11 NFL Enterprises LLC Apparatus and Methods for Personalized Video Delivery
CN111768263A (en) * 2020-03-31 2020-10-13 北京京东尚科信息技术有限公司 Information pushing method and device, server and storage medium
CN112948708A (en) * 2021-03-05 2021-06-11 清华大学深圳国际研究生院 Short video recommendation method
CN113449146A (en) * 2021-07-06 2021-09-28 武汉丰华裕商务有限公司 Short video browsing recommendation method and device based on data analysis and computer storage medium
CN113766281A (en) * 2021-09-10 2021-12-07 北京快来文化传播集团有限公司 Short video recommendation method, electronic device and computer-readable storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160041998A1 (en) * 2014-08-05 2016-02-11 NFL Enterprises LLC Apparatus and Methods for Personalized Video Delivery
CN111768263A (en) * 2020-03-31 2020-10-13 北京京东尚科信息技术有限公司 Information pushing method and device, server and storage medium
CN112948708A (en) * 2021-03-05 2021-06-11 清华大学深圳国际研究生院 Short video recommendation method
CN113449146A (en) * 2021-07-06 2021-09-28 武汉丰华裕商务有限公司 Short video browsing recommendation method and device based on data analysis and computer storage medium
CN113766281A (en) * 2021-09-10 2021-12-07 北京快来文化传播集团有限公司 Short video recommendation method, electronic device and computer-readable storage medium

Also Published As

Publication number Publication date
CN114866845B (en) 2022-09-23

Similar Documents

Publication Publication Date Title
WO2023061087A1 (en) Information recommendation method and apparatus, and electronic device, computer-readable storage medium and computer program product
CN114896454B (en) Short video data recommendation method and system based on label analysis
US20170185894A1 (en) End-to-end deep collaborative filtering
WO2019114423A1 (en) Method and apparatus for merging model prediction values, and device
WO2022041982A1 (en) Data recommendation method and apparatus, computer device, and storage medium
CN109981785B (en) Method and device for pushing information
CN110909942B (en) Method and system for training model and method and system for predicting sequence data
CN111783810B (en) Method and device for determining attribute information of user
CN109117442B (en) Application recommendation method and device
CN111340220A (en) Method and apparatus for training a predictive model
CN112182281B (en) Audio recommendation method, device and storage medium
KR102543838B1 (en) Method and appartus for obtaining information
WO2024051707A1 (en) Recommendation model training method and apparatus, and resource recommendation method and apparatus
CN114866845B (en) Information detection method and system based on short video release
CN112218114B (en) Video cache control method, device and computer readable storage medium
CN109657153B (en) Method and equipment for determining associated financial information of user
CN112308644A (en) Method and device for processing description information
CN112905885B (en) Method, apparatus, device, medium and program product for recommending resources to user
CN114926234A (en) Article information pushing method and device, electronic equipment and computer readable medium
JP5325280B2 (en) Information processing device
CN113298083A (en) Data processing method and device
CN112685516A (en) Multi-channel recall recommendation method and device, electronic equipment and medium
CN113554450A (en) Data model training and data processing method, device, equipment and storage medium
CN111915339A (en) Data processing method, device and equipment
CN116561735B (en) Mutual trust authentication method and system based on multiple authentication sources and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant