CN115203539B - Media content recommendation method, device, equipment and storage medium - Google Patents

Media content recommendation method, device, equipment and storage medium Download PDF

Info

Publication number
CN115203539B
CN115203539B CN202210766812.1A CN202210766812A CN115203539B CN 115203539 B CN115203539 B CN 115203539B CN 202210766812 A CN202210766812 A CN 202210766812A CN 115203539 B CN115203539 B CN 115203539B
Authority
CN
China
Prior art keywords
media content
target
novel
user
information
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
CN202210766812.1A
Other languages
Chinese (zh)
Other versions
CN115203539A (en
Inventor
王一
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.)
Shenzhen Renma Interactive Technology Co Ltd
Original Assignee
Shenzhen Renma Interactive 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 Shenzhen Renma Interactive Technology Co Ltd filed Critical Shenzhen Renma Interactive Technology Co Ltd
Priority to CN202210766812.1A priority Critical patent/CN115203539B/en
Publication of CN115203539A publication Critical patent/CN115203539A/en
Application granted granted Critical
Publication of CN115203539B publication Critical patent/CN115203539B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G06F16/9035—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses a media content recommendation method, a device, equipment and a storage medium, wherein in the scheme of the method, webpage display content is acquired; at least one interactive node is embedded in the webpage display content; acquiring a to-be-selected media content set matched with the characteristic information of the webpage display content; the to-be-selected media content set comprises at least one media content; acquiring user interaction information at a target interaction node in the at least one interaction node; and at least one interactive node, screening target media contents from the to-be-selected media content set based on the user interaction information, and outputting the target media contents to a user. The scheme has better interaction effect, and can play a better advertisement putting effect on an advertisement recommendation scene. Accordingly, the media content recommendation device, the equipment and the storage medium provided by the application also have the technical effects.

Description

Media content recommendation method, device, equipment and storage medium
Technical Field
The present application relates to the field of information recommendation technologies, and in particular, to a method, an apparatus, a device, and a storage medium for recommending media content.
Background
With the development of network technology, many interactive Applications (APPs) have appeared. These interactive applications may present web page presentation content such as web page interactive novels. The webpage display content can receive input of a user and carry out subsequent response according to the input of the user, for example, different scenes are displayed in a targeted mode. This makes the experience of different users in web page presentation content quite different. In addition, there is currently a need to recommend media content in web page presentation content to attract user clicks. However, currently, media content is pushed mainly before logging into an interactive application or entering a certain web scene. The pushing of the media content is too extensive and the content is single, and the pertinence of the pushed media content is not strong, so that the recommendation of the media content is not accurate enough.
Disclosure of Invention
In view of the above, an object of the present application is to provide a method, an apparatus, a device and a storage medium for recommending media content, which can implement accurate recommendation of media content. The specific scheme is as follows:
to achieve the above object, in one aspect, the present application provides a media content recommendation method, including:
acquiring webpage display content; at least one interactive node is embedded in the webpage display content;
acquiring a to-be-selected media content set matched with the characteristic information of the webpage display content; the to-be-selected media content set comprises at least one media content;
acquiring user interaction information at a target interaction node in the at least one interaction node;
and at least one interactive node, screening out target media content from the to-be-selected media content set based on the user interaction information, and outputting the target media content to a user.
Optionally, the web page display content is a web page novel, the web page novel includes at least one novel segment, and the interactive nodes embedded in the web page novel respectively correspond to the novel segments;
the acquiring of the to-be-selected media content set matched with the feature information of the webpage display content comprises the following steps:
predicting information input by a user at the target interaction node based on an input information prediction model to obtain user input prediction information corresponding to the target interaction node;
obtaining feature information of the webpage novel based on the user input prediction information;
and acquiring a to-be-selected media content set matched with the characteristic information of the webpage novel.
Optionally, the acquiring a set of media content to be selected for matching with the feature information of the web novel includes:
acquiring a media content library;
acquiring media content matched with the user input prediction information from the media content library to obtain a first media content set;
and integrating to obtain the media content set to be selected according to the first media content set.
Optionally, the obtaining the feature information of the web novel based on the prediction information input by the user includes:
acquiring a novel subject matter of the webpage novel;
identifying a scene type related to the webpage novel based on a scene identification model;
determining a target novel fragment corresponding to the target interactive node, and performing semantic recognition on the target novel fragment based on a semantic recognition model to obtain novel fragment semantics corresponding to the target novel fragment;
and obtaining the characteristic information of the webpage novel based on the novel subject matter of the webpage novel and the related scene type, the novel segment semantics corresponding to the target novel segment and the prediction information input by the user.
Optionally, the acquiring a set of media content to be selected for matching with the feature information of the web novel further includes:
acquiring a media content library;
acquiring media content matched with the user input prediction information from the media content library to obtain a first media content set;
acquiring media content matched with the novel subject matter of the webpage novel from the media content library to obtain a second media content set;
acquiring media content matched with the scene type related to the webpage novel from the media content library to obtain a third media content set;
acquiring media content matched with the novel fragment semantics corresponding to the target novel fragment from the media content library to obtain a fourth media content set;
and integrating to obtain the media content set to be selected according to at least one of the second media content set, the third media content set and the fourth media content set and the first media content set.
Optionally, the integrating, according to the first media content set and at least one of the second media content set, the third media content set, and the fourth media content set, to obtain the media content set to be selected includes:
and performing duplicate removal on the media contents in the first media content set, the second media content set, the third media content set and the fourth media content set, and arranging and integrating the duplicate-removed media content sets in parallel to obtain the media content set to be selected.
Optionally, the integrating, according to the first media content set and at least one of the second media content set, the third media content set, and the fourth media content set, to obtain the media content set to be selected further includes:
classifying and arranging the second media content set, the third media content set and the fourth media content set according to a preset data structure to obtain a first-level media content set; arranging the first media content set into a sub-level of at least one media content set in the first level media content set to obtain a second level media content set; obtaining the media content set to be selected after arrangement;
correspondingly, the screening of the target media content from the set of media contents to be selected based on the user interaction information includes:
performing matching analysis on the user interaction information and each media content in the second-level media content set according to the preset data structure to obtain matched media content; determining a previous level of media content corresponding to the matched media content in the first set of level of media content; and determining the matched media content and/or the upper-level media content as the target media content.
Optionally, the method further includes:
obtaining a set of candidate media content based on the second set of media content, the third set of media content, and the fourth set of media content;
acquiring user interaction information at each interaction node of the webpage display content, and determining matched media content based on the user interaction information of each interaction node to obtain a reference media content set; comparing the media content in the candidate media content set with the media content in the reference media content set, and when determining that the candidate media content set contains media content which does not exist in the reference media content set, taking the corresponding media content as difference media content;
and embedding a new interactive node in the corresponding novel segment based on the difference media content so as to update the interactive node in the webpage novel.
Optionally, the method further includes:
obtaining a set of candidate media content based on the second set of media content, the third set of media content, and the fourth set of media content;
comparing the media contents in the candidate media content set with the media contents in the first media content set, and when determining that the candidate media content set comprises media contents which do not exist in the first media content set, taking the corresponding media contents as difference media contents;
and embedding a new interactive node in the corresponding novel segment based on the difference media content so as to update the interactive node in the webpage novel.
Optionally, the obtaining, by the target interaction node in the at least one interaction node, user interaction information includes:
receiving current user input information input by a target user corresponding to the target user identification when the target user interacts at the target interaction node, and obtaining the user interaction information; the target user identification is a user identification for logging in the interactive application program; the interactive application program is used for displaying the webpage display content;
correspondingly, the screening of the target media content from the to-be-selected media content set based on the user interaction information includes:
and screening target media content from the to-be-selected media content set based on the current user input information.
Optionally, the screening, based on the current user input information, target media content from the to-be-selected media content set includes:
acquiring historical user input information corresponding to the target user identification; the historical user input information comprises user input information of the target user in an interaction node before the target interaction node;
acquiring a historical operation record corresponding to the target user identification; the historical operation record is used for recording the historical operation executed by the target user in the interactive application program;
and screening target media content from the set of media content to be selected based on at least one of the historical user input information and the historical operation record and the current user input information.
Optionally, the screening, at the at least one interaction node, target media content from the set of media content to be selected based on the user interaction information, and outputting the target media content to the user includes:
screening target media content from the to-be-selected media content set based on the user interaction information;
determining a recommended interaction node in the at least one interaction node based on scene types and/or novel fragment semantics related to novel fragments in the webpage novel;
and outputting the target media content to the user based on the recommended interaction node.
Optionally, the outputting the target media content to the user includes:
determining a recommended interactive node among the at least one interactive node;
generating a media content display window and a media content cancel window which are adjacent in the webpage of the webpage display content based on the recommended interaction node within a preset time period after the user interaction information is obtained; the media content display window is used for displaying the target media content;
if the triggering operation of the media content canceling window is received before the deadline time corresponding to the preset time period is reached, closing the target media content and the media content canceling window;
otherwise, closing the target media content and the media content canceling window when the deadline is reached.
In another aspect, the present application also provides a media content recommendation apparatus, including:
the display content acquisition module is used for acquiring webpage display content; at least one interactive node is embedded in the webpage display content;
the content set acquisition module is used for acquiring a to-be-selected media content set matched with the characteristic information of the webpage display content; the to-be-selected media content set comprises at least one media content;
the interactive information acquisition module is used for acquiring user interactive information from a target interactive node in the at least one interactive node;
and the media content output module is used for screening out target media content from the to-be-selected media content set on the basis of the user interaction information at least one interaction node and outputting the target media content to the user.
In yet another aspect, the present application further provides an electronic device comprising a processor and a memory; wherein the memory is used for storing a computer program which is loaded and executed by the processor to implement the aforementioned media content recommendation method.
In still another aspect, the present application further provides a storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are loaded and executed by a processor, the method for recommending media content as described above is implemented.
According to the media content recommendation method, at least one interactive node is embedded in webpage display content, a to-be-selected media content set is obtained according to the characteristic information of the webpage display content, target media content is screened from the to-be-selected media content set based on user interaction information obtained from a target interactive node, and the target media content is output to a user. Accordingly, the media content recommendation device, the equipment and the storage medium provided by the application also have the technical effects.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flowchart of a method for recommending media content according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a window display interface provided in an embodiment of the present application;
FIG. 3 is a schematic view of a window display interface according to another embodiment of the present application;
FIG. 4 is a schematic view of a window display interface according to yet another embodiment of the present application;
FIG. 5 is a schematic view of a window display interface according to yet another embodiment of the present application;
FIG. 6 is a schematic diagram of a window display interface according to yet another embodiment of the present application;
FIG. 7 is a schematic diagram of a media content recommendation device according to an embodiment of the present application;
fig. 8 is a structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
It should be noted that the terms "first \ second \ third" related to the embodiments of the present invention are only used for distinguishing similar objects, and do not represent a specific ordering for the objects, and it should be understood that "first \ second \ third" may interchange a specific order or sequence when allowed. It should be understood that the terms first, second, third, etc. used herein are interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in other sequences than those illustrated or described herein.
The existing media content is pushed too coarsely and has single content, and the pushed media content has weak pertinence, so that the accurate pushing of the media content cannot be realized.
In view of the above problems existing at present, the present application provides a media content recommendation scheme, which can recommend accurate media content to a user and can implement accurate push of the media content. The media content recommendation method can be applied to electronic equipment such as mobile terminals and personal computers.
Referring to fig. 1, fig. 1 is a flowchart of a media content recommendation method according to an embodiment of the present application. The present embodiment is described by taking an example in which the media content recommendation method is applied to a mobile terminal. As shown in fig. 1, the media content recommendation method may include the steps of:
s101, acquiring webpage display contents; the webpage display content is embedded with at least one interactive node.
The webpage display content is content displayed on the webpage. The webpage display content can be articles, novel, cartoons, videos and the like. Further, the mobile terminal may obtain the web page display content from the server and display the corresponding content on the interface.
The interaction node may be a node for interacting with a user, i.e. a node capable of receiving and responding to an input of a user. In the interactive node, the mobile terminal can obtain the operation record of the user so as to obtain the input information of the user. The operation record may be an operation record corresponding to selection information, input text, input voice, and the like.
It should be noted that the web page display content may include at least one segment, that is, at least one web page content display segment. Taking a web page novel as an example, the web page content presentation segment may be at least one sentence spoken by the novel character, or at least one sentence of a dialogue. Furthermore, the webpage display content is divided into a plurality of webpage content display segments in a mode of embedding the interactive nodes. And correspondingly entering an interactive node when each webpage content display segment is finished. Furthermore, the mobile terminal displays the webpage content display segment before the target interactive node, then enters the target interactive node, and interacts with the user in the target interactive node, and at this time, the mobile terminal can obtain the user input information of the user in the interactive node, so as to obtain the user interactive information. It should be noted that the "plurality" in the embodiments of the present application may be at least two.
Optionally, the implementation process of S101 may be: the mobile terminal obtains each webpage display content segment corresponding to the webpage display content from the server, the webpage display content segments are displayed one by one according to preset interaction nodes, and after each webpage display content segment is finished, the corresponding interaction node is accessed.
S102, acquiring a to-be-selected media content set matched with the characteristic information of the webpage display content; the set of candidate media content includes at least one media content.
The media content may be content displayed based on multimedia, and may be content in the form of pictures, texts, videos, and the like. Alternatively, the media content may be advertising, promotional content, and the like. For advertisements, the media content may be predetermined content carrying merchandise links. When the media content is displayed in the interface of the mobile terminal, the user can select the media content, and at the moment, the mobile terminal can display a commodity detail page in the interface. And further, after the user purchases the corresponding commodity based on the commodity detail page, the conversion amount corresponding to the media content is increased by one, so that the click rate and the conversion rate corresponding to the media content can be obtained. Alternatively, the media content set is a set composed of a plurality of media contents.
The characteristic information of the web page display content is information for characterizing the characteristics of the web page display content, and may include but is not limited to: the content type of the webpage display content, the related scene, semantic information, interaction information with the user, and user input prediction information for predicting the user input in the interaction node.
Further, the set of media content to be selected may be determined based on the content type of the web page presentation content, scene information, semantic information, user input prediction information of the interaction node, and the like. Taking the web page display content as a web page novel as an example, the content type may be a material type of the novel, for example: literature novels, city life novels, workplace novels, swordsman novels, sentiment novels, and the like; the scene information may be scene information related to a place where the novel event occurs, for example: mall scenes, restaurant scenes, etc.; the semantic information is information obtained by performing semantic recognition on at least one webpage display content fragment, and can be text semantics and the like; the user input prediction information is input prediction information obtained by predicting possible user input in the interaction node, for example: voice information that the user may input.
Optionally, the implementation process of S102 may be: and determining at least one media content matched with the characteristic information of the webpage display content, and integrating the media contents into a media content set to obtain a media content set to be selected.
S103, obtaining user interaction information at a target interaction node in the at least one interaction node.
The target interaction node may be a current interaction node, that is, a current node currently used for interacting with the user on the mobile terminal. Optionally, after entering the target user node, the webpage display content may be displayed or may not be displayed. If the webpage display content is displayed, the webpage display content fragments before the target interaction node can be continuously displayed.
The user interaction information is interaction information generated by a user interacting with the mobile terminal in the interaction node, and may be information generated by interaction in a target interaction node, information generated by interaction in other interaction nodes related to the target interaction node, or other information related to user interaction acquired in the target interaction node. Further, the information generated by the interaction may be an operation record of the user and an object corresponding to the operation record, and taking the user ordering in a certain scene as an example, the user interaction information may be semantic content, a timestamp, and the like corresponding to the voice of the user ordering.
Optionally, the implementation process of S103 may be: and receiving user interaction information generated by voice input of a user to the mobile terminal at the target interaction node through the mobile terminal.
S104, at least one interactive node, screening target media content from the media content set to be selected based on the user interaction information, and outputting the target media content to a user.
As previously mentioned, the user interaction information is information related to user interaction, and the user representation may be determined based on the user interaction information. The user representation may characterize user preferences and may also determine user intent. Therefore, the media content can be screened from the media information set to be selected based on the user interaction information, the screened target media content is matched with the user preference or the user intention, and accurate recommendation of the media content can be effectively guaranteed.
Alternatively, the number of target media contents may be one or more. When the number of the target media contents is two or more, the target media contents can be all output to the user, for example, the target media contents are output one by one according to a certain time sequence or priority; in addition, the priorities corresponding to these target media contents or the matching degrees with the user intentions may be acquired, and the target media contents with the highest priority or the highest matching degree may be output to the user.
Optionally, the mode of outputting the target media content to the user may be matched with the form of the target media content, and when the target media content is a text, a picture, a video and other visual content, the corresponding target media content may be displayed on the interface of the mobile terminal, and at this time, the user may see on the interface; when the target media content is auditory content such as voice, the corresponding target media content can be played in the mobile terminal, and the user can hear the target media content.
Further, the user interaction information may be obtained in the target interaction node, and the target media content may be output to the user before displaying the next webpage display content segment, during the displaying of the next webpage display content segment, or even during the displaying of a plurality of subsequent webpage display content segments. In the interaction process, the method for recommending the media content according to the input of the user has better interaction effect, can play better advertisement putting effect on the advertisement recommending scene, and realizes the accurate pushing of the advertisement.
In an alternative embodiment, the targeted media content may be embedded in the accessed web page and presented in the mobile terminal via a browser. Further, the determined advertisement to be recommended is embedded into the accessed novel webpage, and the advertisement is displayed in the mobile terminal through the browser.
Optionally, the implementation process of S104 may be: the mobile terminal screens out target media content from the media content set to be selected based on the user interaction information to serve as recommended media content; and determining a recommended interactive node in the interactive nodes of the webpage display content, and outputting the target media content to the user in the recommended interactive node. Taking the target media content as an example of an advertisement, at this time, the user may learn corresponding information based on the target media content, for example: item details, item price, etc., and may enter the purchase interface.
According to the media content recommendation method, at least one interactive node is embedded in the webpage display content, a to-be-selected media content set is obtained according to the characteristic information of the webpage display content, the target media content is screened from the to-be-selected media content set based on the user interaction information obtained from the target interactive node, the target media content is output to the user, the target media content is matched with the target interactive node and the interaction information of the user, the pertinence and the accuracy are high, and accurate recommendation of the media content can be achieved. Accordingly, the media content recommendation device, the equipment and the storage medium provided by the application also have the technical effects.
In one embodiment, the web page display content is a web page novel, the web page novel comprises at least one novel segment, and the interactive nodes embedded in the web page novel respectively correspond to the novel segments; the acquiring of the to-be-selected media content set matched with the feature information of the webpage display content comprises the following steps: predicting information input by a user at the target interaction node based on an input information prediction model to obtain user input prediction information corresponding to the target interaction node; obtaining feature information of the webpage novel based on the user input prediction information; and acquiring a to-be-selected media content set matched with the characteristic information of the web novel.
Optionally, the interactive node may be embedded into an end node of each novel segment, that is, the corresponding interactive node is entered after the novel segment is ended. In the interactive node, the user can reply to the dialog message output in the previous novel segment, for example: output in the previous novel segment: "show the dish on the menu, the meat braised in soy sauce, the wall is jumped to buddha, which dish you will choose? ", the user can select favorite dish in the interactive node by voice. The mobile terminal can obtain the user input information accordingly.
Optionally, the input information prediction model may be a network model, and specifically may be a neural network model. The trained input information prediction model is capable of outputting the identified user input information based on the input information. For example: and outputting user input prediction information of the corresponding node based on the input webpage novel and the interactive node information. Further, before determining the set of media content to be selected, an input information prediction model may be obtained based on the web novels of various story topics in the interactive application program and the user input labels corresponding to the interactive nodes. The interactive application may be an application capable of interacting with the user, such as a novel application, a cartoon application, and the like.
According to the embodiment, the information input by the user at the target interactive node is predicted to obtain the user input prediction information corresponding to the target interactive node, and then the to-be-selected media content set is determined by taking the user input prediction information as the characteristic information of the webpage novel, so that the obtained to-be-selected media content set is matched with the user input, and then the target media content with high matching degree can be screened from the to-be-selected media content set based on the user interaction information, and further the accurate pushing of the media content is realized.
In one embodiment, the obtaining the set of media content to be selected for matching with the feature information of the web novel includes: acquiring a media content library; acquiring media content matched with the user input prediction information from the media content library to obtain a first media content set; and integrating to obtain the media content set to be selected according to the first media content set.
The media content library may be a database storing various media contents, and the media contents may be various advertisements accessed to the interactive application program.
Optionally, the implementation manner of obtaining the to-be-selected media content set according to the integration of the first media content set may be to directly determine the first media content set as the to-be-selected media content set, or may perform deduplication processing on media content in the first media content set, and the deduplicated media content set is used as the to-be-selected media content set.
In the embodiment, after the prediction information input by the user is determined, the matched at least one media content is acquired from the media content library based on the prediction information input by the user, and a to-be-selected media content set is constructed. The obtained to-be-selected media content set is obtained by screening based on the prediction information input by the user, and then the target media content with high matching degree can be further screened from the to-be-selected media content set based on the user interaction information, so that the media content can be accurately pushed.
In one embodiment, the obtaining feature information of the web novel based on the user input prediction information includes: acquiring a novel subject matter of the webpage novel; identifying a scene type related to the webpage novel based on a scene identification model; determining a target novel segment corresponding to the target interactive node, and performing semantic recognition on the target novel segment based on a semantic recognition model to obtain novel segment semantics corresponding to the target novel segment; and based on the novel subject matter of the web novel and the related scene type, the novel segment semantics corresponding to the target novel segment and the predicted information input by the user, obtaining the characteristic information of the web novel.
Optionally, the scene recognition model and the semantic recognition model may be network models, and may specifically be neural network models. The trained scene recognition model can output the recognized scene type based on the input information, and the trained semantic recognition model can output the recognized novel segment semantics based on the input information. Further, before the media content set to be selected is determined, a scene recognition model can be obtained based on the web page novel of various novel subjects in the interactive application program and the corresponding scene type labels through training, and a semantic recognition model is obtained based on the web page novel of various novel subjects in the interactive application program and the corresponding semantic labels of novel fragments through training.
Alternatively, the target novel segment may be a novel segment between the target interactive node and the previous interactive node, and may be at least one sentence, i.e., at least one sentence spoken for a novel role. Further, performing semantic recognition on the target novel fragment based on a semantic recognition model to obtain novel fragment semantics corresponding to the target novel fragment, including: and performing semantic recognition on each sentence in the target novel fragment based on the semantic recognition model, and taking the recognized semantics as novel fragment semantics corresponding to the target novel fragment.
In this embodiment, a corresponding media content set is obtained based on a novel subject, a scene type, a novel segment semantic, and a predicted user input, and at least two items in the media content sets are integrated to obtain a media content set to be selected. The determination process of the to-be-selected media content set integrates information of multiple dimensions, so that the media content set is more comprehensive, and further the media content set can be more accurately recommended to the user.
In an embodiment, the obtaining a set of media content to be selected for matching with the feature information of the web novel further includes: acquiring a media content library; acquiring media content matched with the user input prediction information from the media content library to obtain a first media content set; acquiring media content matched with the novel subject matter of the webpage novel from the media content library to obtain a second media content set; acquiring media content matched with the scene type related to the webpage novel from the media content library to obtain a third media content set; acquiring media content matched with the novel fragment semantics corresponding to the target novel fragment from the media content library to obtain a fourth media content set; and integrating to obtain the media content set to be selected according to at least one of the second media content set, the third media content set and the fourth media content set and the first media content set.
Optionally, according to the first media content set and at least one of the second media content set, the third media content set, and the fourth media content set, the implementation process of obtaining the to-be-selected media content set through integration may be:
1) And integrating one of the second media content set, the third media content set and the fourth media content set with the first media content set to obtain a to-be-selected media content set.
2) And performing parallel integration on the first media content set, the second media content set, the third media content set and the fourth media content set to obtain a to-be-selected media content set.
3) And integrating the first media content set, the second media content set, the third media content set and the fourth media content set according to a certain hierarchical structure to obtain a to-be-selected media content set.
In the embodiment, the corresponding media content sets are obtained from different angles such as the story topic, the scene type, the story segment semantics, the predicted user input and the like, and then the media content sets are integrated according to a certain mode to obtain the media content sets to be selected. The obtained candidate media content set integrates information of a plurality of angles, namely, the media content of the plurality of angles is comprehensively contained, and subsequently, the target media content which is as accurate as possible can be obtained from the candidate media content set, so that the accurate pushing of the media content is realized.
In an alternative embodiment, the set of candidate media content may be derived based on the third set of media content. An example of such an implementation is as follows: advertisements adapted to the novel scenes can be directly presented according to the novel scenes, for example, the novel scenes arrive at a clothing store scene, and the clothing advertisements are directly recommended.
In an alternative embodiment, the set of candidate media content may be derived based on the first set of media content and the third set of media content. An example of such an implementation is as follows: the user enters a mall scene in the novel, and the clothing brand advertisement is recommended according to the storefront (clothes) stroked by the user; the user enters a restaurant scene in the novel, and related food advertisements are recommended according to the ordering of the user.
In an embodiment, the integrating the to-be-selected media content set according to the first media content set and at least one of the second media content set, the third media content set, and the fourth media content set includes: and performing duplicate removal on the media contents in the first media content set, the second media content set, the third media content set and the fourth media content set, and arranging and integrating the duplicate-removed media content sets in parallel to obtain the media content set to be selected.
The duplicate removal implementation process may be: determining repeated media contents in the first media content set, the second media content set, the third media content set and the fourth media content set, deleting at least one media content in the repeated media contents, and only retaining one media content.
Optionally, the media contents in the first media content set, the second media content set, the third media content set, and the fourth media content set are subjected to deduplication processing, and the remaining media contents are sorted and sorted, for example, sorted into: clothing media content, video media content, book media content and the like, and the whole after classification and sorting is used as a to-be-selected media content set.
In the embodiment, the first media content set, the second media content set, the third media content set and the fourth media content set are subjected to deduplication processing, the obtained to-be-selected media content set integrates the media content sets at four angles, media contents at multiple angles are comprehensively contained, and then target media contents as accurate as possible can be obtained from the to-be-selected media content set, so that accurate pushing of the media contents is realized.
In one embodiment, the integrating the to-be-selected media content set according to the first media content set and at least one of the second media content set, the third media content set and the fourth media content set includes: classifying and arranging the first media content set, the second media content set, the third media content set and the fourth media content set step by step according to a preset data structure to obtain the media content set to be selected after arrangement; correspondingly, the screening of the target media content from the to-be-selected media content set based on the user interaction information includes: and performing matching analysis on the user interaction information and each media content in the to-be-selected media content set step by step according to the preset data structure, and taking the media content matched with the terminal structure as the target media content.
The matching analysis may be similarity calculation, keyword matching, intention matching, or the like. Further, the process of determining the target media content based on the similarity is exemplified as follows: and calculating the similarity between the user interaction information and the corresponding media content, and further determining the media content with the highest similarity as the matched media content. Additionally, the process of determining target media content based on intent matching is exemplified as follows: two entity words are extracted from the user interaction information, the semantic relation between the entity words is determined, the intention of the user interaction information is obtained by combining the semantic relation, and then the matched media content is obtained based on the intention.
Optionally, the implementation process of performing the step-by-step classification arrangement on the first media content set, the second media content set, the third media content set, and the fourth media content set may be: and determining the second media content set as a root node, determining the third media content set as a secondary node, determining the fourth media content set as a tertiary node, determining the first media content set as a quaternary node, and determining the constructed tree structure and the media content on the tree structure as a media content set to be selected. When constructing nodes of each hierarchy, media contents in the media content collection are classified, so that the classified media contents serve as child nodes of a previous hierarchy.
In an embodiment, the integrating the to-be-selected media content set according to the first media content set and at least one of the second media content set, the third media content set, and the fourth media content set further includes: classifying and arranging the second media content set, the third media content set and the fourth media content set according to a preset data structure to obtain a first-level media content set; arranging the first media content set into the sub-level of at least one media content set in the first level media content set to obtain a second level media content set; obtaining the media content set to be selected after arrangement; correspondingly, the screening of the target media content from the set of media contents to be selected based on the user interaction information includes: performing matching analysis on the user interaction information and each media content in the second-level media content set according to the preset data structure to obtain matched media content; determining a previous level of media content in the first set of level of media content corresponding to the matching media content; and determining the matched media content and/or the upper-level media content as the target media content.
Optionally, before the first media content set, the second media content set, the third media content set, and the fourth media content set are sorted and arranged stage by stage, the first media content set, the second media content set, the third media content set, and the fourth media content set may be subjected to deduplication processing, so that no duplicate media content exists in each tier, and the integrated media content set to be selected is simplified.
In the above embodiment, each media content set is classified and arranged step by step according to a preset data structure, and multi-level and multi-angle classification and integration are performed, so that comprehensive and accurate target media content can be obtained based on user interaction information, and accurate pushing of the media content is further realized.
In one embodiment, the method further comprises: obtaining a set of candidate media content based on the second set of media content, the third set of media content, and the fourth set of media content; acquiring user interaction information at each interaction node of the webpage display content, and determining matched media content based on the user interaction information of each interaction node to obtain a reference media content set; comparing the media contents in the candidate media content set with the media contents in the reference media content set, and when determining that the candidate media content set comprises the media contents which do not exist in the reference media content set, taking the corresponding media contents as the differential media contents; and embedding a new interactive node in the corresponding novel segment based on the difference media content so as to update the interactive node in the webpage novel.
The candidate media content set is a media content set composed of a second media content set, a third media content set and a fourth media content set, and the media content set integrates a novel subject matter, a scene type, novel semantics and the like of a webpage novel. The reference media content set is a media content set matched based on the user interaction information of each interaction node, and is equivalent to a media content set determined after interaction is carried out based on the webpage display content segments corresponding to the interaction nodes.
Therefore, the difference media content obtained by comparing the candidate media content set with the reference media content set is the new media content obtained by performing the pertinence analysis on the novel segment. However, the reference media content set does not have the new media content, which indicates that the new media content does not have the corresponding interactive node, so it is necessary to embed the new interactive node into the novel segment, so as to output the corresponding media content in the new interactive node subsequently. If the interactive novel has a clothing store scene, but no interactive node is set in the clothing store scene, the user cannot input the interactive novel in the clothing store scene, and at the moment, the meaning of the user cannot be interpreted according to the input of the user, so that the advertisement can be accurately pushed according to the input meaning of the user. And although the advertisement can be inserted at the non-set interactive node, the insertion of the advertisement influences the continuity of the broadcast scenario, influences the consistent thought of the user and generates bad use experience, so that the preferred scheme is that the advertisement is not inserted at the non-set interactive node at the current time, and if the advertisement is to be inserted, the effect of modifying the newly-added interactive node is better. By the implementation mode of the embodiment of the application, the output scene of the media content can be expanded, and the accurate pushing of the media content is realized.
Optionally, embedding a new interaction node in the corresponding novel segment based on the difference media content to update the interaction node in the webpage novel, including: an interactive node is newly embedded in the corresponding novel scene based on the differential media content.
Optionally, after the interactive node in the web novel is updated, if an instruction for displaying the web novel again is received, the web novel is displayed based on the updated interactive node, that is, when the novel is gradually pushed into the new embedded interactive node, a to-be-selected media content set corresponding to the new interactive node may be obtained, user interaction information is obtained in the new interactive node, a target media content is determined from the to-be-selected media content set based on the obtained user interaction information, and then the corresponding target media content is output to the user.
In the embodiment, after the difference media content is determined, the display of the web page display content is not directly interrupted, but the interactive node is updated, and the media content is output in the newly added interactive node when the web page novel is displayed again. The method and the device not only can improve the interactive nodes, but also can prevent the output of the media content from influencing the display of the normal webpage novel, and can improve the experience of a user when the user looks over the webpage novel.
In one embodiment, the method further comprises: obtaining a set of candidate media content based on the second set of media content, the third set of media content, and the fourth set of media content; comparing the media contents in the candidate media content set with the media contents in the first media content set, and when determining that the candidate media content set comprises media contents which do not exist in the first media content set, taking the corresponding media contents as difference media contents; and embedding a new interactive node in the corresponding novel segment based on the difference media content so as to update the interactive node in the webpage novel.
Optionally, the candidate media content set is compared with the media content in the first media content set, specifically, the media content set obtained based on the user prediction input information is compared with the media content set obtained based on the story subject, the scene content and the story segment semantics, so that the purpose of obtaining the different media content by comparing the media content sets obtained based on different angles is achieved. The scene that no user input exists but the user input can be embedded can be accurately determined, and then the interactive nodes can be embedded and updated.
In the embodiment, after the difference media content is determined, the display of the web page display content is not directly interrupted, but the interactive node is updated, and the media content is output in the newly added interactive node when the web page novel is displayed again. The interaction node can be more perfect, the influence of the output of the media content on the display of the normal webpage novel can be prevented, and the experience of a user in the process of viewing the webpage novel can be improved.
In one embodiment, the obtaining user interaction information by a target interaction node in the at least one interaction node comprises: receiving current user input information input by a target user corresponding to the target user identification when the target user interacts at the target interaction node, and obtaining the user interaction information; the target user identification is a user identification for logging in the interactive application program; the interactive application program is used for displaying the webpage display content; correspondingly, the screening of the target media content from the set of media contents to be selected based on the user interaction information includes: and screening target media content from the to-be-selected media content set based on the current user input information.
The target user identifier may be a user account registered by the target user in the interactive application program, and the like. In the webpage display content, the target user identification corresponds to the role identification. Taking a webpage novel as an example, the target user identification corresponds to a novel role, and the webpage novel can perform targeted response based on the input of the target user identification, so as to promote the novel plot to develop towards the direction expected by the target user. The current user input information may include voice input, etc.
Optionally, the implementation process of receiving the current user input information input by the target user corresponding to the target user identifier interacting with the target interaction node may be: and acquiring the voice input of a target user and determining a selection target corresponding to the voice input as the input information of the current user. Corresponding target media content may then be determined based on the current user input information, and the target media content may be output to the target user. Taking a restaurant scene in a webpage novel as an example, when ordering a target interaction node, the ordered dish is determined based on the voice input of a target user, when the user selects abalone and sea cucumber porridge, the target media content is determined to be the advertisement related to seafood, and when the user selects preserved egg and lean meat porridge, the target media content is determined to be the advertisement related to eggs.
Optionally, as shown in fig. 2, the novel segments are output one by one, and "the ball, where do you want to try? "then enter target interaction node where the user can make voice input (the user can be alerted by means of an icon, such as" listen in 8230; "indicate that voice input is being received" in fig. 2). As shown in fig. 3, when the character corresponding to the target user identifier says "bright primary school", the next novel segment is determined, and "where, we see" bright primary school.
In the above embodiment, the target media content is screened from the to-be-selected media content set based on the current user input information, so that the screened target media content is a response to the current user input, and therefore, not only can the accuracy of the determined target media content be improved, but also the use experience of the user can be improved.
In one embodiment, the screening target media content from the set of media content to be selected based on the current user input information includes: acquiring historical user input information corresponding to the target user identification; the historical user input information comprises user input information in interaction nodes before the target interaction node by the target user; acquiring a historical operation record corresponding to the target user identifier; the historical operation record is used for recording the historical operation executed by the target user in the interactive application program; and screening target media content from the to-be-selected media content set based on at least one of the historical user input information and the historical operation record and the current user input information.
Optionally, the target media content is screened from the to-be-selected media content set based on the current user input information and the historical user input information, the target media content is also screened from the to-be-selected media content set based on the current user input information and the historical operation record, and the target media content is also screened from the to-be-selected media content set based on the current user input information, the historical user input information and the historical operation record.
In addition, in some embodiments, the user interaction information may also be the current user input information directly, i.e., without regard to historical user input information and historical operating records. Furthermore, when semantic information corresponding to the current user input information is determined, the target media content is determined from the media content set to be selected based on the semantic information, and then the target media content is recommended to the target user.
Optionally, the implementation process of obtaining the historical user input information of the target user identifier may be: and determining all historical interaction nodes before the target interaction node, and determining the user input information acquired from the historical interaction nodes to obtain the historical user input information.
Optionally, the historical operation record may be an operation record corresponding to a click operation, a sliding operation, a text input operation, and the like performed by the target user in the interactive application before entering the target interactive node. In some embodiments, the historical operation record may be a resource transfer operation, i.e., a resource transfer from one resource account to the interactive application, and the resource transfer may be a transfer operation, etc. When the user interaction information is determined, the resource transfer frequency and the resource transfer value of the target user can be determined based on the resource transfer operation of the target user, so that the consumption tendency corresponding to the target user is obtained, and the user portrait of the target user is obtained based on the consumption tendency.
Optionally, based on at least one of the historical user input information and the historical operation record and the current user input information, screening out target media content from the set of media content to be selected, including: performing voice recognition on the current user input information and the historical user input information to obtain target user input information; and screening target media content from the to-be-selected media content set according to the target user input information and the historical operation records.
In the above embodiment, the current user input information generated by the user interacting at the target interaction node is obtained, and the historical user input information and the historical operation record corresponding to the target user are obtained, so that the user portrait of the target user can be obtained accordingly, and the target media content with high matching degree with the target user can be obtained from the media content set to be selected based on the user portrait, for example: according to the input of the user, the side-writing user picture is gentlewoman type, and then when the clothes advertisement is recommended, the gentlewoman type clothes advertisement is found from a plurality of clothes advertisements and is pushed to the user. The accuracy of the recommended target media content can be improved, and the accurate pushing of the media content is realized.
In one embodiment, the screening, at the at least one interaction node, target media content from the set of media content to be selected based on the user interaction information, and outputting the target media content to a user includes: screening target media contents from the media content set to be selected based on the user interaction information; determining a recommended interaction node in the at least one interaction node based on scene types and/or novel fragment semantics related to novel fragments in the webpage novel; and outputting the target media content to the user based on the recommended interaction node.
The recommended interactive node is an interactive node for outputting the target media content, that is, when entering the recommended interactive node, the target media content can be output to the user. The recommended interaction nodes suitable for outputting the target media content can be determined from the interaction nodes of the web novel based on at least one of scene types and novel segment semantics related to novel segments in the web novel.
The target media content can be displayed on a display screen of the terminal device in an interface display mode. Further, the target media content may be displayed in a media content presentation window of the mobile terminal. The media content presentation window may be in the upper region of the display screen of the mobile terminal (as shown in fig. 4), or in other regions. In addition, the media content display window may be overlaid on the web page display content or may be displayed separately from the web page display content (as shown in fig. 4).
Optionally, in the recommendation interaction node, user interaction information may be obtained, a target media content is screened from the to-be-selected media content set based on the user interaction information, and the target media content is output in the interaction node.
In the above embodiment, the target media content is output at the recommended interaction node, that is, the target media content is output after interacting with the user, so that the user can pay more attention to the target media content, and the click rate and the conversion rate of the media content can be improved.
In one embodiment, the outputting the target media content to the user includes: determining a recommended interactive node among the at least one interactive node; generating a media content display window and a media content cancel window which are adjacent in the webpage of the webpage display content based on the recommended interaction node within a preset time period after the user interaction information is acquired; the media content display window is used for displaying the target media content; if the triggering operation of the media content canceling window is received before the deadline time corresponding to the preset time period is reached, closing the target media content and the media content canceling window; otherwise, closing the target media content and the media content canceling window when the deadline is reached.
Alternatively, as shown in fig. 5, a media content presentation window and a media content cancellation window 501 are displayed in the display screen of the mobile terminal. And when the triggering operation of the media content canceling window is received, the mobile terminal closes the media content displaying window and the target media content and the media content canceling window in the media content displaying window. Further, for the case that the media content display window and the web page display content are displayed separately, after the media content cancellation window and the media content display window are closed, the display position of the web page display content may be adjusted, as shown in fig. 6, the content in the web page novel is moved up to the position of the media content display window.
Optionally, the preset time period may be a fixed time period, which may be 1 second, or may be determined in a targeted manner based on the story emergency state of the web novel, where the preset time period when the story is urgent is shorter than the preset time period when the story is relaxed. Broadcasting the advertisement within a plurality of seconds after the user inputs can ensure that the user is in the interactive process, and the advertisement putting effect can be better under the condition of advertisement putting, and the advertisement related to the user input can also be enjoyed to see a little when the user waits for the follow-up plot broadcasting. The intelligence is embodied, and the use effect of a user is not influenced.
According to the embodiment, in the process of outputting the target media content, active closing of the target media content is achieved through a mode of displaying the window, timely closing of the target media content which is not interested by a user is guaranteed, meanwhile, if the user does not operate the target media content, automatic closing can be performed regularly, diversified closing of the target media content is achieved through the two modes, large influence on normal display of the webpage display content is avoided, and user experience can be effectively improved under the condition of achieving accurate pushing.
Optionally, the click proportion and the click time of the target user to the media content cancel window in the interactive application program may be determined, and when the click number proportion is higher than a preset proportion threshold and the click time is smaller than a preset time threshold, it is determined that the user portrait of the target user is: excluding the mode of recommending the media content through the popup in the interactive application program, the times and frequency of outputting the media content in the interactive nodes can be reduced, and only the media content which is most matched with the target user is output in the remaining interactive nodes which can output the media content; the output form of the media content can also be converted, for example, the output form is converted into the form that the target media content is broadcasted in the target interactive node in a voice playing mode; the interactive nodes may also be updated to reduce the number of interactive nodes.
In an optional embodiment, the mobile terminal determines a webpage display content segment before a target interactive node, generates a webpage content display window for displaying the webpage display content segment, acquires the webpage display content segment from a website server by using a webpage link, and displays the webpage display content segment on an interface of an interactive application program by using the webpage content display window; after the webpage display content segments are displayed, a target interaction node is entered, target media content is determined, a media content display window used for displaying the media content is generated, webpage information corresponding to the target media content is obtained from a website server through webpage links, the target media content is displayed on the media content display window, a media content cancellation window is displayed at the same time, when the triggering operation of the media content cancellation window is received, the display of the media content display window and the target media content is cancelled, and then the next webpage display content segments are displayed. In addition, a media content display window can be generated in the process of displaying the next webpage display content segment, webpage information corresponding to the target media content is acquired from the website server by using the webpage link, the target media content is displayed in the media content display window, a media content cancel window is displayed at the same time, and when the triggering operation of the media content cancel window is received, the media content display window and the target media content are canceled and displayed.
According to the embodiment, the webpage display content and the target media content are displayed through the specific window, the user can cancel the display of the target media content based on the media content canceling window, the webpage novel and the advertisement can be displayed intuitively in the scene of recommending the advertisement in the novel, and meanwhile, the normal reading of the user on the novel is not influenced.
In the following, a media content recommendation apparatus provided in an embodiment of the present application is introduced, and a media content recommendation apparatus described below, a media content recommendation method described above, and corresponding technical effects may be mutually referred to.
Referring to fig. 7, fig. 7 is a schematic diagram of a media content recommendation device 700 according to an embodiment of the present application, including:
a display content obtaining module 701, configured to obtain webpage display content; at least one interactive section is embedded in the webpage display content;
a content set acquiring module 702, configured to acquire a to-be-selected media content set used for matching with feature information of the webpage display content; the to-be-selected media content set comprises at least one media content;
an interaction information obtaining module 703, configured to obtain user interaction information at a target interaction node in the at least one interaction node;
a media content output module 704, configured to, at least one interaction node, screen out target media content from the to-be-selected media content set based on the user interaction information, and output the target media content to a user.
According to the media content recommendation device, at least one interactive node is embedded in the webpage display content, a to-be-selected media content set is obtained according to the characteristic information of the webpage display content, the target media content is screened from the to-be-selected media content set based on the user interaction information obtained from the target interactive node, and the target media content is output to the user, is matched with the target interactive node and the user interaction information, has strong pertinence and accuracy, and can be accurately recommended.
In one embodiment, the web page display content is a web page novel, the web page novel comprises at least one novel segment, and the interactive nodes embedded in the web page novel respectively correspond to the novel segments;
a content collection acquisition module comprising:
the prediction information acquisition submodule is used for predicting information input by a user at the target interaction node based on an input information prediction model to obtain user input prediction information corresponding to the target interaction node;
the characteristic information acquisition sub-module is used for acquiring the characteristic information of the webpage novel based on the prediction information input by the user;
and the content set acquisition sub-module is used for acquiring a to-be-selected media content set matched with the characteristic information of the web novel.
In one embodiment, the content collection acquisition sub-module includes:
a first content library acquisition unit for acquiring a media content library;
a content set acquiring unit, configured to acquire media content matching the user input prediction information from the media content library, to obtain a first media content set;
and the first content set integration unit is used for integrating the to-be-selected media content set according to the first media content set.
In one embodiment, the feature information obtaining sub-module includes:
the subject acquisition unit is used for acquiring the novel subject of the webpage novel;
the scene type acquisition unit is used for identifying the scene type related to the webpage novel based on a scene identification model;
a novel fragment semantic acquisition unit, configured to determine a target novel fragment corresponding to the target interaction node, and perform semantic recognition on the target novel fragment based on a semantic recognition model to obtain a novel fragment semantic corresponding to the target novel fragment;
and the characteristic information determining unit is used for obtaining the characteristic information of the webpage novel based on the novel subject matter of the webpage novel, the related scene type, the novel segment semantics corresponding to the target novel segment and the user input prediction information.
In one embodiment, the content collection obtaining sub-module further includes:
a second content library acquisition unit for acquiring a media content library;
a first set acquiring unit, configured to acquire media content matching the user input prediction information from the media content library, so as to obtain a first media content set;
a second set acquiring unit, configured to acquire, from the media content library, media content that matches the novel subject matter of the web novel, to obtain a second media content set;
a third set acquiring unit, configured to acquire, from the media content library, media content that matches the scene type related to the web novel, to obtain a third media content set;
a fourth set acquiring unit, configured to acquire, from the media content library, media content that matches the fiction of the novel segment corresponding to the target novel segment, to obtain a fourth media content set;
and the second content set integration unit is used for integrating the second media content set, the third media content set and the fourth media content set to obtain the to-be-selected media content set according to the first media content set and at least one of the second media content set, the third media content set and the fourth media content set.
In an embodiment, the second content set integration unit is further configured to perform deduplication on media content in the first media content set, the second media content set, the third media content set, and the fourth media content set, and arrange and integrate the deduplicated media content sets in parallel to obtain the media content set to be selected.
In an embodiment, the second content set integration unit is further configured to sort and arrange the second media content set, the third media content set, and the fourth media content set according to a preset data structure to obtain a first hierarchical media content set; arranging the first media content set into a sub-level of at least one media content set in the first level media content set to obtain a second level media content set; obtaining the media content set to be selected after arrangement;
correspondingly, the media content output module is further configured to perform matching analysis on the user interaction information and each media content in the second-level media content set according to the preset data structure to obtain matched media content; determining a previous level of media content in the first set of level of media content corresponding to the matching media content; and determining the matched media content and/or the upper-level media content as the target media content.
In one embodiment, the apparatus further comprises:
a first candidate set obtaining module, configured to obtain a candidate media content set based on the second media content set, the third media content set, and the fourth media content set;
the first difference content acquisition module is used for acquiring user interaction information at each interaction node of the webpage display content, and determining matched media content based on the user interaction information of each interaction node to obtain a reference media content set; comparing the media content in the candidate media content set with the media content in the reference media content set, and when determining that the candidate media content set contains media content which does not exist in the reference media content set, taking the corresponding media content as difference media content;
and the first node updating module is used for embedding a new interactive node in the corresponding novel segment based on the difference media content so as to update the interactive node in the webpage novel.
In one embodiment, the apparatus further comprises:
a second candidate set obtaining module, configured to obtain a candidate media content set based on the second media content set, the third media content set, and the fourth media content set;
a second difference content obtaining module, configured to compare media content in the candidate media content set with media content in the first media content set, and when it is determined that the candidate media content set includes media content that does not exist in the first media content set, take corresponding media content as difference media content;
and the second node updating module is used for embedding a new interactive node in the corresponding novel segment based on the difference media content so as to update the interactive node in the webpage novel.
In one embodiment, the interaction information obtaining module is further configured to receive current user input information input by a target user interacting at the target interaction node and corresponding to the target user identifier, and obtain the user interaction information; the target user identification is a user identification for logging in the interactive application program; the interactive application program is used for displaying the webpage display content;
correspondingly, the media content output module is further configured to screen the target media content from the to-be-selected media content set based on the current user input information.
In one embodiment, a media content output module includes:
a historical input acquisition sub-module, configured to acquire historical user input information corresponding to the target user identifier; the historical user input information comprises user input information in interaction nodes before the target interaction node by the target user;
a history record obtaining submodule, configured to obtain a history operation record corresponding to the target user identifier; the historical operation record is used for recording the historical operation executed by the target user in the interactive application program;
and the first content screening submodule is used for screening target media content from the to-be-selected media content set on the basis of at least one of the historical user input information and the historical operation record and the current user input information.
In one embodiment, a media content output module includes:
the second content screening submodule is used for screening target media content from the media content set to be selected on the basis of the user interaction information;
the first recommendation node determining submodule is used for determining a recommendation interaction node in the at least one interaction node based on the scene type and/or the novel fragment semantics related to the novel fragment in the webpage novel;
and the media content output sub-module is used for outputting the target media content to the user based on the recommended interactive node.
In one embodiment, a media content output module includes:
a second recommended node determining submodule, configured to determine a recommended interaction node in the at least one interaction node;
the window display submodule is used for generating a media content display window and a media content cancel window which are adjacent in the page of the webpage display content based on the recommended interaction node in a preset time period after the user interaction information is obtained; the media content display window is used for displaying the target media content;
a first window closing submodule, configured to close the target media content and the media content cancellation window if a trigger operation for the media content cancellation window is received before a deadline time corresponding to the preset time period arrives;
and the second window closing submodule is used for closing the target media content and the media content canceling window when the deadline is reached otherwise.
In the following, an electronic device provided by an embodiment of the present application is introduced, and a media content recommendation method, a media content recommendation device, and a corresponding technical effect described above may be referred to each other.
The embodiment of the application further provides the electronic equipment. The electronic device may be as shown in fig. 8. FIG. 8 is a block diagram of an electronic device, shown in accordance with an exemplary embodiment, and the contents of the diagram should not be construed as limiting the scope of use of the present application in any way. The electronic device may be a terminal device. As shown in fig. 8, the electronic device 80 may specifically include: a processor 81 and a memory 82.
Among other things, processor 81 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and so forth. The processor 81 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), PLA (Programmable Logic Array). The processor 81 may also include a main processor and a coprocessor, where the main processor is a processor for Processing data in an awake state, and is also called a Central Processing Unit (CPU); a coprocessor is a low power processor for processing data in a standby state. In some embodiments, the processor 81 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed on the display screen. In some embodiments, the processor 81 may further include an AI (Artificial Intelligence) processor for processing computing operations related to machine learning.
The memory 82 may include one or more computer-readable storage media, which may be non-transitory. Memory 82 may also include high-speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In this embodiment, the memory 82 is at least used for storing a computer program 821, wherein after being loaded and executed by the processor 81, the computer program can realize relevant steps in the media content recommendation method executed by the terminal side disclosed in any of the foregoing embodiments. In addition, the resources stored by the memory 82 may also include an operating system 822, data 823, and the like, and the storage may be transient storage or permanent storage. The operating system 822 may include Windows, unix, linux, etc. Data 823 may include, but is not limited to, update information for applications.
In some embodiments, the terminal may also include a display 83, an input-output interface 84, a communication interface 85, sensors 86, a power supply 87, and a communication bus 88.
Those skilled in the art will appreciate that the configuration shown in fig. 8 is not intended to be limiting of terminal devices and may include more or fewer components than those shown.
A storage medium provided by an embodiment of the present application is described below, and a storage medium described below and a media content recommendation method, apparatus, device and corresponding technical effects described above may be mutually referred to.
The embodiment of the application also discloses a storage medium, wherein computer-executable instructions are stored in the storage medium, and when the computer-executable instructions are loaded and executed by a processor, the media content recommendation method disclosed by any one of the embodiments is realized. For the specific steps of the method, reference may be made to the corresponding contents disclosed in the foregoing embodiments, which are not described herein again.
It should be noted that the above-mentioned embodiments are only preferred embodiments of the present application, and are not intended to limit the present application, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device disclosed in the embodiment corresponds to the method disclosed in the embodiment, so that the description is simple, and the relevant points can be referred to the description of the method part.
The principle and the implementation of the present application are explained herein by applying specific examples, and the above description of the embodiments is 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 (13)

1. A method for recommending media contents, comprising:
acquiring webpage display content; at least one interactive node is embedded in the webpage display content;
acquiring a to-be-selected media content set matched with the characteristic information of the webpage display content; the to-be-selected media content set comprises at least one media content;
acquiring user interaction information at a target interaction node in the at least one interaction node;
at least one interactive node, screening target media content from the to-be-selected media content set based on the user interaction information, and outputting the target media content to a user;
the webpage display content is a webpage novel, the webpage novel comprises at least one novel fragment, and the interactive nodes embedded in the webpage novel respectively correspond to the novel fragments;
acquiring a to-be-selected media content set matched with the characteristic information of the webpage display content, wherein the acquiring comprises the following steps of:
acquiring a media content library;
acquiring media content matched with user input prediction information from the media content library to obtain a first media content set;
acquiring media contents matched with the novel topics of the web novel from the media content library to obtain a second media content set;
acquiring media content matched with the scene type related to the webpage novel from the media content library to obtain a third media content set;
acquiring media content matched with the novel fragment semantics corresponding to the target novel fragment from the media content library to obtain a fourth media content set;
integrating to obtain the media content set to be selected according to the first media content set and at least one of the second media content set, the third media content set and the fourth media content set;
further comprising:
obtaining a set of candidate media content based on the second set of media content, the third set of media content, and the fourth set of media content;
acquiring user interaction information at each interaction node of the webpage display content, and determining matched media content based on the user interaction information of each interaction node to obtain a reference media content set; comparing the media content in the candidate media content set with the media content in the reference media content set, and when determining that the candidate media content set contains media content which does not exist in the reference media content set, taking the corresponding media content as difference media content;
or the like, or, alternatively,
comparing the media contents in the candidate media content set with the media contents in the first media content set, and when determining that the candidate media content set comprises media contents which do not exist in the first media content set, taking the corresponding media contents as difference media contents;
and embedding a new interactive node in the corresponding novel segment based on the difference media content so as to update the interactive node in the webpage novel.
2. The method for recommending media contents according to claim 1, wherein said obtaining a set of media contents to be selected for matching with the feature information of the web page display contents comprises:
predicting information input by a user at the target interaction node based on an input information prediction model to obtain user input prediction information corresponding to the target interaction node;
obtaining characteristic information of the web novel based on the user input prediction information;
and acquiring a to-be-selected media content set matched with the characteristic information of the webpage novel.
3. The method for recommending media contents according to claim 2, wherein the obtaining a set of media contents to be selected for matching with the feature information of the novel of the web page comprises:
acquiring a media content library;
acquiring media content matched with the user input prediction information from the media content library to obtain a first media content set;
and integrating to obtain the media content set to be selected according to the first media content set.
4. The method of claim 2, wherein the obtaining feature information of the web novel based on the user input prediction information comprises:
acquiring a novel subject matter of the webpage novel;
identifying a scene type related to the webpage novel based on a scene identification model;
determining a target novel segment corresponding to the target interactive node, and performing semantic recognition on the target novel segment based on a semantic recognition model to obtain novel segment semantics corresponding to the target novel segment;
and based on the novel subject matter of the web novel and the related scene type, the novel segment semantics corresponding to the target novel segment and the predicted information input by the user, obtaining the characteristic information of the web novel.
5. The method of claim 4, wherein the integrating the set of media content to be selected according to at least one of the second set of media content, the third set of media content, and the fourth set of media content, and the first set of media content comprises:
and performing duplicate removal on the media contents in the first media content set, the second media content set, the third media content set and the fourth media content set, and performing parallel arrangement and integration on the duplicate-removed media content sets to obtain the media content sets to be selected.
6. The method of claim 4, wherein the integrating the set of media content to be selected according to at least one of the second set of media content, the third set of media content, and the fourth set of media content, and the first set of media content further comprises:
classifying and arranging the second media content set, the third media content set and the fourth media content set according to a preset data structure to obtain a first-level media content set; arranging the first media content set into a sub-level of at least one media content set in the first level media content set to obtain a second level media content set; obtaining the media content set to be selected after arrangement;
correspondingly, the screening of the target media content from the set of media contents to be selected based on the user interaction information includes:
performing matching analysis on the user interaction information and each media content in the second-level media content set according to the preset data structure to obtain matched media content; determining a previous level of media content corresponding to the matched media content in the first set of level of media content; and determining the matched media content and/or the upper-level media content as the target media content.
7. The media content recommendation method according to any of claims 1 to 6, wherein the obtaining user interaction information at the target interaction node of the at least one interaction node comprises:
receiving current user input information input by a target user corresponding to the target user identification when the target user interacts at the target interaction node, and obtaining the user interaction information; the target user identification is a user identification for logging in the interactive application program; the interactive application program is used for displaying the webpage display content;
correspondingly, the screening of the target media content from the to-be-selected media content set based on the user interaction information includes:
and screening target media contents from the to-be-selected media content set based on the current user input information.
8. The media content recommendation method according to claim 7, wherein the filtering out target media content from the set of media content to be selected based on the current user input information comprises:
acquiring historical user input information corresponding to the target user identification; the historical user input information comprises user input information in interaction nodes before the target interaction node by the target user;
acquiring a historical operation record corresponding to the target user identification; the historical operation record is used for recording the historical operation executed by the target user in the interactive application program;
and screening target media content from the to-be-selected media content set based on at least one of the historical user input information and the historical operation record and the current user input information.
9. The method of any one of claims 2 to 6, wherein the step of, at the at least one interaction node, filtering out target media content from the set of media content to be selected based on the user interaction information, and outputting the target media content to a user, comprises:
screening target media content from the to-be-selected media content set based on the user interaction information;
determining a recommended interaction node in the at least one interaction node based on scene types and/or novel fragment semantics related to novel fragments in the webpage novel;
and outputting the target media content to the user based on the recommended interaction node.
10. The method of any one of claims 2 to 6, wherein the outputting the target media content to the user comprises:
determining a recommended interactive node among the at least one interactive node;
generating a media content display window and a media content cancel window which are adjacent in the webpage of the webpage display content based on the recommended interaction node within a preset time period after the user interaction information is obtained; the media content display window is used for displaying the target media content;
if the triggering operation of the media content canceling window is received before the deadline time corresponding to the preset time period is reached, closing the target media content and the media content canceling window;
otherwise, closing the target media content and the media content cancellation window when the deadline is reached.
11. A media content recommendation device, comprising:
the display content acquisition module is used for acquiring webpage display content; at least one interactive node is embedded in the webpage display content;
the content set acquisition module is used for acquiring a to-be-selected media content set matched with the characteristic information of the webpage display content; the to-be-selected media content set comprises at least one media content;
the interaction information acquisition module is used for acquiring user interaction information at a target interaction node in the at least one interaction node;
the media content output module is used for screening out target media content from the media content set to be selected based on the user interaction information at least one interaction node and outputting the target media content to a user;
the webpage display content is a webpage novel, the webpage novel comprises at least one novel fragment, and the interactive nodes embedded in the webpage novel respectively correspond to the novel fragments;
a content collection acquisition module comprising:
a second content library acquisition unit configured to acquire a media content library;
the first set acquisition unit is used for acquiring media contents matched with user input prediction information from the media content library to obtain a first media content set;
a second set acquiring unit, configured to acquire, from the media content library, media content that matches the novel subject matter of the web novel, to obtain a second media content set;
a third set acquiring unit, configured to acquire, from the media content library, media content that matches the scene type related to the web novel, to obtain a third media content set;
a fourth set acquiring unit, configured to acquire, from the media content library, media content that matches the fiction of the novel segment corresponding to the target novel segment, to obtain a fourth media content set;
a second content set integration unit, configured to integrate, according to the first media content set and at least one of the second media content set, the third media content set, and the fourth media content set, the to-be-selected media content set;
further comprising:
a first candidate set obtaining module, configured to obtain a candidate media content set based on the second media content set, the third media content set, and the fourth media content set;
the first difference content acquisition module is used for acquiring user interaction information at each interaction node of the webpage display content, and determining matched media content based on the user interaction information of each interaction node to obtain a reference media content set; comparing the media content in the candidate media content set with the media content in the reference media content set, and when determining that the candidate media content set contains media content which does not exist in the reference media content set, taking the corresponding media content as difference media content;
or the like, or, alternatively,
a second difference content obtaining module, configured to compare media content in the candidate media content set with media content in the first media content set, and when it is determined that the candidate media content set includes media content that does not exist in the first media content set, take corresponding media content as difference media content;
and the first node updating module is used for embedding a new interactive node in the corresponding novel segment based on the difference media content so as to update the interactive node in the webpage novel.
12. An electronic device, wherein the electronic device comprises a processor and a memory; wherein the memory is adapted to store a computer program which is loaded and executed by the processor to implement the method of any of claims 1 to 10.
13. A storage medium having stored thereon computer-executable instructions which, when loaded and executed by a processor, carry out a method according to any one of claims 1 to 10.
CN202210766812.1A 2022-07-01 2022-07-01 Media content recommendation method, device, equipment and storage medium Active CN115203539B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210766812.1A CN115203539B (en) 2022-07-01 2022-07-01 Media content recommendation method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210766812.1A CN115203539B (en) 2022-07-01 2022-07-01 Media content recommendation method, device, equipment and storage medium

Publications (2)

Publication Number Publication Date
CN115203539A CN115203539A (en) 2022-10-18
CN115203539B true CN115203539B (en) 2023-03-03

Family

ID=83578820

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210766812.1A Active CN115203539B (en) 2022-07-01 2022-07-01 Media content recommendation method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN115203539B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115357704B (en) * 2022-10-19 2023-02-10 深圳市人马互动科技有限公司 Processing method and related device for heterogeneous plot nodes in voice interaction novel
CN115760296B (en) * 2022-11-24 2023-10-17 阿里巴巴(中国)网络技术有限公司 Page data processing and browsing method, terminal equipment and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104581409A (en) * 2015-01-22 2015-04-29 广东小天才科技有限公司 Virtual interactive video playing method and device
CN110618774A (en) * 2019-09-17 2019-12-27 腾讯科技（深圳）有限公司 Progress information generation method and device, electronic device and storage medium
CN113297399A (en) * 2020-03-27 2021-08-24 阿里巴巴集团控股有限公司 Personalized recommendation method, personalized recommendation device and electronic equipment
CN113347498A (en) * 2021-05-28 2021-09-03 北京爱奇艺科技有限公司 Video playing method and device and computer readable storage medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10282752B2 (en) * 2009-05-15 2019-05-07 Excalibur Ip, Llc Computerized system and method for displaying a map system user interface and digital content

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104581409A (en) * 2015-01-22 2015-04-29 广东小天才科技有限公司 Virtual interactive video playing method and device
CN110618774A (en) * 2019-09-17 2019-12-27 腾讯科技（深圳）有限公司 Progress information generation method and device, electronic device and storage medium
CN113297399A (en) * 2020-03-27 2021-08-24 阿里巴巴集团控股有限公司 Personalized recommendation method, personalized recommendation device and electronic equipment
CN113347498A (en) * 2021-05-28 2021-09-03 北京爱奇艺科技有限公司 Video playing method and device and computer readable storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
选择与留白――哔哩哔哩互动视频的两个认知角度;刘思源;《新媒体研究》;20200725(第14期);全文 *

Also Published As

Publication number Publication date
CN115203539A (en) 2022-10-18

Similar Documents

Publication Publication Date Title
CN110134931B (en) Medium title generation method, medium title generation device, electronic equipment and readable medium
US9201959B2 (en) Determining importance of scenes based upon closed captioning data
CN105786977B (en) Mobile search method and device based on artificial intelligence
CN115203539B (en) Media content recommendation method, device, equipment and storage medium
CN107577807B (en) Method and device for pushing information
US20200097538A1 (en) Book analysis and recommendation
CN111818370B (en) Information recommendation method and device, electronic equipment and computer-readable storage medium
JP2019537106A (en) Content recommendation and display
CN113079417B (en) Method, device and equipment for generating bullet screen and storage medium
US10440435B1 (en) Performing searches while viewing video content
CN110781376A (en) Information recommendation method, device, equipment and storage medium
WO2023279843A1 (en) Content search method, apparatus and device, and storage medium
CN110489649B (en) Method and device for associating content with tag
CN109275047A (en) Video information processing method and device, electronic equipment, storage medium
Lavid Ben Lulu et al. Functionality-based clustering using short textual description: Helping users to find apps installed on their mobile device
WO2022247220A9 (en) Interface processing method and apparatus
CN114564666A (en) Encyclopedic information display method, encyclopedic information display device, encyclopedic information display equipment and encyclopedic information display medium
CN112733024A (en) Information recommendation method and device
CN110475158B (en) Video learning material providing method and device, electronic equipment and readable medium
CN109116718B (en) Method and device for setting alarm clock
CN117171406A (en) Application program function recommendation method, device, equipment and storage medium
WO2017135889A1 (en) Ontology determination methods and ontology determination devices
CN115827978A (en) Information recommendation method, device, equipment and computer readable storage medium
CN113704630A (en) Information pushing method and device, readable storage medium and electronic equipment
CN114780790A (en) Content search method, device, equipment and storage medium

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