CN112165639B - Content distribution method, device, electronic equipment and storage medium - Google Patents

Content distribution method, device, electronic equipment and storage medium Download PDF

Info

Publication number
CN112165639B
CN112165639B CN202011009503.7A CN202011009503A CN112165639B CN 112165639 B CN112165639 B CN 112165639B CN 202011009503 A CN202011009503 A CN 202011009503A CN 112165639 B CN112165639 B CN 112165639B
Authority
CN
China
Prior art keywords
content
distributed
interaction
information
timeliness
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.)
Active
Application number
CN202011009503.7A
Other languages
Chinese (zh)
Other versions
CN112165639A (en
Inventor
朱朝悦
衡阵
马连洋
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.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen 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 Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN202011009503.7A priority Critical patent/CN112165639B/en
Publication of CN112165639A publication Critical patent/CN112165639A/en
Application granted granted Critical
Publication of CN112165639B publication Critical patent/CN112165639B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/262—Content or additional data distribution scheduling, e.g. sending additional data at off-peak times, updating software modules, calculating the carousel transmission frequency, delaying a video stream transmission, generating play-lists
    • H04N21/26208—Content or additional data distribution scheduling, e.g. sending additional data at off-peak times, updating software modules, calculating the carousel transmission frequency, delaying a video stream transmission, generating play-lists the scheduling operation being performed under constraints
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
    • H04L65/60—Network streaming of media packets
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H04L67/1004—Server selection for load balancing
    • H04L67/1014—Server selection for load balancing based on the content of a request
    • 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
    • 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
    • H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
    • 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]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
    • H04N21/84—Generation or processing of descriptive data, e.g. content descriptors

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Databases & Information Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

The application discloses a content distribution method, a device, an electronic device and a storage medium, comprising: acquiring associated information of content to be distributed, wherein the associated information of the content to be distributed comprises text description information corresponding to a plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed; extracting features of the collected content interaction information to obtain interaction features corresponding to the collected content interaction information; predicting content timeliness of the content to be distributed according to semantics of the text description information and interaction characteristics corresponding to the collected content interaction information; the content to be distributed is distributed based on the predicted content timeliness, and the content to be distributed can be accurately pushed to the target user within the validity period of the content timeliness, so that the accuracy of content distribution is improved.

Description

Content distribution method, device, electronic equipment and storage medium
Technical Field
The present invention relates to the field of computer technologies, and in particular, to a content distribution method, a device, an electronic apparatus, and a storage medium.
Background
With the development of modern technology, the way in which media release information is more and more convenient. Such media may register an account number on the network platform and then publish information based on the account number, such as text information, audio information, video information, and the like. These media also include self-media, which refers to the fact that the general public publishes themselves and the way news propagates through networks and the like. In recent years, the wind gap for content creation is adopted, all large internet companies actively enter the content market, various self-media are emerging like spring bamboo shoots after rain, and people can create self-media by writing. A huge number of self-media creates a huge number of articles every day, but some content published by the self-media account may be copied from the self-media platform or the original content of the self-media account may be reprocessed and patched, so that the content distributed by the self-media account needs to be checked.
At present, a manual auditing scheme is adopted to audit the content released by the media account, however, due to the huge number of the media account and limited by manpower and auditing time, part of highly-timeliness content may not have been audited for a period of time, so that the content to be distributed cannot be accurately pushed to a target user in the period of time-timeliness of the content in the current content distribution scheme.
Disclosure of Invention
The content distribution method, the device, the electronic equipment and the storage medium can accurately push the content to be distributed to the target user in the validity period of the timeliness of the content, and the accuracy of content distribution is improved.
The application provides a content distribution method, which comprises the following steps:
acquiring associated information of content to be distributed, wherein the associated information of the content to be distributed comprises text description information corresponding to a plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed;
extracting features of the collected content interaction information to obtain interaction features corresponding to the collected content interaction information;
predicting content timeliness of the content to be distributed according to semantics of the text description information and interaction characteristics corresponding to the collected content interaction information;
and distributing the content to be distributed based on the predicted content timeliness.
Correspondingly, the application also provides a content distribution device, which comprises:
the system comprises an acquisition module, a distribution module and a distribution module, wherein the acquisition module is used for acquiring the associated information of the content to be distributed, and the associated information of the content to be distributed comprises text description information corresponding to a plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed;
The extraction module is used for extracting the characteristics of the collected content interaction information to obtain the interaction characteristics corresponding to the collected content interaction information;
the prediction module is used for predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the collected content interaction information;
and the distribution module is used for distributing the content to be distributed based on the predicted content timeliness.
Optionally, in some embodiments of the present application, the prediction module includes:
a first extraction submodule block, configured to extract a content tag of the content to be distributed from the text description information;
the second extraction sub-module is used for extracting the characteristics of the content label to obtain the label characteristics corresponding to the content label;
and the prediction sub-module is used for predicting the content timeliness of the content to be distributed based on the semantics and the tag characteristics of the content tags and the interaction characteristics corresponding to the acquired content interaction information.
Optionally, in some embodiments of the present application, the call prediction submodule includes:
the word segmentation unit is used for segmenting the label text corresponding to the content label based on the semantic meaning of the content label;
The segmentation unit is used for segmenting the tag features according to the word segmentation result to obtain element features corresponding to each text element in the tag text;
and the predicting unit is used for predicting the content timeliness of the content to be distributed based on the element characteristics corresponding to each text element in the tag text and the interaction characteristics corresponding to the acquired content interaction information.
Optionally, in some embodiments of the present application, the prediction unit includes:
the acquisition subunit is used for acquiring a preset timeliness prediction model;
and the prediction subunit is used for predicting the content timeliness of the content to be distributed through the timeliness prediction model based on element characteristics corresponding to each text element in the tag text and interaction characteristics corresponding to the acquired content interaction information.
Optionally, in some embodiments of the present application, the prediction subunit is specifically configured to:
combining the characteristics of the interaction characteristics corresponding to the collected content interaction information to obtain combined characteristics;
predicting the attention degree of the user of the content to be distributed in different preset time periods according to the combination characteristics to obtain a first attention degree corresponding to each preset time period;
Predicting the user attention degree of the content to be distributed in different preset time periods according to element characteristics corresponding to each text element in the tag text, and obtaining a second attention degree corresponding to each preset time period;
and fusing the first attention degree and the second attention degree corresponding to each preset period to obtain the content timeliness of the content to be distributed in each preset period.
Optionally, in some embodiments of the present application, the method further includes an identification module, where the identification module is specifically configured to:
identifying the intention of the collected content interaction information;
according to the intention recognition result, determining the interaction information corresponding to the intention as the forward intention as forward interaction information, and;
determining the interaction information corresponding to the intention as the negative intention as negative interaction information;
reserving the forward interaction information;
the prediction module is specifically configured to: and predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the forward interaction information.
Optionally, in some embodiments of the present application, the distribution module is specifically configured to:
ordering the content to be distributed in the associated information of the content to be distributed based on the predicted content timeliness to obtain a content list to be distributed;
And distributing the content to be distributed in the content list to be distributed according to the sequence of the content to be distributed in the content list to be distributed.
After the associated information of the content to be distributed is collected, the associated information of the content to be distributed comprises text description information corresponding to a plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed, the collected content interaction information is subjected to feature extraction to obtain interaction features corresponding to the collected content interaction information, then, according to the semantics of the text description information and the interaction features corresponding to the collected content interaction information, the content timeliness of the content to be distributed is predicted, and finally, the content to be distributed is distributed based on the predicted content timeliness. Therefore, the scheme can accurately push the content to be distributed to the target user in the validity period of the timeliness of the content, and improves the accuracy of content distribution.
Drawings
In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1a is a schematic view of a scenario of a content distribution method provided herein;
FIG. 1b is a flow chart of a content distribution method provided herein;
FIG. 1c is a schematic structural diagram of the time-dependent predictive model provided herein;
FIG. 2a is another flow chart of the content distribution method provided herein;
FIG. 2b is another schematic diagram of the time-dependent predictive model provided herein;
FIG. 3a is a schematic view of the structure of the content distribution apparatus provided in the present application;
FIG. 3b is another schematic structural view of the content distribution apparatus provided in the present application;
fig. 4 is a schematic structural diagram of an electronic device provided in the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention.
Artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use the knowledge to obtain optimal results. In other words, artificial intelligence is an integrated technology of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence, i.e. research on design principles and implementation methods of various intelligent machines, enables the machines to have functions of sensing, reasoning and decision.
The artificial intelligence technology is a comprehensive subject, and relates to the technology with wide fields, namely the technology with a hardware level and the technology with a software level. Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.
Natural language processing (Nature Language processing, NLP) is an important direction in the fields of computer science and artificial intelligence. It is studying various theories and methods that enable effective communication between a person and a computer in natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Thus, the research in this field will involve natural language, i.e. language that people use daily, so it has a close relationship with the research in linguistics. Natural language processing techniques typically include text processing, semantic understanding, machine translation, robotic questions and answers, knowledge graph techniques, and the like.
Machine Learning (ML) is a multi-domain interdisciplinary, involving multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. It is specially studied how a computer simulates or implements learning behavior of a human to acquire new knowledge or skills, and reorganizes existing knowledge structures to continuously improve own performance. Machine learning is the core of artificial intelligence, a fundamental approach to letting computers have intelligence, which is applied throughout various areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, confidence networks, reinforcement learning, transfer learning, induction learning, teaching learning, and the like.
With research and advancement of artificial intelligence technology, research and application of artificial intelligence technology is being developed in various fields, such as common smart home, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned, automatic driving, unmanned aerial vehicles, robots, smart medical treatment, smart customer service, etc., and it is believed that with the development of technology, artificial intelligence technology will be applied in more fields and with increasing importance value.
The scheme provided by the embodiment of the application relates to artificial intelligence natural language processing, machine learning and other technologies, and is specifically described through the following embodiments.
The application provides a content distribution method, a content distribution device, an electronic device and a storage medium.
The content distribution device can be integrated in a server, wherein the server can be an independent physical server, can be a server cluster or a distributed system formed by a plurality of physical servers, and can also be a cloud server for providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs, basic cloud computing services such as big data and artificial intelligent platforms. The terminal may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, etc. The terminal and the server may be directly or indirectly connected through wired or wireless communication, which is not limited herein.
For example, referring to fig. 1a, the content distribution apparatus is integrated on a server, the server collects related information of content to be distributed, where the related information of the content to be distributed includes text description information corresponding to a plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed, the text description information is text information that can be used to describe a type of the content to be distributed, for example, a title of the content to be distributed, etc., the content interaction information may be information generated after a user performs an interaction action with the distribution account in a historical period, for example, the released content under the distribution account is endorsed, etc., then the server may perform feature extraction on the collected content interaction information to obtain interaction features corresponding to the collected content interaction information, then predict content timeliness of the content to be distributed according to semantics of the text description information and interaction features corresponding to the collected content interaction information, and finally, the server performs content distribution on the content to be distributed based on the predicted content timeliness.
According to the content distribution method, the content timeliness of the content to be distributed can be predicted according to the content interaction information and the text description information, the content timeliness of the content to be distributed is predicted through different dimensions, the problem that the predicted content timeliness is inaccurate due to the fact that the content timeliness of the content to be distributed is predicted through the information of the single dimension is avoided, and therefore the content to be distributed can be accurately pushed to a target user within the validity period of the content timeliness, and the accuracy of content distribution is improved.
The following will describe in detail. It should be noted that the following description order of embodiments is not a limitation of the priority order of embodiments.
A content distribution method, comprising: collecting associated information of the content to be distributed, extracting features of the collected content interaction information to obtain interaction features corresponding to the collected content interaction information, predicting content timeliness of the content to be distributed according to semantics of text description information and the interaction features corresponding to the collected content interaction information, and distributing the content to be distributed based on the predicted content timeliness.
Referring to fig. 1b, fig. 1b is a flow chart of a content distribution method provided in the present application. The specific flow of the content distribution method can be as follows:
101. and collecting the associated information of the content to be distributed.
The associated information of the content to be distributed comprises text description information corresponding to the plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed. It may be understood that each content to be distributed has corresponding text description information, where the text description information may include a content text corresponding to the content to be distributed, a content title of the content to be distributed, and a classification tag of the content to be distributed. In the application, a content title of a content to be distributed and a classification label of the content to be distributed are collectively called a content label, a distribution account refers to an account authenticated by a content distribution system (also called a content distribution platform), the distribution account is an account with a content distribution function, and the distribution account can be a self-media account. The text description information may be obtained directly or indirectly. Specifically, if the title, the label and other direct text presentation contents of the content to be distributed are directly obtained; such as indirectly acquiring the caption in video or audio, the text corresponding to background voice, etc. through picture recognition, voice recognition, etc.
It may be understood that, the self-Media (We Media) refers to a personalized and autonomous propagator, and uses a modern and electronic means to transfer normative and non-normative information to a large number of unspecified individuals or to a new Media generic term of the normative and non-normative information, where the self-Media account may be an account number (such as a microblog account number) which is registered in an independent content distribution platform and capable of autonomously distributing content, or may be an account number which is registered in a content distribution platform integrated in a social platform and capable of autonomously distributing content, and the content interaction information is information generated after a user interacts with the distribution account number in a historical period, for example, the user performs praise on the distributed content under the distribution account number, for example, the user forwards the distributed content under the distribution account number, and the content interaction information records the time, operation and the targeted distribution account number when the user interacts with the distribution account number.
Here, the definition of the timeliness of the content is explained, and the distributed content has a certain effect in a period of time, and the effect is measured by the interest degree of the user in the distributed content. In general, the timeliness refers to that video is pushed to a user within a time range and is not too long, and the timeliness plays an important role in retaining, clicking and clicking passing of the user on a side line of the opposite end, and pushing content to the user within the time-efficiency validity range of the content can play a positive role, otherwise, the user's dislike can be caused.
102. And extracting the characteristics of the collected content interaction information to obtain the interaction characteristics corresponding to the collected content interaction information.
The content interaction information comprises user praise behavior information, user forwarding behavior information, user comment behavior information and the like, and for a computer, the information needs to be converted into information which can be identified and processed by a structured computer, namely, the information is scientifically abstracted, and a mathematical model of the information is established for describing and replacing the information.
The feature extraction mode adopted for different types of behavior information may be different, but the concept is consistent, and comment behavior information is taken as an example, currently, comments of users for distribution content are generally embodied in a text form, in the application, a vector space model can be adopted to describe a text vector, but if feature items obtained by a word segmentation algorithm and a word frequency statistical method are directly used for representing each dimension in the text vector, the dimension of the vector is very large. The unprocessed text vector not only brings huge calculation cost for subsequent work, so that the efficiency of the whole processing process is very low, but also the accuracy of classification and clustering algorithms is damaged, and the obtained result is difficult to be satisfied. Therefore, further purification treatment is required to be carried out on the text vector, and the text feature most representative to the text feature class is found out on the basis of guaranteeing the original text meaning. To solve this problem, the most effective approach is to reduce the dimension by feature selection. There are 4 common:
(1) The original features are transformed into fewer new features by mapping or transformation methods.
(2) Some of the most representative features are selected from the original features.
(3) The most influential features are chosen based on a priori knowledge.
(4) The method is a relatively accurate method, has less interference of human factors, is particularly suitable for application of an automatic text classification mining system, and is particularly selected according to actual conditions, and the application is not limited.
103. And predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the acquired content interaction information.
The content to be distributed may be one or more of an image, a text, a voice and a video, so in order for a user to know the domain or the type of the content to be distributed, a corresponding content label is usually added to the content to be distributed, where the content label includes a content title of the content to be distributed and a classification label of the content to be distributed.
For the content to be distributed with more texts, extracting the content text of the content to be distributed, and predicting the content timeliness of the content to be distributed by combining the semantics of text description information and the interaction characteristics corresponding to the acquired content interaction information; for the content to be distributed with less text information such as images, voices and videos, taking the video as an example, the available semantic features of the video titles and video labels are not abundant, and the video timeliness is difficult to accurately judge through the video titles and the video labels in general, so that the accuracy of the subsequent directional delivery is affected. Therefore, in the present application, the content timeliness of the content to be distributed is predicted by using the semantics of the text description information and the interaction characteristics corresponding to the collected content interaction information, and optionally, in some embodiments, the step of predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the collected content interaction information may specifically include:
(11) Extracting a content tag of the content to be distributed from the text description information;
(12) Extracting features of the content labels to obtain label features corresponding to the content labels;
(13) Based on the semantics of the content tags, the tag characteristics and the interaction characteristics corresponding to the acquired content interaction information, the content timeliness of the content to be distributed is predicted.
The method for predicting the content timeliness of the content to be distributed based on the semantic meaning of the content tag, the tag feature and the collected interaction feature of the content interaction information comprises the following steps:
(21) Word segmentation is carried out on the label text corresponding to the content label based on the semantic meaning of the content label;
(22) Dividing the tag characteristics according to the word segmentation result to obtain element characteristics corresponding to each text element in the tag text;
(23) Based on element characteristics corresponding to each text element in the tag text and interaction characteristics corresponding to the collected content interaction information, the content timeliness of the content to be distributed is predicted.
Specifically, a preset timeliness prediction model may be obtained, and similar to the conventional search, one challenge of the content delivery system is how to obtain accuracy and expansibility of the delivery result at the same time. The recommended content is accurate content, the user interests are converged, the freshness is not felt, and the long-term user retention is not facilitated; the recommendation content is too generalized, the accurate interests of the users cannot be met, and the loss risk of the users is high, so that in the application, the time-based prediction model can be a "wide & deep model". Referring to fig. 1c, the timeliness prediction model includes a linear model and a depth model, wherein the linear model is used for processing interaction characteristics, the depth model is used for processing element characteristics corresponding to each text element, and finally, the results output by the two parts are spliced to output the timeliness of the content to be distributed, and the linear model can achieve the purpose of accurate classification by utilizing the cross characteristics to efficiently realize memory capacity. The linear model achieves a certain generalization capability by adding some broad class features. But limited to training data, linear models do not allow generalization that did not occur in training data. In the depth model, the generalization capability of the model can be realized through the learned low-latitude dense vector, so that generalization recommendation of unseen contents is realized. When the matrix of the model is sparse, the model is excessively generalized, a lot of content without correlation is recommended, and the accuracy cannot be ensured.
It should be noted that, the content interaction information may include comments of a plurality of users for the same content, the relation between the comments is not a linear relation, but in the present application, the content interaction information is processed through a linear model, so, in order to solve the nonlinear problem, feature combinations may be performed on the interaction features corresponding to the collected content interaction information, where the feature combinations are synthesized features that code the nonlinear rule in the feature space by multiplying two or more input features, then, based on the combined features, predict the degree of user attention of the content to be distributed in different preset periods, to obtain a first degree of attention corresponding to each preset period, and simultaneously, predict the degree of user attention of the content to be distributed in different preset periods according to the element features corresponding to each text element in the tag text, to obtain a second degree of attention corresponding to each preset period, and finally, combine the first degree of attention and the second degree of attention corresponding to the same preset period, to obtain the content timeliness of the content to be distributed in each preset period, that is, optionally, in some embodiments, the steps "based on the timeliness of the collected text element corresponding to the text element, the content to be distributed may include the timeliness of the text corresponding to the text element, based on the predicted content information, and the corresponding to the text feature, and the timeliness of the text element, which is predicted by the corresponding to the text element, and the content interaction information, and the feature, which is the method, and the method to be used by the method and the method to be provided by:
(31) Combining the characteristics of the interaction characteristics corresponding to the collected content interaction information to obtain combined characteristics;
(32) Predicting the user attention degree of the content to be distributed in different preset time periods according to the combined characteristics to obtain a first attention degree corresponding to each preset time period;
(33) Predicting the user attention degree of the content to be distributed in different preset time periods according to the element characteristics corresponding to each text element in the tag text, and obtaining a second attention degree corresponding to each preset time period;
(34) And fusing the first attention degree and the second attention degree corresponding to each preset period to obtain the content timeliness of the content to be distributed in each preset period.
As described above, the content timeliness is measured by means of the user's interest degree (i.e., attention degree) in the distributed content, so that the user's interest degree of the content to be distributed in different preset time periods is predicted according to the combined feature, the user's interest degree of the content to be distributed in different preset time periods is predicted according to the element feature corresponding to each text element in the tag text, and finally, the content timeliness of the content to be distributed is obtained by fusing the two predicted attention degrees, where the size of the preset time period may be determined according to the actual situation, for example, in terms of day, and the size of one preset time period is 3 days, so that the content timeliness of the content to be distributed in the future 3 days, the future 6 days, the future 9 days, and so on can be predicted.
In addition, for text processing in natural language processing, machine Learning (ML) technology is generally used to implement text processing. The machine learning is a multi-field interdisciplinary, and relates to a plurality of disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and the like. It is specially studied how a computer simulates or implements learning behavior of a human to acquire new knowledge or skills, and reorganizes existing knowledge structures to continuously improve own performance. Machine learning is the core of artificial intelligence, a fundamental approach to letting computers have intelligence, which is applied throughout various areas of artificial intelligence. Machine learning and deep learning generally include techniques such as artificial neural networks, confidence networks, reinforcement learning, transfer learning, induction learning, teaching learning, etc., wherein natural language processing (Nature Language processing, NLP) is an important direction in the fields of computer science and artificial intelligence. It is studying various theories and methods that enable effective communication between a person and a computer in natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Thus, the research in this field will involve natural language, i.e. language that people use daily, so it has a close relationship with the research in linguistics. Natural language processing techniques typically include text processing, semantic understanding, machine translation, robotic questions and answers, knowledge graph techniques, and the like.
It may be appreciated that, the content interaction information may include content interaction information corresponding to positive interaction behavior and content interaction information corresponding to negative interaction behavior, where the positive interaction behavior refers to positive interaction behavior, such as collection, forwarding, sharing praise, and positive comments on the content, and the negative interaction behavior refers to negative interaction behavior, such as reporting and negative comments on the content, so as to improve accuracy of subsequent delivery, in some embodiments of the present application, the negative interaction information may be removed to avoid an influence of the negative interaction information on the subsequent delivery, that is, the content delivery method provided by the present application may specifically further include:
(41) Identifying the intention of the collected content interaction information;
(42) According to the intention recognition result, retaining interaction information corresponding to the intention as the forward intention;
the step of predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the collected content interaction information may specifically include: and predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the interaction information of which the intention is the forward intention.
104. Content distribution is performed on the content to be distributed based on the predicted content timeliness.
For example, specifically, the content to be distributed may be ordered according to the predicted content timeliness, and then content distribution is performed on the content to be distributed based on the ordered sequence, that is, optionally, in some embodiments, the step of "content distribution is performed on the content to be distributed based on the predicted content timeliness" may specifically include:
(51) Ordering the content to be distributed in the content to be distributed based on the predicted content timeliness to obtain a content list to be distributed;
(52) And distributing the content to be distributed in the content list to be distributed according to the sequence of the content to be distributed in the content list to be distributed.
It should be noted that in practical application, there may be a situation that the content timeliness of the content a to be distributed and the content B to be distributed is the same, for example, the content timeliness of the content a to be distributed and the content timeliness of the content B to be distributed are both 3 days, the domain to which the content a to be distributed belongs is news information class, the domain to which the content B to be distributed belongs is science popularization course class, and at this time, the content delivery of the content a to be distributed may be performed preferentially according to the different domains; for example, the timeliness of the content a to be distributed and the timeliness of the content B to be distributed are 3 days, and the fields to which the content a to be distributed and the content B to be distributed belong are science popularization courses, and at this time, the content to be distributed with high click rate can be preferentially put in, and of course, the content to be distributed can also be put in according to other strategies, which is only illustrative and not limiting.
After the associated information of the content to be distributed is collected, the collected content interaction information is subjected to feature extraction to obtain interaction features corresponding to the collected content interaction information, then, according to the semantics of the text description information and the interaction features corresponding to the collected content interaction information, the content timeliness of the content to be distributed is predicted, and finally, the content to be distributed is distributed based on the predicted content timeliness. According to the content distribution method, the content timeliness of the content to be distributed can be predicted according to the content interaction information and the text description information, the content timeliness of the content to be distributed is predicted through different dimensions, the problem that the predicted content timeliness is inaccurate due to the fact that the content timeliness of the content to be distributed is predicted through the information of the single dimension is avoided, and therefore the content to be distributed can be accurately pushed to a target user within the validity period of the content timeliness, and the accuracy of content distribution is improved.
The method according to the embodiment will be described in further detail by way of example.
In this embodiment, a description will be given of an example in which the content distribution apparatus is specifically integrated in a server.
Referring to fig. 2a, a specific flow of a content distribution method may be as follows:
201. the server collects the associated information of the content to be distributed.
The associated information of the content to be distributed includes text description information corresponding to a plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed, and it can be understood that each content to be distributed has corresponding text description information, where the text description information may include a content text corresponding to the content to be distributed, a content title of the content to be distributed and a classification tag of the content to be distributed, and the server may collect the associated information of the content to be distributed uploaded by the distribution account through network connection.
202. And the server performs feature extraction on the collected content interaction information to obtain interaction features corresponding to the collected content interaction information.
203. And the server predicts the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the collected content interaction information.
Specifically, the server may segment the tag features according to the semantics of the tag text, then predict the content timeliness of the content to be distributed based on the segmented features and the interaction features corresponding to the collected content interaction information, further, the server may obtain a preset timeliness prediction model, then perform feature combination on the interaction features corresponding to the collected content interaction information to obtain combined features, then predict the degree of attention of the user of the content to be distributed in different preset time periods according to the combined features to obtain a first degree of attention corresponding to each preset time period, and predict the degree of attention of the user of the content to be distributed in different preset time periods according to the element features corresponding to each text element in the tag text to obtain a second degree of attention corresponding to each preset time period, and finally, the server fuses the first degree of attention and the second degree of attention corresponding to each preset time period to obtain the content timeliness of the content to be distributed in each preset time period.
204. The server distributes the content to be distributed based on the predicted content timeliness.
For example, specifically, the server may sort the content to be distributed according to the predicted content timeliness, and then distribute the content to be distributed based on the sorted order.
For further understanding of the content distribution scheme of the present application, please refer to fig. 2b, which illustrates the architecture of the time-efficient prediction model, the wick & deep model proposed in the present application balances the memory and generalization capabilities of the linear model and the depth model, and the linear model is denoted as y=wx+b. X is the feature portion, which includes the base feature and the cross feature. Cross features are important in the linear part, and interaction among features can be captured, so that the effect of adding nonlinearity is achieved. The cross-over feature can be expressed as:
the depth part is a feed forward network model. Features are first converted to low-dimensional dense vectors, typically between O (10) and O (100) in dimension. And (3) randomly initializing vectors, and training a model through minimizing a function at any time. The activation function employs Relu. The feedforward section is represented as follows:
a (l+1) ＝f(W (l) a (l) +b (l) )
the outputs of the Wide and Deep parts are combined together by weighting and the final output is made by logistic loss function, which can be expressed by:
Taking content to be distributed as video as an example, in a training stage, a server collects sample video information, wherein the sample video information comprises text description information of the sample video, the sample video and content interaction information under a corresponding distribution account, real content timeliness is marked on the sample video, then the server performs feature extraction on the collected content interaction information to obtain interaction features corresponding to the collected content interaction information, then the server extracts video titles and classification tags of the sample video from the text description information, performs feature extraction on the video titles and the classification tags respectively to obtain title features of the video titles and tag features corresponding to the classification tags, then the server inputs the title features and the tag features into a depth model of a timeliness prediction model, inputs the interaction features corresponding to the content interaction information into a linear model of the timeliness prediction model, performs feature combination on the collected interaction features by using the linear model to obtain combined features, and then the server processes the combined features and the interaction features by using the linear model to obtain the degree of attention of users of the sample video in different preset periods; in addition, the server processes the title feature and the label feature of the depth model to obtain the attention degree of the user of the sample video in different preset time periods, wherein the title feature and the label feature can be respectively subjected to embedded vectorization, and parameters of the depth model can be updated according to a final loss function; the use stage is the same as that of the previous embodiment, and specific reference is made to the previous embodiment, and details are not repeated here.
After the server collects the associated information of the content to be distributed, the server performs feature extraction on the collected content interaction information to obtain interaction features corresponding to the collected content interaction information, then predicts the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction features corresponding to the collected content interaction information, and finally performs content distribution on the content to be distributed based on the predicted content timeliness. According to the server provided by the application, the content timeliness of the content to be distributed can be predicted according to the content interaction information and the text description information, the content timeliness of the content to be distributed is predicted through different dimensions, the problem that the predicted content timeliness is inaccurate due to the fact that the content timeliness of the content to be distributed is predicted through the information of the single dimension is avoided, and therefore the content to be distributed can be accurately pushed to a target user within the validity period of the content timeliness, and the accuracy of content distribution is improved.
In order to facilitate better implementation of the content distribution method of the present application, the present application further provides a content distribution device (abbreviated as a distribution device) based on the above-mentioned content distribution device. Where the meaning of a noun is the same as in the content distribution method described above, specific implementation details may be referred to the description in the method embodiment.
Referring to fig. 3a, fig. 3a is a schematic structural diagram of a content distribution apparatus provided in the present application, where the distribution apparatus may include an acquisition module 301, an extraction module 302, a prediction module 303, and a distribution module 304, and may specifically be as follows:
the collection module 301 is configured to collect association information of content to be distributed.
The associated information of the content to be distributed includes text description information corresponding to a plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed, and it can be understood that each content to be distributed has corresponding text description information, where the text description information may include a content text corresponding to the content to be distributed, a content title of the content to be distributed, and a classification tag of the content to be distributed, and the acquisition module 301 may acquire the associated information of the content to be distributed uploaded by the distribution account through network connection.
The extracting module 302 is configured to perform feature extraction on the collected content interaction information, so as to obtain interaction features corresponding to the collected content interaction information.
And the prediction module 303 is configured to predict content timeliness of the content to be distributed according to semantics of the text description information and interaction characteristics corresponding to the collected content interaction information.
For example, specifically, the prediction module 303 predicts the content timeliness of the content to be distributed by using the semantics of the text description information and the interaction characteristics corresponding to the collected content interaction information, that is, optionally, in some embodiments, the prediction module 303 may specifically include:
the first extraction submodule block is used for extracting content labels of the content to be distributed from the text description information;
the second extraction submodule is used for extracting the characteristics of the content labels to obtain the label characteristics corresponding to the content labels;
and the prediction sub-module is used for predicting the content timeliness of the content to be distributed based on the semantics of the content tag, the tag characteristics and the interaction characteristics corresponding to the acquired content interaction information.
Optionally, in some embodiments, the prediction submodule may specifically include:
the word segmentation unit is used for segmenting the label text corresponding to the content label based on the semantic meaning of the content label;
the segmentation unit is used for segmenting the tag characteristics according to the word segmentation result to obtain element characteristics corresponding to each text element in the tag text;
the predicting unit is used for predicting the content timeliness of the content to be distributed based on element characteristics corresponding to each text element in the tag text and interaction characteristics corresponding to the acquired content interaction information.
Optionally, in some embodiments, the prediction unit may specifically include:
the acquisition subunit is used for acquiring a preset timeliness prediction model;
the prediction subunit is used for predicting the content timeliness of the content to be distributed through the timeliness prediction model based on element characteristics corresponding to each text element in the tag text and interaction characteristics corresponding to the acquired content interaction information.
Alternatively, in some embodiments, the prediction subunit may specifically be configured to: and carrying out feature combination on the interaction features corresponding to the acquired content interaction information to obtain combined features, predicting the user attention degree of the content to be distributed in different preset time periods according to the combined features to obtain a first attention degree corresponding to each preset time period, predicting the user attention degree of the content to be distributed in different preset time periods according to the element features corresponding to each text element in the tag text to obtain a second attention degree corresponding to each preset time period, and fusing the first attention degree and the second attention degree corresponding to each preset time period to obtain the content timeliness of the content to be distributed in each preset time period.
Optionally, in some embodiments, referring to fig. 3b, the dispensing apparatus specifically may further include an identification module 305, and the identification module 305 may specifically be configured to: identifying the intention of the collected content interaction information, and reserving the interaction information corresponding to the intention as the forward intention according to the intention identification result;
The prediction module 303 may specifically be configured to: and predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the forward interaction information.
A distribution module 304, configured to distribute the content to be distributed based on the predicted content timeliness.
For example, in particular, the distribution module 304 may sort the content to be distributed according to the predicted content timeliness, and then the distribution module 304 distributes the content to be distributed based on the sorted order, and optionally, in some embodiments, the distribution module 304 may be specifically configured to: ordering the content to be distributed in the content to be distributed based on the predicted content timeliness to obtain a content list to be distributed, and distributing the content to be distributed in the content list to be distributed according to the sequence of the content to be distributed in the content list to be distributed.
After the acquisition module 301 acquires the associated information of the content to be distributed, the extraction module 302 performs feature extraction on the acquired content interaction information to obtain interaction features corresponding to the acquired content interaction information, then the prediction module 303 predicts the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction features corresponding to the acquired content interaction information, and finally the distribution module 304 distributes the content based on the predicted content timeliness of the associated information of the content to be distributed. According to the content distribution device, the content timeliness of the content to be distributed can be predicted according to the content interaction information and the text description information, the content timeliness of the content to be distributed is predicted through different dimensions, the problem that the predicted content timeliness is inaccurate due to the fact that the content timeliness of the content to be distributed is predicted through the information of the single dimension is avoided, and therefore the content to be distributed can be accurately pushed to a target user within the validity period of the content timeliness, and the accuracy of content distribution is improved.
In addition, the present application further provides an electronic device, as shown in fig. 4, which shows a schematic structural diagram of the electronic device according to the present application, specifically:
the electronic device may include one or more processing cores 'processors 401, one or more computer-readable storage media's memory 402, power supply 403, and input unit 404, among other components. Those skilled in the art will appreciate that the electronic device structure shown in fig. 4 is not limiting of the electronic device and may include more or fewer components than shown, or may combine certain components, or may be arranged in different components. Wherein:
the processor 401 is a control center of the electronic device, connects various parts of the entire electronic device using various interfaces and lines, and performs various functions of the electronic device and processes data by running or executing software programs and/or modules stored in the memory 402, and calling data stored in the memory 402, thereby performing overall detection of the electronic device. Optionally, processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a user interface, an application program, etc., and the modem processor mainly processes wireless communication. It will be appreciated that the modem processor described above may not be integrated into the processor 401.
The memory 402 may be used to store software programs and modules, and the processor 401 executes various functional applications and data processing by executing the software programs and modules stored in the memory 402. The memory 402 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data created according to the use of the electronic device, etc. In addition, memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.
The electronic device further comprises a power supply 403 for supplying power to the various components, preferably the power supply 403 may be logically connected to the processor 401 by a power management system, so that functions of managing charging, discharging, and power consumption are performed by the power management system. The power supply 403 may also include one or more of any of a direct current or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.
The electronic device may further comprise an input unit 404, which input unit 404 may be used for receiving input digital or character information and generating keyboard, mouse, joystick, optical or trackball signal inputs in connection with user settings and function control.
Although not shown, the electronic device may further include a display unit or the like, which is not described herein. In particular, in this embodiment, the processor 401 in the electronic device loads executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 executes the application programs stored in the memory 402, so as to implement various functions as follows:
collecting associated information of the content to be distributed, extracting features of the collected content interaction information to obtain interaction features corresponding to the collected content interaction information, predicting content timeliness of the content to be distributed according to semantics of text description information and the interaction features corresponding to the collected content interaction information, and distributing the content to be distributed based on the predicted content timeliness.
The specific implementation of each operation above may be referred to the previous embodiments, and will not be described herein.
After the associated information of the content to be distributed is collected, the collected content interaction information is subjected to feature extraction to obtain interaction features corresponding to the collected content interaction information, then, according to the semantics of the text description information and the interaction features corresponding to the collected content interaction information, the content timeliness of the content to be distributed is predicted, and finally, the content to be distributed is distributed based on the predicted content timeliness. According to the content distribution method, the content timeliness of the content to be distributed can be predicted according to the content interaction information and the text description information, the content timeliness of the content to be distributed is predicted through different dimensions, the problem that the predicted content timeliness is inaccurate due to the fact that the content timeliness of the content to be distributed is predicted through the information of the single dimension is avoided, and therefore the content to be distributed can be accurately pushed to a target user within the validity period of the content timeliness, and the accuracy of content distribution is improved.
Those of ordinary skill in the art will appreciate that all or a portion of the steps of the various methods of the above embodiments may be performed by instructions, or by instructions controlling associated hardware, which may be stored in a computer-readable storage medium and loaded and executed by a processor.
To this end, the present application provides a storage medium having stored therein a plurality of instructions capable of being loaded by a processor to perform the steps of any of the content distribution methods provided herein. For example, the instructions may perform the steps of:
collecting associated information of the content to be distributed, extracting features of the collected content interaction information to obtain interaction features corresponding to the collected content interaction information, predicting content timeliness of the content to be distributed according to semantics of text description information and the interaction features corresponding to the collected content interaction information, and distributing the content to be distributed based on the predicted content timeliness.
The specific implementation of each operation above may be referred to the previous embodiments, and will not be described herein.
Wherein the storage medium may include: read Only Memory (ROM), random access Memory (RAM, random Access Memory), magnetic or optical disk, and the like.
The steps in any content distribution method provided in the present application may be executed due to the instructions stored in the storage medium, so that the beneficial effects that any content distribution method provided in the present application may achieve are achieved, which are detailed in the previous embodiments and are not described herein.
According to one aspect of the present application, there is provided a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The computer instructions are read from the computer-readable storage medium by a processor of a computer device, and executed by the processor, cause the computer device to perform the methods provided in the various alternative implementations described above.
The foregoing has described in detail a content distribution method, apparatus, electronic device and storage medium provided in the present application, and specific examples have been applied to illustrate the principles and embodiments of the present invention, and the above examples are only used to help understand the method and core idea of the present invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present invention, the present description should not be construed as limiting the present invention.

Claims (6)

1. A content distribution method, comprising:
acquiring associated information of content to be distributed, wherein the associated information of the content to be distributed comprises text description information corresponding to a plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed;
Extracting features of the collected content interaction information to obtain interaction features corresponding to the collected content interaction information;
predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the collected content interaction information, wherein the method comprises the following steps: extracting a content tag of the content to be distributed from the text description information; extracting features of the content tags to obtain tag features corresponding to the content tags; word segmentation is carried out on the label text corresponding to the content label based on the semantic meaning of the content label; dividing the tag characteristics according to word segmentation results to obtain element characteristics corresponding to each text element in the tag text; combining the characteristics of the interaction characteristics corresponding to the collected content interaction information to obtain combined characteristics; predicting the attention degree of the user of the content to be distributed in different preset time periods according to the combination characteristics to obtain a first attention degree corresponding to each preset time period; predicting the user attention degree of the content to be distributed in different preset time periods according to element characteristics corresponding to each text element in the tag text, and obtaining a second attention degree corresponding to each preset time period; fusing a first attention degree and a second attention degree corresponding to each preset period to obtain the content timeliness of the content to be distributed in each preset period;
And distributing the content to be distributed based on the predicted content timeliness.
2. The method as recited in claim 1, further comprising:
identifying the intention of the collected content interaction information;
according to the intention recognition result, retaining interaction information corresponding to the intention as the forward intention;
the predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the collected content interaction information comprises the following steps: and predicting the content timeliness of the content to be distributed according to the semantics of the text description information and the interaction characteristics corresponding to the interaction information of which the intention is the forward intention.
3. The method according to claim 1 or 2, wherein the content distribution of the content to be distributed based on the predicted content timeliness comprises:
ordering the content to be distributed in the content to be distributed based on the predicted content timeliness to obtain a content list to be distributed;
and distributing the content to be distributed in the content list to be distributed according to the sequence of the content to be distributed in the content list to be distributed.
4. A content distribution apparatus, comprising:
The system comprises an acquisition module, a distribution module and a distribution module, wherein the acquisition module is used for acquiring the associated information of the content to be distributed, and the associated information of the content to be distributed comprises text description information corresponding to a plurality of content to be distributed and content interaction information under a distribution account corresponding to the content to be distributed;
the extraction module is used for extracting the characteristics of the collected content interaction information to obtain the interaction characteristics corresponding to the collected content interaction information;
the prediction module is configured to predict content timeliness of the content to be distributed according to semantics of the text description information and interaction characteristics corresponding to the collected content interaction information, and includes: extracting a content tag of the content to be distributed from the text description information; extracting features of the content tags to obtain tag features corresponding to the content tags; word segmentation is carried out on the label text corresponding to the content label based on the semantic meaning of the content label; dividing the tag characteristics according to word segmentation results to obtain element characteristics corresponding to each text element in the tag text; combining the characteristics of the interaction characteristics corresponding to the collected content interaction information to obtain combined characteristics; predicting the attention degree of the user of the content to be distributed in different preset time periods according to the combination characteristics to obtain a first attention degree corresponding to each preset time period; predicting the user attention degree of the content to be distributed in different preset time periods according to element characteristics corresponding to each text element in the tag text, and obtaining a second attention degree corresponding to each preset time period; fusing a first attention degree and a second attention degree corresponding to each preset period to obtain the content timeliness of the content to be distributed in each preset period;
And the distribution module is used for distributing the content to be distributed based on the predicted content timeliness.
5. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the content distribution method of any of claims 1-3 when the program is executed by the processor.
6. A computer readable storage medium, having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of the content distribution method according to any of claims 1-3.
CN202011009503.7A 2020-09-23 2020-09-23 Content distribution method, device, electronic equipment and storage medium Active CN112165639B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011009503.7A CN112165639B (en) 2020-09-23 2020-09-23 Content distribution method, device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011009503.7A CN112165639B (en) 2020-09-23 2020-09-23 Content distribution method, device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN112165639A CN112165639A (en) 2021-01-01
CN112165639B true CN112165639B (en) 2024-02-02

Family

ID=73862840

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011009503.7A Active CN112165639B (en) 2020-09-23 2020-09-23 Content distribution method, device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112165639B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113301376B (en) * 2021-05-24 2023-04-07 成都威爱新经济技术研究院有限公司 Live broadcast interaction method and system based on virtual reality technology
CN114330295A (en) * 2021-08-04 2022-04-12 腾讯科技（深圳）有限公司 Time efficiency identification, model training and pushing method, device and medium of information
CN115730111B (en) * 2021-09-01 2024-02-06 腾讯科技（深圳）有限公司 Content distribution method, apparatus, device and computer readable storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109086345A (en) * 2018-07-12 2018-12-25 北京奇艺世纪科技有限公司 A kind of content identification method, content distribution method, device and electronic equipment
CN111008278A (en) * 2019-11-22 2020-04-14 厦门美柚股份有限公司 Content recommendation method and device
CN111026969A (en) * 2019-12-18 2020-04-17 腾讯科技（深圳）有限公司 Content recommendation method and device, storage medium and server
CN111339404A (en) * 2020-02-14 2020-06-26 腾讯科技（深圳）有限公司 Content popularity prediction method and device based on artificial intelligence and computer equipment
CN111581510A (en) * 2020-05-07 2020-08-25 腾讯科技（深圳）有限公司 Shared content processing method and device, computer equipment and storage medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106202394B (en) * 2016-07-07 2021-03-19 腾讯科技（深圳）有限公司 Text information recommendation method and system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109086345A (en) * 2018-07-12 2018-12-25 北京奇艺世纪科技有限公司 A kind of content identification method, content distribution method, device and electronic equipment
CN111008278A (en) * 2019-11-22 2020-04-14 厦门美柚股份有限公司 Content recommendation method and device
CN111026969A (en) * 2019-12-18 2020-04-17 腾讯科技（深圳）有限公司 Content recommendation method and device, storage medium and server
CN111339404A (en) * 2020-02-14 2020-06-26 腾讯科技（深圳）有限公司 Content popularity prediction method and device based on artificial intelligence and computer equipment
CN111581510A (en) * 2020-05-07 2020-08-25 腾讯科技（深圳）有限公司 Shared content processing method and device, computer equipment and storage medium

Also Published As

Publication number Publication date
CN112165639A (en) 2021-01-01

Similar Documents

Publication Publication Date Title
CN111444428B (en) Information recommendation method and device based on artificial intelligence, electronic equipment and storage medium
CN111177575B (en) Content recommendation method and device, electronic equipment and storage medium
CN112165639B (en) Content distribution method, device, electronic equipment and storage medium
CN111382361B (en) Information pushing method, device, storage medium and computer equipment
CN111966914B (en) Content recommendation method and device based on artificial intelligence and computer equipment
CN111885399A (en) Content distribution method, content distribution device, electronic equipment and storage medium
CN113254711B (en) Interactive image display method and device, computer equipment and storage medium
CN108959323B (en) Video classification method and device
CN111625715B (en) Information extraction method and device, electronic equipment and storage medium
CN112131430A (en) Video clustering method and device, storage medium and electronic equipment
CN111324773A (en) Background music construction method and device, electronic equipment and storage medium
CN111563158A (en) Text sorting method, sorting device, server and computer-readable storage medium
CN114201516A (en) User portrait construction method, information recommendation method and related device
CN115131052A (en) Data processing method, computer equipment and storage medium
CN112101015B (en) Method and device for identifying multi-label object
CN116484085A (en) Information delivery method, device, equipment, storage medium and program product
CN114840771A (en) False news detection method based on news environment information modeling
Li et al. Enterprise precision marketing effectiveness model based on data mining technology
CN115130453A (en) Interactive information generation method and device
CN112446738A (en) Advertisement data processing method, device, medium and electronic equipment
CN114926192A (en) Information processing method and device and computer readable storage medium
CN111091198A (en) Data processing method and device
CN112749335B (en) Lifecycle state prediction method, lifecycle state prediction apparatus, computer device, and storage medium
CN115630170B (en) Document recommendation method, system, terminal and storage medium
CN117711001B (en) Image processing method, device, equipment and medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
REG Reference to a national code

Ref country code: HK

Ref legal event code: DE

Ref document number: 40037427

Country of ref document: HK

SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant