Comment method and device, medium and computing equipment

Abstract

According to the comment method and device, the medium and the computing equipment, the emotional attitude of the user to the selected multimedia content is visually displayed on the comment page in the form of the emotional label, so that the emotional attitude of the user to the multimedia content and the overall evaluation trend are comprehensively and visually displayed; the method improves the efficiency and satisfaction of the user to obtain information, supports the interaction between the user and the tag display area, allows the user to dynamically update the displayed emotion tag and / or tag popularity through the interaction operation, provides an interaction mode for expressing the emotion attitude of the user more conveniently, quickly and visually, increases the interactivity and participation sense, and improves the user experience. And the immersion of the user in browsing and commenting processes is improved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of network technology, and more specifically, embodiments of the present invention relate to comment methods, apparatuses, media, and computing devices. Background Art

[0002] This section is intended to provide a background or context to embodiments of the invention that are recited in the claims. No description herein is admitted to be prior art by inclusion in this section.

[0003] In current digital content consumption scenarios, such as interactive scenarios of multimedia in the form of music, articles, videos or other social dynamics, users' interaction and feedback methods for different types of multimedia content are becoming increasingly diverse.

[0004] At present, common interaction and feedback methods include at least one of the like / collection function and the picture and text comment function. The like / collection function can reflect the popularity of multimedia content, but its emotional expression is single and users cannot express more complex emotions about multimedia content; the picture and text comment function allows users to express their feelings by posting text, uploading pictures or videos, but it takes a long time to manually input text and pictures in the picture and text comment function, and the accumulation of a large number of single comments causes users to spend more time and energy to screen and read when browsing, which increases the information processing burden of users and reduces the efficiency of user interaction.

[0005] The above two methods have limitations in satisfying the public's need to express their diverse emotional attitudes towards multimedia content conveniently and quickly. It is difficult to form a strong emotional resonance, which limits the depth of interaction between users and multimedia content and affects the efficiency of user interaction and overall satisfaction. Summary of the invention

[0006] In view of this, in order to solve the above problems, the embodiments of the present invention provide a comment method, apparatus, medium and computing device.

[0007] In a first aspect of the embodiments of the present invention, a comment method is provided, the method comprising:

[0008] Displaying a comment page for the selected multimedia; the comment page includes a tag display area for displaying emotion tags, the tag display area displays emotion tags and tag heat of each emotion tag; the emotion tags are obtained by semantically analyzing the comment text of the selected multimedia;

[0009] In response to an interactive operation on the tag display area, the emotion tag and / or tag heat displayed in the tag display area is updated.

[0010] Optionally, the method further comprises a step of pre-acquiring emotion labels:

[0011] Acquire all comment texts of the selected multimedia; use a pre-built emotion classification model to perform semantic analysis on the comment texts to obtain emotion classification results corresponding to each comment text; the emotion classification results include the confidence corresponding to each emotion classification preset in the emotion classification model; select the emotion label corresponding to the emotion classification with the highest confidence in the emotion classification results as the emotion label corresponding to the comment text.

[0012] Optionally, the method further comprises:

[0013] For the emotion tag corresponding to each comment text of the selected multimedia, the number of comment texts corresponding to the same emotion tag is determined as the tag heat of the emotion tag.

[0014] Optionally, for different types of multimedia, the emotion classification model pre-sets emotion classification sets that match different multimedia content classification categories of the multimedia; the multimedia content category refers to classification according to the content theme or style of the multimedia; the method further includes:

[0015] The multimedia content category of the selected multimedia is input into the emotion classification model; the emotion classification model obtains an emotion classification set matching the multimedia content category, and uses each emotion classification included in the emotion classification set as an emotion classification preset in the emotion classification model.

[0016] Optionally, in response to the interactive operation on the tag display area, updating the emotion tag and / or tag heat displayed in the tag display area includes:

[0017] In response to a first selection operation on any emotion tag displayed in the tag display area, the tag heat value of the first emotion tag selected by the first selection operation is set to increase by a first preset value, and the tag display area is updated to display the increased tag heat of the first emotion tag.

[0018] Optionally, the method further comprises:

[0019] A selection animation effect related to the first emotion tag is displayed on the comment page; wherein the selection animation effect includes an animation effect for an emoticon; and the emoticon matches the emotion represented by the first emotion tag.

[0020] Optionally, the animation effect for the emoticon includes at least any one of the following:

[0021] The expression explosion animation effect for the emoticon is used to generate multiple identical emoticons on the terminal screen starting from the selection position of the first selection operation, and fade out of the terminal screen after moving along a set path; the full-screen expression instant effect for the emoticon is used to expand the emoticon to the entire terminal screen at a set speed, and fade out of the terminal screen after full-screen display for a set time.

[0022] Optionally, in response to an interactive operation on the label display area, updating the emotion label and / or label heat displayed in the label display area includes: in response to an emotion label adding operation on the label display area, displaying a pre-set emotion label library; the emotion label library includes a plurality of predefined candidate emotion labels; in response to a confirmation adding operation on at least one candidate emotion label, updating the label area to display the at least one newly added candidate emotion label and its label heat; the label heat of the at least one candidate emotion label is taken as an initialization value.

[0023] Optionally, the tag display area is provided with a tag adding control, and the emotion tag adding operation includes a touch operation acting on the tag adding control or a selection operation of an external input device.

[0024] Optionally, in response to interactive operations on the label display area, updating the emotion label and / or label heat displayed in the label display area includes: displaying a custom editing interface in response to a custom operation on an emotion label acting on the label display area; receiving emotion text input by a user through the custom editing interface; generating a corresponding custom emotion label based on the emotion text, and updating the label display area to display the newly added custom emotion label and its label heat; the label heat of the custom emotion label is taken as an initialization value.

[0025] Optionally, the custom editing interface provides a plurality of emoticons for expressing different emotions; the method further comprises:

[0026] In response to a second selection operation for any emoticon, a corresponding custom emotion tag is generated according to the emotion text and the emoticon selected by the second selection operation.

[0027] Optionally, the method further includes: visually updating the target emotion tag on the comment page to show that the target emotion tag is in a selected state; the target emotion tag includes the first emotion tag, the at least one newly displayed candidate emotion tag, and the custom emotion tag.

[0028] Optionally, the method also includes: in response to any user's new comment posting operation on the selected multimedia, obtaining a new comment text corresponding to the new comment posting operation; using a pre-built emotion classification model to perform semantic analysis on the new comment text to obtain a second emotion label corresponding to the new comment text; updating the label display area to display the second emotion label and its label popularity.

[0029] Optionally, updating the label display area includes:

[0030] In the case where the second emotion label is an emotion label that has been displayed in the label display area, setting the label heat value of the second emotion label to increase by a second preset value, and updating the label display area to display the increased label heat of the second emotion label;

[0031] When the second emotion tag is not displayed in the tag display area, the tag heat value of the second emotion tag is set to an initialization value, and the tag display area is updated to display the newly added second emotion tag and its tag heat.

[0032] Optionally, the method also includes: in response to a like operation by any user on any comment text of the selected multimedia, setting the label heat value of the third emotion label corresponding to the liked comment text associated with the like operation to increase by a second preset value, and updating the label display area to display the increased label heat of the third emotion label.

[0033] Optionally, the method also includes: in response to an interaction detail viewing operation for any emotion tag in the tag display area, displaying other account information of a fourth emotion tag selected in the history of the interaction detail viewing operation; the other account information supports viewing the account homepage indicated by the other account information.

[0034] Optionally, the method further includes: in response to a resonance interaction operation for any of the other account information, sending a resonance prompt message to other accounts indicated by the other account information; the resonance prompt message is used to prompt the other account that there is a user emotional resonance with the emotion tag under the selected multimedia.

[0035] Optionally, displaying the comment page of the selected multimedia includes: arranging and displaying each emotion tag and its tag heat in the tag display area in descending order of the tag heat corresponding to the emotion tag of the selected multimedia.

[0036] Optionally, the tag display area is used to display a set number of emotion tags; the method also includes: when the total number of emotion tags of the selected multimedia exceeds the set number, in response to an emotion tag expansion operation, loading and displaying all emotion tags of the selected multimedia.

[0037] Optionally, the comment page also includes a text display area for displaying comment text; the method also includes: in response to a comment filtering operation based on the emotion tag, obtaining all comment texts matching the fifth emotion tag selected in the comment filtering operation according to the stored comment texts and their corresponding emotion tags, and displaying them in the text display area.

[0038] In a second aspect of the embodiments of the present invention, a comment device is provided, the device comprising:

[0039] An emotion tag viewing module is used to display a comment page of the selected multimedia; the comment page includes a tag display area for displaying emotion tags, the tag display area displays emotion tags and the tag heat of each emotion tag; the emotion tags are obtained by semantically analyzing the comment text of the selected multimedia;

[0040] The emotion tag interaction module is used to update the emotion tags and / or tag heat displayed in the tag display area in response to the interaction operation on the tag display area.

[0041] Optionally, the device further comprises a step of pre-acquiring emotion tags:

[0042] Obtaining all comment texts of the selected multimedia;

[0043] Using a pre-built emotion classification model to perform semantic analysis on the comment text, obtaining an emotion classification result corresponding to each comment text; the emotion classification result includes a confidence level corresponding to each emotion classification preset in the emotion classification model;

[0044] The emotion label corresponding to the emotion classification with the highest confidence in the emotion classification results is selected as the emotion label corresponding to the comment text.

[0045] Optionally, the device further comprises: for the emotion tag corresponding to each comment text of the selected multimedia, determining the number of comment texts corresponding to the same emotion tag as the tag heat of the emotion tag.

[0046] Optionally, for different types of multimedia, the emotion classification model pre-sets emotion classification sets that match different multimedia content classification categories of the multimedia; the multimedia content category refers to classification according to the content theme or style of the multimedia; the device also includes:

[0047] The multimedia content category of the selected multimedia is input into the emotion classification model; the emotion classification model obtains an emotion classification set matching the multimedia content category, and uses each emotion classification included in the emotion classification set as an emotion classification preset in the emotion classification model.

[0048] Optionally, the emotion label interaction module is specifically used to:

[0049] In response to a first selection operation on any emotion tag displayed in the tag display area, the tag heat value of the first emotion tag selected by the first selection operation is set to increase by a first preset value, and the tag display area is updated to display the increased tag heat of the first emotion tag.

[0050] Optionally, the device further comprises:

[0051] A selection animation effect related to the first emotion tag is displayed on the comment page; wherein the selection animation effect includes an animation effect for an emoticon; and the emoticon matches the emotion represented by the first emotion tag.

[0052] Optionally, the animation effect for the emoticon includes at least any one of the following:

[0053] The expression explosion animation effect for the emoticon is used to generate multiple identical emoticons on the terminal screen starting from the selection position of the first selection operation, and fade out of the terminal screen after moving along a set path; the full-screen expression instant effect for the emoticon is used to expand the emoticon to the entire terminal screen at a set speed, and fade out of the terminal screen after full-screen display for a set time.

[0054] Optionally, the emotion label interaction module is specifically used to:

[0055] In response to an emotion tag adding operation performed on the tag display area, a pre-set emotion tag library is displayed; the emotion tag library includes a plurality of predefined candidate emotion tags; in response to a confirmation adding operation performed on at least one candidate emotion tag, the tag area is updated to display the at least one newly added candidate emotion tag and its tag heat; the tag heat of the at least one candidate emotion tag is taken as an initialization value.

[0056] Optionally, the tag display area is provided with a tag adding control, and the emotion tag adding operation includes a touch operation acting on the tag adding control or a selection operation of an external input device.

[0057] Optionally, the emotion tag interaction module is specifically used to: respond to the emotion tag customization operation acting on the tag display area, display a custom editing interface; receive emotion text input by the user through the custom editing interface; generate a corresponding custom emotion tag based on the emotion text, and update the tag display area to display the newly added custom emotion tag and its tag heat; the tag heat of the custom emotion tag is taken as an initialization value.

[0058] Optionally, the custom editing interface provides a variety of emoticons for expressing different emotions; the device also includes:

[0059] In response to a second selection operation for any emoticon, a corresponding custom emotion tag is generated according to the emotion text and the emoticon selected by the second selection operation.

[0060] Optionally, the device also includes: visually updating the target emotion tag on the comment page to show that the target emotion tag is in a selected state; the target emotion tag includes the first emotion tag, the at least one newly displayed candidate emotion tag, and the custom emotion tag.

[0061] Optionally, the device also includes: a new comment text acquisition module, which is used to respond to any user's new comment posting operation on the selected multimedia and obtain a new comment text corresponding to the new comment posting operation; a semantic analysis module, which is used to perform semantic analysis on the new comment text using a pre-built emotion classification model to obtain a second emotion label corresponding to the new comment text; and an update module, which is used to update the label display area to display the second emotion label and its label popularity.

[0062] Optionally, the update module is specifically used to:

[0063] In the case where the second emotion label is an emotion label that has been displayed in the label display area, setting the label heat value of the second emotion label to increase by a second preset value, and updating the label display area to display the increased label heat of the second emotion label;

[0064] When the second emotion tag is not displayed in the tag display area, the tag heat value of the second emotion tag is set to an initialization value, and the tag display area is updated to display the newly added second emotion tag and its tag heat.

[0065] Optionally, the device also includes: in response to a like operation by any user on any comment text of the selected multimedia, setting the label heat value of the third emotion label corresponding to the liked comment text associated with the like operation to increase by a second preset value, and updating the label display area to display the increased label heat of the third emotion label.

[0066] Optionally, the device also includes: in response to an interaction detail viewing operation for any emotion tag in the tag display area, displaying other account information of a fourth emotion tag selected in the history of the interaction detail viewing operation; the other account information supports viewing the account homepage indicated by the other account information.

[0067] Optionally, the device also includes: in response to a resonance interaction operation for any of the other account information, sending a resonance prompt message to other accounts indicated by the other account information; the resonance prompt message is used to prompt the other account that there is a user emotional resonance with the emotion tag under the selected multimedia.

[0068] Optionally, the emotion tag viewing module is specifically used to: display the emotion tags and their tag heats in the tag display area in order in descending order of the tag heats corresponding to the emotion tags of the selected multimedia.

[0069] Optionally, the tag display area is used to display a set number of emotion tags; the device also includes: when the total number of emotion tags of the selected multimedia exceeds the set number, in response to an emotion tag expansion operation, loading and displaying all emotion tags of the selected multimedia.

[0070] Optionally, the comment page further includes a text display area for displaying the comment text; the device further includes:

[0071] In response to the comment filtering operation based on the emotion tag, all comment texts matching the fifth emotion tag selected in the comment filtering operation are obtained according to the stored comment texts and their corresponding emotion tags, and are displayed in the text display area.

[0072] In a third aspect of the embodiments of the present invention, a readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned comment method is implemented.

[0073] In a fourth aspect of the embodiments of the present invention, a computing device is provided, comprising: a processor and a memory; the memory is used to store a computer program; the processor is used to execute the above-mentioned comment method by calling the computer program.

[0074] The embodiments of the present invention provide a comment interaction method suitable for different types of multimedia, which intuitively displays the user's emotional attitude towards selected multimedia content in the form of emotional tags on the comment page. It not only analyzes the comment text to generate emotional tags, but also displays the popularity of each emotional tag in real time, thereby achieving a comprehensive and intuitive display of the user's emotional attitude and overall evaluation trend towards multimedia content, improving the efficiency and satisfaction of users in obtaining information, and at the same time supporting the interaction between the user and the tag display area, allowing the user to dynamically update the displayed emotional tags and / or tag popularity through interactive operations, providing a more convenient, fast and intuitive interactive method for expressing the user's emotional attitude, increasing interactivity and participation, and enhancing the user's immersion in the browsing and commenting process. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:

[0076] Figure 1 A schematic diagram of a comment interaction and feedback method using a video as an example in the related art is shown schematically;

[0077] Figure 2A A flowchart of a review method according to an embodiment of the present invention is schematically shown;

[0078] Figure 2B A schematic diagram of a comment page layout of a selected multimedia according to an embodiment of the present invention is schematically shown;

[0079] Figure 3 A schematic diagram of comparing the display of emotion labels before and after responding to the first selection operation according to an embodiment of the present invention is schematically shown;

[0080] Figure 4A A static diagram of an expression explosion animation effect of a smiley face symbol according to an embodiment of the present invention is schematically shown;

[0081] Figure 4B A schematic diagram of a full-screen instant expression effect of a smiley face symbol according to an embodiment of the present invention is schematically shown;

[0082] Figure 5 A schematic diagram of an update process for an emotion tag adding operation according to an embodiment of the present invention is schematically shown;

[0083] Figure 6 A schematic diagram of an update process for a custom operation of an emotion tag according to an embodiment of the present invention is schematically shown;

[0084] Figure 7 A schematic diagram of indicating that an emotion tag is selected through visual updating according to an embodiment of the present invention is schematically shown;

[0085] Figure 8 Schematically shows an interactive diagram of a process of updating a label display area based on newly posted comments according to an embodiment of the present invention;

[0086] Fig. 9 A schematic diagram of an interactive process for achieving emotional resonance according to an embodiment of the present invention is shown;

[0087] Fig.10 A block diagram of a comment device according to an embodiment of the present invention is schematically shown;

[0088] Fig.11 A schematic diagram of a medium according to an embodiment of the present invention is schematically shown;

[0089] Fig.12 A schematic diagram of a computing device according to an embodiment of the present invention is schematically shown.

[0090] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0091] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0092] In the current digital content consumption scenario, various platforms or programs not only provide users with massive, high-quality multimedia content choices, but also enhance user participation and satisfaction by continuously improving the interaction and feedback mechanism between users and different types of multimedia content.

[0093] At present, common ways of interaction and feedback between users and multimedia include like / collection functions and picture and text comment functions. The number of likes / collections can be used as a rough indicator of the popularity of content. The like function can express the user's positive recognition or love of multimedia content. The collection function further meets the personalized needs of users and allows users to save content of interest, reflecting the user's personal interests. The picture and text comment function allows users to post text comments and upload pictures or videos as auxiliary explanations, allowing users to express their views and feelings intuitively, which to some extent makes up for the shortcomings of the like / collection function.

[0094] For example, Figure 1 An example of an interactive method for a video is shown, in which a user can express his or her likes by clicking the like / favorite button, share the video with others by clicking the share button, and express his or her own opinions by posting text comments in the comment area.

[0095] However, the above-mentioned interaction method based on the like / collection function has a single emotional expression, and users cannot express more complex emotions towards multimedia content. In the picture and text comment function, it takes a long time to manually input text and pictures. Moreover, as the number of comments continues to increase, the accumulation of a large number of single comments causes users to spend more time and energy to screen and read when browsing, and it is easy to miss valuable or emotionally resonant comments, which increases the user's information processing burden and reduces the efficiency of user interaction. It can be seen that the current multimedia interaction and feedback methods still have limitations, that is, they cannot meet the public's needs for conveniently and quickly expressing their diverse emotional attitudes towards multimedia content, and it is difficult to form a strong emotional resonance, which affects the efficiency and satisfaction of user interaction.

[0096] In view of this, the present invention proposes a comment method suitable for commenting on different types of multimedia such as songs, videos, social dynamics, articles, posts, etc. The method optimizes the presentation of multimedia content comment pages by introducing a display mechanism of emotional tags and tag heat, thereby improving the interactive experience and satisfaction between users and multimedia content.

[0097] Specifically, the method performs semantic analysis on the comment text of the selected multimedia, extracts emotional tags to reflect the user's emotional attitude towards the multimedia content, and combines the tag popularity to intuitively and comprehensively display the public's diverse attitudes towards the selected multimedia, making it easy for users to quickly understand the public's overall emotional tendency towards the selected multimedia. It also provides interaction and response between users and the emotional tags displayed on the comment page, allowing users to express their attitudes towards the selected multimedia more conveniently and intuitively, thereby promoting in-depth interaction between users and multimedia content.

[0098] After introducing the basic principles of the present invention, various non-limiting embodiments of the present invention are described in detail below.

[0099] Exemplary Methods

[0100] Reference below Figures 2A to 12To describe the comment method according to an exemplary embodiment of the present invention. The method can be widely applied to various multimedia content interactive programs or platforms that support posting user comments, such as video websites, social media platforms, online forums, song playback software, and article / post sharing websites. It is understandable that the application scenarios described here are only shown to facilitate understanding of the spirit and principles of the present invention, and the embodiments of the present invention are not limited in this respect. On the contrary, the embodiments of the present invention can be applied to any applicable scenario.

[0101] refer to Figure 2A , Figure 2A A comment method according to an embodiment of the present invention is schematically shown, and the method may at least include the following steps:

[0102] S201, displaying a comment page of the selected multimedia; the comment page includes a tag display area for displaying emotion tags, the tag display area displays emotion tags and the tag heat of each emotion tag; the emotion tags are obtained by semantically analyzing the comment text of the selected multimedia;

[0103] Selected multimedia refers to multimedia that the current user requests to view the comment page through a terminal device (such as a mobile phone, computer, tablet, etc.). The selected multimedia may include but is not limited to any type of multimedia such as songs, videos, articles, posts or social dynamics. The specific content of the selected multimedia is determined according to the application currently being used by the user and the operation location that triggers the viewing of the comment page. For example, in a song playing program, the user clicks the comment control on the song playing interface, and the selected multimedia is the song that is being played or paused on the song playing interface; on a video or article sharing platform, the user swipes up to the bottom of the video or article display interface to view the comments, and the selected multimedia is the currently displayed video or article.

[0104] The comment page of the selected multimedia is used to display the comment information related to the selected multimedia content. In this embodiment, a label display area is set on the comment page, which is dedicated to displaying the emotion labels related to the selected multimedia content and the label heat of each emotion label. The emotion label represents the diverse emotional attitudes of the public towards the selected multimedia content, and the label heat of the emotion label represents the popularity / popularity of the emotion label.

[0105] Among them, the emotion tag can generate matching tags by semantically analyzing all the comment texts of the selected multimedia, which are used to represent the emotional tendency expressed in the comment text, such as "happy", "shocked", "sad", "angry", etc. It can be understood that the emotion tags described in this embodiment include not only the user's emotional feelings, but also a broader, multi-dimensional emotional attitude and state description, which may include but is not limited to the evaluation of the content of the selected multimedia, the user status (such as studying, exercising, good morning), etc.

[0106] For example, for a song interaction scenario, the emotional label may include emotional feelings such as happy, wanting to cry, and healed, song evaluations such as the melody is catchy, the lyrics are touching, and the rhythm is dynamic, as well as user states such as good morning, insomnia, and studying.

[0107] The emotion tag may include only text content used to express emotion or status, or may include both text content and emoticons that match the emotion represented by the text content; the tag heat of each emotion tag may be updated and determined in real time by counting the number of texts in the comment text that express the emotional tendency indicated by the emotion tag, combined with other interactive operations that trigger changes in the tag heat value.

[0108] The comment page can be displayed in response to a comment viewing operation trigger, and the comment viewing operation may include but is not limited to user operations such as clicking a comment viewing trigger control or link, sliding the screen to a specific position, and long pressing a page element. The specific operation mode of the comment viewing operation is determined according to the type of interactive interface setting of the application, and the present invention does not limit this. For example, for a song playing program, the comment page of the selected song can be triggered to be displayed by clicking the comment control of the song playing interface; for a news browsing web page, the page layout can be adjusted by sliding the news page when the sliding direction and distance reach the threshold for displaying the comment page, and the comment page content can be dynamically loaded and displayed.

[0109] Regarding the emotion label, all the comment texts of the selected multimedia can be obtained in advance, and a pre-built and trained emotion classification model can be used to perform semantic analysis on all the comment texts of the selected multimedia to obtain the emotion label corresponding to each comment text, and the same emotion labels can be merged to obtain the emotion label of the selected multimedia and the label heat of each emotion label.

[0110] Among them, the emotion classification model pre-sets multiple emotion classifications, each emotion classification corresponds to an emotion label, and the text set annotated with emotion labels can be used as a training sample to train the emotion classification model to identify the emotional tendencies in different texts. The text set annotated with emotion labels can include comment texts from different fields annotated with emotion labels, such as movie reviews, social media posts, news reviews, etc. This type of data set is manually or automatically annotated into different emotion categories. In order to better adapt the training model to real-world application scenarios, the comment texts of the training samples can include real user comments on platforms such as social media, forums, and blogs.

[0111] The application program where the selected multimedia is located is configured with a data storage space for storing the relevant data information of each multimedia content, which may include but is not limited to the multimedia identification, multimedia comment text, emotional tags and their tag heat, etc. All the comment texts of each multimedia are obtained in advance for semantic analysis, and the corresponding emotional tags and their tag heat are obtained and stored in the data storage space. When the user operation triggers the comment text, or the emotional tags and their tag heat change, the stored relevant data information is synchronized in real time to maintain the consistency of the information displayed on the comment interface and the data storage space.

[0112] Based on the data storage space, when a comment page of a selected multimedia is requested, the matching emotion tag and its tag heat can be searched in the data storage space through the multimedia identifier of the selected multimedia, and rendered and displayed on the comment interface.

[0113] The comment page may also include a text display area for displaying the comment text of the selected multimedia, which is dedicated to displaying the comments made by the public on the content of the selected multimedia. When displaying the comment page of the selected multimedia, a request is made to the application program where the selected multimedia is located to obtain all the comment text of the selected multimedia, and the obtained comment text is loaded and displayed in the text display area.

[0114] In order to enable those skilled in the art to better understand the above review page, Figure 2B A schematic diagram of a selected multimedia comment page is shown as an example. Figure 2B As shown, a specific area is set at the top of the comment page to display the emotion tags and tag heat of the selected multimedia. In addition, a text display area is also set at the bottom of the comment page to display the comments made by the public on the selected multimedia.

[0115] Regarding displaying the comment text of the selected multimedia in the text display area, the text display area can display the comment text in order according to the release time and the number of likes of the comment text, or can also display it in a random order. Among them, the number of likes represents the cumulative number of times a single comment text has been liked by different users. Figure 2B In the example of a single comment text in the text display area shown, in order to measure the popularity of the comment among users, a like control is also provided in the display area of ​​each comment text in the text display area, which can be implemented in the form of a button or icon. In response to a selection operation on the like control (such as a touch operation on the control through a touch screen, or a click operation on the control through an external input device), the number of likes for the comment text increases by one. That is, when a user is browsing the comment text in the text display area and sees content that he or she is interested in or thinks is valuable, the user can choose to click the "like" button / icon to express approval or support. The like operation will be recorded and the number of likes corresponding to the comment text will increase accordingly each time a like operation occurs.

[0116] S202: In response to an interactive operation on the tag display area, update the emotion tag and / or tag heat displayed in the tag display area.

[0117] In this embodiment, under the premise that the comment interface displays the emotion tags and their tag heat, in order to provide users with a more convenient and quick way to express their emotional attitudes towards the selected multimedia content, a variety of interactive operations related to emotion tags and an update process associated with each interactive operation are provided. Users can use different interactive operations to achieve dynamic updates of emotion tags and / or tag heat in different ways.

[0118] The interactive operation may include at least three types: a selection operation acting on each emotion tag displayed in the tag display area, or an emotion tag adding operation that triggers adding emotion tags from the emotion tag library, or an emotion tag customization operation that triggers an emotion tag customization operation implemented through a custom interface. Different interactive operations correspond to different processing flows for updating emotion tags / tag heat, thereby realizing different interaction methods with emotion tags and meeting the personalized needs of users' emotional attitudes. For example, in response to a selection operation acting on an emotion tag, the tag heat of the selected emotion tag is adjusted; in response to an emotion tag customization operation, a user-defined emotion tag is obtained through a series of steps and displayed in the tag display area together with the displayed emotion tags.

[0119] It can be understood that the updating of the emotion labels and / or label heat displayed in the label display area described in this step is not a single action execution, but a complex and flexible update processing flow. The update processing flow matches different types of interactive operations. For different interactive operations, different methods are used to achieve the updating of the emotion labels and / or label heat displayed in the label display area.

[0120] In the disclosed embodiment, the emotional tendencies of the public's comment text on the selected multimedia are obtained by analysis, and are intuitively displayed on the comment page in the form of emotional tags. The popularity of the emotional tags is quantified in combination with the tag heat, thereby comprehensively and intuitively displaying the public's diverse emotional attitudes and overall emotional tendencies towards the content of the selected multimedia, providing an intuitive visual presentation of emotional feedback, which is convenient for users to quickly understand the public's emotional tendencies. In addition, on the premise of displaying emotional tags, interactive operations on the tag display area and a response process for real-time updating of emotional tags and tag heat corresponding to the interactive operations are provided, so that the user's emotional expression is not limited to a single like or collection. The user can quickly and directly express complex and diverse emotional attitudes by selecting or creating tags, thereby enhancing the depth of interaction and emotional resonance, and thereby improving the overall satisfaction of user interaction.

[0121] In addition, interactive operations with the label display area can express users' emotional attitudes more conveniently and quickly. Compared with manually writing comment text, it reduces the time and energy of user input, lowers the usage threshold and operation complexity for users to express complex emotional attitudes, and improves the efficiency of user interaction.

[0122] In some embodiments, for selected multimedia such as video, audio, picture, social dynamics, etc., based on the above-mentioned embodiment, semantic analysis is performed on all comment texts of the selected multimedia to obtain emotion tags related to the selected multimedia. This embodiment uses a pre-built emotion classification model to achieve pre-acquisition of emotion tags, which may at least include the following steps a1-a3:

[0123] a1, obtaining all comment texts of the selected multimedia;

[0124] The comment text can be read from the database of the application program where the selected multimedia is located. For example, for song A in a song playing program, all comment texts of the song A are read from the database of the song playing program.

[0125] a2, using a pre-built emotion classification model to perform semantic analysis on the comment text to obtain an emotion classification result corresponding to each comment text; the emotion classification result includes a confidence level corresponding to each emotion classification preset in the emotion classification model;

[0126] The emotion classification model adopted in this embodiment can be a comprehensive classification model suitable for the comment text analysis of various types of multimedia, or a model dedicated to the comment text analysis of each type of multimedia can be set separately. When analyzing the comment text of the selected multimedia, the emotion classification model dedicated to the selected multimedia is called.

[0127] The sentiment classification model can be implemented using machine learning or deep learning models. After being trained with a large amount of sentiment-labeled comment data, it can identify the type of emotion expressed in the text. The sentiment classification model parses the input comment text and, based on its internal algorithm and training data, calculates the confidence corresponding to each sentiment classification preset in the model for each comment text. The confidence indicates the degree of certainty of the model that the comment belongs to a certain sentiment classification.

[0128] For example, the song A includes 1,000 comment texts, which are respectively input into the emotion classification model so that the emotion classification model generates 1,000 corresponding emotion classification results. Assuming that 20 emotion classifications are set in the model, each emotion classification result will include the confidence level for each of the 20 emotion classifications, and the sum of the 20 confidence levels is 1.

[0129] To facilitate understanding of the emotion classification results, take the five emotion classifications preset in the model as an example: Classification 1 - Relaxation, Classification 2 - Excitement, Classification 3 - Sensation, Classification 4 - Good to Burn, Classification 5 - Single Loop. Each emotion classification result includes the five classifications and their confidence levels, such as:

[0130] Comment 1:

[0131] [Classification 1: 0.03, classification 2: 0.05, classification 3: 0.81, classification 4: 0.1, classification 5: 0.01].

[0132] a3, selecting an emotion label corresponding to the emotion classification that meets the set confidence requirement from the emotion classification results as the emotion label corresponding to the comment text.

[0133] In this embodiment, the step of determining the emotion label corresponding to the comment text from the emotion classification result can be performed by the output layer of the emotion classification model, or can also be performed by the application program where the selected multimedia is located.

[0134] Regarding setting the confidence requirement, it can be flexibly set according to the needs. For example, the confidence requirement can be set to the highest confidence, that is, after obtaining the emotion classification result of the comment text, the corresponding emotion label can be selected according to the highest confidence emotion classification of each comment to ensure that each comment text is assigned a most likely emotion label, so as to more accurately reflect the user's emotional attitude towards multimedia content. Alternatively, based on the existence of a comment text expressing two or more emotions at the same time, in order to more reasonably count the emotion labels, the confidence requirement can also be set to a confidence greater than a set threshold such as 40%, and the set threshold is less than 50%, so as to be able to capture multiple emotional tendencies in a comment text at the same time and avoid missing some emotion labels.

[0135] Taking the emotion label corresponding to the emotion classification with the highest confidence in the emotion classification results as the emotion label corresponding to the comment text as an example, for the emotion classification results of the above comment 1 (classification 1: 0.03, class 2: 0.05, class 3: 0.81, class 4: 0.1, class 5: 0.01), the emotion label corresponding to the highest confidence class 3 (feeling) is used as the emotion label of the comment text.

[0136] In the disclosed embodiment, by utilizing a pre-built sentiment classification model to perform efficient semantic analysis on all comment texts of the selected multimedia, sentiment tags closely related to the multimedia content are obtained, thereby achieving fast and accurate extraction of sentiment information from a large amount of comment text, greatly improving the efficiency and accuracy of sentiment analysis.

[0137] Based on the emotion labels corresponding to each comment text of the selected multimedia obtained through the above-mentioned emotion classification model, in order to further quantify the user's emotional inclination and attention to multimedia content, the frequency of occurrence of the same emotion label in the comment text can also be counted, and the label heat can be calculated based on this, that is, for the emotion label corresponding to each comment text of the selected multimedia, the number of comment texts corresponding to the same emotion label is determined, and the label heat of the emotion label is determined based on the number of comment texts.

[0138] Regarding determining the label heat of the emotion label based on the number of comment texts, it can be determined based on the set exchange ratio between the number of comment texts and the label heat. For example, if the label heat value corresponding to 1 comment text is set to be plus 1, the number of comment texts corresponding to the same emotion label can be directly used as the label heat of the emotion label to truly reflect the number of users who have selected the emotion label, thereby accurately quantifying the user's emotional inclination and attention to multimedia content.

[0139] For example, for 1,000 comments on song A, after using the emotion classification model to obtain the emotion label corresponding to each comment, we can group them by emotion label, merge the same emotion labels, and count the number of comments corresponding to the same emotion label as the label heat of the emotion label. Assume that the emotion labels obtained after emotion classification of 1,000 comments are classified as follows (some are listed as examples): excited: 230; moved: 220; liked: 350; calm: 100; bored: 50; angry: 50, then the heat value of the corresponding emotion label is the number of corresponding comment texts.

[0140] In some embodiments, considering that different types of multimedia with different themes or styles usually have different emotional attributes in actual application scenarios, it reflects the unique emotional attitudes held by users towards various types of content. For example, taking articles as an example, they can be divided into multiple multimedia content categories such as politics, education, entertainment, food, economy, beauty, music, and home life according to the theme. For articles under each theme, the emotional attitudes expressed by users are different: for example, the emotional attitudes of users' evaluations under beauty articles tend to be asking for tutorials (seeking makeup, skin care, etc. skills), asking for links (asking about product purchase channels), expressing amazement (surprised at the effect), expressing praise (liking makeup, products), etc., while the emotional attitudes under music tend to be liking, willing to recommend and share, agreeing with opinions, etc.

[0141] In order to more accurately capture the differences in users' emotional attitudes under different themes or styles of the same multimedia type, and to capture the differences in emotional attitudes between different multimedia types, this embodiment pre-sets an emotion classification set that matches different multimedia content categories of the multimedia in the emotion classification model for different types of multimedia; the multimedia content category refers to the classification based on the content theme or style of the multimedia.

[0142] That is, for each type of multimedia, multiple multimedia content categories of the multimedia are obtained by dividing the multimedia according to the content theme or style of the multimedia. According to the characteristics of each multimedia content category, an emotion classification set is separately set to represent the user's emotional inclination towards the content of the multimedia content category, including multiple emotion classifications and each emotion classification corresponds to an emotion label applicable to the multimedia content category.

[0143] For example, when the multimedia is in the form of songs, a corresponding set of emotion classifications is set for each style of song according to the style information of the song. For example, if the multimedia content categories obtained by dividing the style information of the song include electronic music, Chinese style, light music, folk songs, children's songs, classical music, jazz, rock, etc., then an emotion classification set is set for each music style. For example, the emotion classification set corresponding to the jazz category may include relevant emotion categories such as appreciation, enjoyment, nostalgia, and skill discussion, and the emotion classification set corresponding to the Chinese style category may include relevant emotion categories such as identification, immersion, pride, and love for traditional culture.

[0144] Based on the emotion classification set of each multimedia in different multimedia content classification categories pre-stored in the emotion classification model, when using the emotion classification model to generate the emotion classification result of the input comment text, the following steps may also be included:

[0145] The multimedia content category of the selected multimedia is input into the emotion classification model, so that the emotion classification model obtains an emotion classification set matching the multimedia content category, and each emotion classification included in the emotion classification set is used as the emotion classification preset in the emotion classification model.

[0146] That is, when performing semantic analysis on all comment texts of the selected multimedia, the multimedia content category of the selected multimedia is first input into the emotion classification model. For example, if the selected multimedia is song A, the multimedia content category of song A, such as jazz, is input into the emotion classification model.

[0147] After receiving the multimedia content category, the emotion classification model first obtains the emotion classification set corresponding to the multimedia content category (such as the emotion classification set corresponding to the jazz category) as the emotion classification preset in the model. Next, for each comment text of the selected multimedia received, the confidence that the comment text belongs to each preset emotion classification is calculated through semantic analysis. For example, the emotion classification set corresponding to the jazz category corresponds to categories 15-25, and the emotion classification result of each comment text of the selected multimedia of the jazz category includes the confidence that the comment text belongs to emotion categories 15-25.

[0148] In the embodiments of the present disclosure, by separately setting an emotion classification set for each multimedia content category based on the actual user emotions of that type of content, and when the emotion classification model is running, the matching emotion classification set is automatically selected for emotion analysis according to the input multimedia content category, so that the model can more flexibly adapt to the emotion analysis needs in different scenarios, and can more finely capture the differences in users' emotional attitudes under different themes or styles, thereby more accurately reflecting the true emotional tendencies in users' comment texts.

[0149] In some embodiments, for the interactive operation on the tag display area described in the above embodiments, the interactive operation can be used to realize direct interaction between the user and the emotion tag displayed in the tag display area, so as to quickly express the user's attitude towards the selected multimedia content. That is, the interactive operation can include a first selection operation acting on any emotion tag displayed in the tag display area. Based on the first selection operation, the above step S202 described in response to the interactive operation on the tag display area to update the emotion tag and / or tag heat displayed in the tag display area can be realized in the following way:

[0150] In response to a first selection operation on any emotion tag displayed in the tag display area, the tag heat value of the first emotion tag selected by the first selection operation is set to increase by a first preset value, and the tag display area is updated to display the increased tag heat of the first emotion tag.

[0151] As an interactive element in the tag display area, the emotion tag supports interaction with the user. The user can implement the first selection operation for the emotion tag through a variety of input methods to adapt to the operating environment of different user terminals. The first selection operation may include but is not limited to: the user touches the first emotion tag through a touch screen, or the user selects or clicks the first emotion tag through an external device such as a mouse or keyboard.

[0152] The first preset value is used to increase the heat value of a certain emotion tag each time a user selects the emotion tag. The size of the first preset value can be adjusted according to actual needs to control the sensitivity of the tag heat change. In order to truly reflect the number of users who select the emotion tag and accurately quantify the user's emotional tendency and attention to multimedia content, the first preset value can be set to 1, so that when the user selects the first emotion tag through the first selection operation, the tag heat value of the first emotion tag is set to be increased by one. Or, the first preset value can be determined according to the duration of the operation of the first selection operation, and the exchange ratio of the operation duration and the increase in heat value is set. For example, if the operation duration of the first selection operation increases by one second, the heat value increases by 1. Then, the longer the user's selection operation for the first emotion tag lasts, the more the tag heat value of the emotion tag will increase accordingly, that is, the larger the value of the first preset value is, the more detailed the user's attention and investment in a specific emotion tag will be.

[0153] To better understand the interactive operations and update results described in this embodiment, see Figure 3An exemplary diagram of a comparison of emotion tag display before and after in response to the above-mentioned first selection operation is shown. The left figure is a comment page that displays the emotion tags of the selected video in the tag display area. When the user selects any emotion tag (such as the emotion tag marked with a gray box in the figure) through touch or an external input device, in response to the first selection operation on the emotion tag, the tag heat value of the emotion tag is increased by one, and the tag display area is updated, so as to obtain the tag heat of the emotion tag after the value is increased by one as shown in the right figure.

[0154] In the disclosed embodiment, by optimizing the interactive operation of the tag display area, users can directly, quickly and diversely express their emotional attitudes towards multimedia content through direct interactive operations with emotional tags. It introduces the selection and popularity update of emotional tags, allowing users to directly select emotional tags displayed on the interface when watching or reading multimedia content. In response to the user selecting an emotional tag, the popularity value of the tag is increased, and the interface is updated in real time to intuitively reflect the audience feedback on the current content at the emotional level, enriching the user's way of expressing emotions and enhancing the emotional interaction and resonance between users and content. Compared with manually writing comment texts, the selection operation of emotional tags is more convenient and fast, greatly reducing the time and energy cost of user input.

[0155] At the same time, since the popularity of tags can be updated and displayed in real time, users can immediately see that their feedback is being paid attention to and responded to, thereby improving interaction efficiency and user satisfaction. In addition, it provides a more friendly expression platform for users who are not good at or unwilling to write comments. Users can effectively convey their emotional attitudes with a simple click, reducing the threshold and complexity of emotional expression.

[0156] In some embodiments, in order to enhance the fun and intuitiveness of the emotional interaction between the user and the multimedia content, and to improve the immersion and satisfaction of the user experience, this embodiment proposes an animation effect display mechanism, which responds to the user's selection operation of the emotion tag instantly through visual dynamic feedback, so that the user can get a more vivid and interesting interactive experience while expressing emotions. Based on this, for the first selection operation of the emotion tag described in the above embodiment, the method may also include a selection animation effect display step, that is, in response to the first selection operation of the emotion tag, the selection animation effect related to the first emotion tag is displayed on the comment page; wherein, the selection animation effect includes an animation effect for an emoticon; and the emoticon matches the emotion represented by the first emotion tag. For example, if the user selects the emotion tag representing "happy" through the first selection operation, the first selection operation may trigger the display of a dynamic animation effect about a smiley face on the comment page.

[0157] The step of displaying the selection animation effect and the step of setting the label heat value of the first emotion label to increase the first preset value and updating the label display area in the above embodiment can be performed in parallel, or the step of displaying the selection animation effect can be performed first. This step gives immediate visual feedback to the user's selection behavior in the form of animation, which enhances the interactivity and fun of the user's operation. Among them, the selection animation effect is designed to include animation effects for emoticons, which not only closely match the emotions represented by the first emotion label, but also can convey the emotions contained in the emotion label more vividly through dynamic display methods, such as zooming in, zooming out, rotating or color changing, so that when the user selects the emotion label, he can intuitively feel the emotional color of the selected emotion, thereby deepening the understanding of the content and emotional resonance.

[0158] The selection animation effect includes an animation effect for an emoticon, and the animation effect for the emoticon may include at least any of the following implementation methods: an expression explosion animation effect for the emoticon, or a full-screen expression instantaneous effect for the emoticon. The expression explosion animation effect is used to generate multiple identical emoticons on the terminal screen starting from the selection position of the first selection operation, and fade out of the terminal screen after moving along a set path. The set path may be a random path or a specified path with a pre-set direction by the user. The full-screen expression instantaneous effect for the emoticon is used to expand the emoticon to the entire terminal screen at a set speed, and fade out of the terminal screen after full-screen display for a set time.

[0159] Take the first selection operation to select the emotion of happiness and match the smiley face emoticon as an example, see Figure 4A A static diagram showing an example of an expression explosion animation effect, where multiple smiling faces gradually disappear after moving in a random path on the terminal screen. Figure 4B The schematic diagram of the full-screen expression instantaneous effect is exemplarily shown, where the smiley face expression slowly expands to the entire screen and gradually disappears from the entire screen after being displayed for a few seconds.

[0160] It is understandable that other animation forms can also be used for the animation effects of emoticons, such as the zooming and bouncing effect of emoticons, that is, after the emoticon is selected, it will perform a series of zooming and bouncing actions to simulate cheerful or excited emotions; or the gradual color change effect of emoticons, which uses the gradual change of colors to show the transition and change of emotions and increase the delicate expression of emotions; and the spinning and dancing effect of emoticons, which allows the emoticons to spin and slowly dance on the screen. This animation form can be selected and customized according to specific emotional labels and user needs to achieve a better interactive experience.

[0161] In some embodiments, based on considerations of enhancing the diversity of user emotional expression and personalized needs, and in order to enable users to more accurately convey their complex emotions, this embodiment proposes a function for adding emotion tags. In addition to the displayed emotion tags, users can also add emotion tags that match their attitudes from the emotion tag library provided by the program according to their needs, so as to enrich the user's emotional expression methods and make the communication content more vivid, interesting and layered.

[0162] Based on the function of adding emotion tags, for the interactive operation on the tag display area described in the above embodiment, the interactive operation can be used to implement adding emotion tags that are not displayed in the tag display area from the emotion tag library. That is, the interactive operation can include an emotion tag adding operation acting on the tag display area; specifically, in the case where a tag adding control is set in the tag display area, the emotion tag adding operation can include a user's touch operation on the tag adding control, or a user's selection operation on the tag adding control through an external input device; in the case where a tag adding control is not set, the emotion tag adding operation can include a preset touch screen interactive gesture operation (such as long press, sliding a specific path, etc.), a physical button operation (such as a specific combination key), or a specified input operation of an external input device (such as a mouse), etc., so as to provide users with a variety of intuitive and easy-to-use ways to trigger the emotion tag adding process to meet the usage habits and preferences of different users.

[0163] Based on the emotion tag adding operation, the updating of the emotion tags and / or tag heat displayed in the tag display area in response to the interactive operation on the tag display area in the aforementioned step S202 can be achieved in the following manner:

[0164] In response to the emotion tag adding operation, a preset emotion tag library is displayed; the emotion tag library includes a plurality of predefined candidate emotion tags;

[0165] In response to a confirmation adding operation on at least one candidate emotion tag, the tag area is updated to display the at least one newly added candidate emotion tag and its tag heat; the tag heat value of the at least one candidate emotion tag is an initialization value.

[0166] Among them, the emotion tag adding operation triggers the display of the emotion tag library so that an emotion tag can be selected and added therefrom. The emotion tag library refers to a collection of multiple predefined candidate emotion tags. The user can currently view each emotion tag in the tag library and choose to add it to the tag display area. The confirmation adding operation includes a confirmation operation performed after the user selects at least one candidate emotion tag. The confirmation adding operation is used to trigger the addition of the selected candidate emotion tag to the tag display area and trigger the update of the tag popularity. The confirmation adding operation may include but is not limited to the selection operation of the confirmation control, or the closing operation of the display page of the emotion tag library, the direct dragging operation of the candidate emotion tag, etc., which can be dynamically adjusted according to the setting method of the interactive interface.

[0167] A label display area is set on the comment page of the selected multimedia. One or more areas or elements on this area (such as buttons, blank areas, etc.) can be set as hotspots that trigger the operation of adding emotion labels. When the user performs a preset trigger operation (such as click, long press, etc.), it is detected that the operation is an emotion label addition operation, and the pre-set emotion label library is loaded from the backend or local storage and displayed to the user in the form of pop-up windows, drop-down lists, etc.

[0168] The user can select one or more candidate emotion tags from the displayed emotion tag library by clicking, checking, etc., and when the user's confirmation adding operation is recognized, the candidate emotion tags selected by the user are recorded and the tag display area is updated, and the candidate emotion tags selected by the user are added to the area.

[0169] Among them, for the emotion tags that are not displayed in the tag display area, an initialization value is assigned to the newly added emotion tag as the tag heat, and the newly added emotion tag and the tag heat information are stored in the data storage space for subsequent updates. The initialization value can be 1 to accurately reflect the number of users who select the emotion tag, or the initialization value can also be determined according to the growth rule of the tag heat. For the emotion tags selected by the user and already displayed in the tag display area, the tag heat value of the emotion tag selected by the user can be set to increase by one or other preset values, and the tag display area can be updated to display the increased tag heat.

[0170] See also Figure 5The schematic diagram of the update process for the operation of adding emotion tags is shown as an example. The left figure shows a comment page including a tag display area, and the tag display area is provided with a tag adding control. When the user selects the tag adding control through touch or an external input device, the tag adding operation is detected, triggering the display of the emotion tag library such as the multiple candidate emotion tags shown in the middle figure. The user's confirmation addition operation on at least one candidate emotion tag is detected, such as the addition operation of the candidate emotion tags "Studying" and "Good morning" in the figure, which triggers the update of the tag display area to display the newly added emotion tags on the basis of the originally displayed emotion tags. As shown in the tag display area in the right figure, compared with the left figure, the emotion tags "Studying" and "Good morning" are newly added, and the tag heat of the two newly added emotion tags is set to the initialization value 1.

[0171] In the embodiments of the present disclosure, by introducing a predefined plurality of candidate emotion tag libraries and providing a tag adding operation based on the emotion tag library, users are allowed to directly select or add emotion tags to the display area through simple interactive operations, and real-time updating of tags displayed in the tag display area and their popularity is achieved, thereby providing a quick interaction method, lowering the threshold and complexity for users to express complex emotional attitudes, making more users willing to participate in the interaction, and thus improving the overall efficiency and satisfaction of user interaction.

[0172] In some embodiments, based on the users' growing demand for personalized emotional expression and the fact that existing predefined candidate emotion tags may not fully cover all users' emotional attitudes towards selected multimedia content, in order to further improve the user experience and enhance the flexibility and richness of emotional expression, this embodiment provides an emotion tag customization function that allows users to freely create and add personalized emotion tags according to their needs.

[0173] Based on the emotion tag customization function, for the interactive operation on the tag display area described in the above embodiment, the interactive operation can be used to implement adding user-defined emotion tags in the tag display area, that is, the interactive operation can include the emotion tag customization operation acting on the tag display area. Specifically, in the case where a tag customization control is set in the tag display area, the emotion tag adding operation can include the user's touch operation on the tag customization control, or the user's selection operation on the tag customization control through an external input device; in the case where a tag customization control is not set, the emotion tag adding operation can also include preset touch screen interactive gesture operations, physical button operations, or specified input operations of external input devices, etc., to provide users with a variety of intuitive and easy-to-use ways to trigger the custom process of emotion tags.

[0174] Based on the emotion tag customization operation, the above-mentioned step S202 of updating the emotion tags and / or tag heat displayed in the tag display area in response to the interactive operation on the tag display area can be implemented by the following steps b1-b3:

[0175] b1, in response to the emotion tag customization operation acting on the tag display area, displaying a customization editing interface;

[0176] The custom editing interface is used to receive the custom emotion text input by the user, which may include at least a text input box and operation buttons (such as confirm, cancel, etc.). The text input box is a key area for the user to input the custom emotion label name or description, and can support multiple input methods, such as keyboard input, handwriting input or voice input, to meet the input habits and needs of different users.

[0177] After the user completes the emotional text input, they can click the Confirm button to submit the custom label to process the user input, generate the corresponding custom emotional label, and update the label display area. If the user decides not to create a custom label midway, they can click the Cancel button to exit the custom editing interface without saving any changes.

[0178] b2, receiving the emotion text input by the user through the custom editing interface;

[0179] b3, generating a corresponding custom emotion tag according to the emotion text, and updating the tag display area to display the newly added custom emotion tag and its tag heat; the tag heat value of the custom emotion tag is taken as the initialization value.

[0180] The emotion tag includes text content representing the emotion. After receiving the emotion text input by the user through the custom editing interface, the input emotion text can be directly used as the text content of the emotion represented by the custom emotion tag, and the tag heat value of the custom emotion tag is set to the aforementioned initialization value such as 1 or other values. Next, the tag display area is updated in real time to display the newly added custom emotion tag and its tag heat to the user.

[0181] For example, see Figure 6The schematic diagram of the update process for the custom operation of emotion tags is shown as an example. The left figure shows a comment page including a tag display area, which is provided with a tag custom control. When the user selects the tag custom control through touch or an external input device, the emotion tag custom operation is detected, triggering a custom editing interface as shown in the middle figure to be displayed in the form of a pop-up window, drop-down list, etc.; the user can enter emotion text such as "strong sense of rhythm" in the text input box of the custom editing interface. After clicking the confirmation button, a custom emotion tag is generated according to the received emotion text, and the tag heat of the custom emotion tag is set to the initialization value 1, and the tag display area is updated to obtain the tag display area as shown in the right figure. Compared with the left figure, a new custom emotion tag "strong sense of rhythm" is added and its tag heat is displayed.

[0182] In the disclosed embodiments, by introducing the emotion tag customization operation, users can freely create and share emotion tags for selected multimedia content according to their own needs and preferences, thereby expanding the scope and depth of user emotional expression, providing users with a more personalized and liberalized interaction method, and enhancing the fun of interacting with multimedia content.

[0183] In some embodiments, based on the need to improve user experience and increase the diversity of emotional expression, the custom editing interface can not only provide a text input box for users to input emotional text, but also provide a variety of emoticons for expressing different emotions, which are implemented using various existing emoticons or emoticons that represent different user emotions. Based on the emoticons representing different emotions provided by the custom editing interface; when inputting emotional text through the custom editing interface, a corresponding custom emotional label can also be generated according to the emotional text and the emoticon selected by the second selection operation in response to a second selection operation for any emoticon.

[0184] That is, after the user enters the custom editing interface, the displayed custom editing interface includes not only a text input box, but also an emoticon panel or library, which contains various preset emoticons, which are organized into a layout that is easy to browse and select, such as classified by emotion type or sorted by frequency of use; the user can select the desired emoticon by clicking, dragging or other interactive methods. After the custom input is completed, the corresponding custom emotion label is generated according to the emotion text entered by the user and the emoticon selected by the second selection operation, that is, the emoticon can be inserted into the emotion text as part of the text, or the emoticon and text can be processed separately but displayed in association, depending on the system design and user preference; the label display area is updated to display the custom emotion label that combines text and emoticons.

[0185] In the disclosed embodiment, by providing the function of selecting emoticons, the customized emotion tags are made more vivid and interesting, thereby enhancing the attractiveness and depth of the interaction.

[0186] In some embodiments, in order to enhance the user interaction experience and provide clear feedback on the user's selection of emotion tags, for the three types of emotion tags, namely, the first emotion tag selected by the first selection operation described in the aforementioned embodiments, or at least one candidate emotion tag newly displayed in the emotion tag adding operation, or the custom emotion tag created by the emotion tag customization operation, when updating the tag display area, the emotion tags displayed in the tag display area and that meet any of the three types of emotion tags can also be visually updated to show that they are in a selected state through visual updates.

[0187] When updating the visual display to indicate that the emotion tag is in the selected state, a variety of visual display methods can be used to enhance the intuitiveness and interactivity of the user interface. The visual display method can include but is not limited to any of the following methods: changing the color of the selected emotion tag to make it contrast with other unselected tags, for example, changing the color of the selected tag to a bright color, a contrasting color, or a theme color to attract the user's attention; or, adding a highlight border to the selected emotion tag, and the color and width of the border can be adjusted according to design requirements; or, changing the background color of the selected emotion tag or adding a background pattern to make it more prominent, and the background change can be consistent with the tag color. Coordinate to form a harmonious visual effect; or, adjust the font style, size or color of the selected emotion label to make it more eye-catching, for example, make the font bold, larger or change it to a specific color to emphasize the user's choice; or, add a specific icon or mark next to the selected emotion label, such as a check box, a check mark, a star, etc., to clearly indicate that the label has been selected; or, add animation effects to the selected emotion label, such as slight enlargement, reduction, rotation or flashing, etc., to increase the visual appeal; or, add a shadow or relief effect to the selected emotion label to make it look more three-dimensional and prominent, enhance the layering of the label, and improve the aesthetics of the overall interface.

[0188] For example, Figure 7 As shown, the left picture shows a comment page containing a label display area, which displays multiple emotion labels. When the user selects the emotion label "Favorite" (i.e., the first emotion label) in the picture through touch operation, the label heat of the emotion label is updated while its visual effect is updated to indicate that the emotion label has been selected by the user, such as the first emotion label shown in the right picture. By adjusting the background color of the selected emotion label to gray, it is distinguished from the unselected emotion labels.

[0189] In the disclosed embodiment, by visually updating and showing that it is in a selected state, the user's eyes can be guided to focus on the selected emotion tag, and the user is given instant feedback on the emotion tag selection, which reduces the user's repeated operations or checks due to uncertainty about whether the selection is successful, thereby improving the smoothness of the overall interaction process and improving the accuracy and efficiency of the operation.

[0190] In some embodiments, the comment text based on the selected multimedia is dynamically updated, and users who view the selected multimedia can continuously post new comments, which will carry new emotional expressions. In order to maintain the real-time and accuracy of the emotional tags in the comment page, thereby providing users with richer and more accurate emotional feedback, the aforementioned comment method also includes an operation of updating the label display area based on the new comments, that is, in response to any user's new comment posting operation on the selected multimedia, obtaining a new comment text corresponding to the new comment posting operation, using a pre-built emotion classification model to perform semantic analysis on the new comment text, and obtaining a second emotion label corresponding to the new comment text, and then updating the label display area to display the second emotion label and its label popularity.

[0191] That is, if Figure 8 An exemplary interactive diagram of the processing process is shown, in which a user posts a new comment through the user interaction interface of the application where the selected media is located. The backend of the application will accordingly detect the new comment posting operation for the selected multimedia, and first obtain the new comment text entered by the user through the user interaction interface, wherein the user interaction interface can be any interaction interface that includes a comment posting entry, such as the aforementioned comment page.

[0192] Next, for the newly acquired comment text, call the trained sentiment classification model to perform semantic analysis on the new comment text, generate a sentiment classification result for the new comment text, including the confidence corresponding to each sentiment classification preset in the model, and based on the sentiment classification result, select the sentiment label corresponding to the sentiment classification whose confidence meets the preset confidence requirement as the second sentiment label corresponding to the new comment text. The specific implementation process can refer to the aforementioned implementation method of determining the emotion label of the comment text using the sentiment classification model, which will not be repeated here.

[0193] Finally, the tag display area on the comment page of the selected multimedia is updated to display the second emotion tag and its tag popularity. Based on the fact that the second emotion tag corresponding to the new comment text belongs to or does not belong to the emotion tags already displayed in the tag display area, the update of the tag display area to display the second emotion tag and its tag popularity is divided into adding the newly identified emotion tag to the tag display area or updating the tag popularity of the existing emotion tag. That is, combined with Figure 8As shown in the interaction diagram, the updating of the label display area to display the second emotion label and its label popularity can be achieved through the following implementation process:

[0194] In the case where the second emotion label is an emotion label that has been displayed in the label display area, setting the label heat value of the second emotion label to increase by a second preset value, and updating the label display area to display the increased label heat of the second emotion label;

[0195] When the second emotion tag is not displayed in the tag display area, the tag heat value of the second emotion tag is set to an initialization value, and the tag display area is updated to display the newly added second emotion tag and its tag heat.

[0196] Among them, the second preset value is used to increase the label heat value. In this embodiment, it means that when the emotion label corresponding to the new comment text is already displayed in the label display area, its heat needs to be increased. The size of the second preset value can be adjusted according to actual needs to control the sensitivity of the label heat change when adding new comments. In order to truly reflect the number of users who select the emotion label, the second preset value can be set to 1, so that when the user selects the first emotion label through the first selection operation, the label heat value of the first emotion label is set to be increased by one.

[0197] For example, a user can watch a short video on a social media platform and comment "It's so funny", and identify that the comment uses "funny" as an emotional tag. If the tag display area on the comment page of the short video has already displayed the "funny" tag, and has a heat value such as 50, then the heat value of the "funny" tag is set to increase by a second preset value, such as from a heat value of 50 to a heat value of 51, and the heat value of the "funny" tag in the tag display area is updated to be displayed as 51. If the "funny" tag is not displayed in the tag display area on the comment page of the short video, the tag heat value of the "funny" tag can be set to the aforementioned initialization value, such as 1. The initialization value is the same as the initialization value involved in the aforementioned embodiment, and has the same value. The tag display area is updated to display the newly added "funny" tag and its heat value 1 on the basis of the displayed emotional tags.

[0198] In the disclosed embodiment, the comment method can not only display the existing emotion tags and popularity of the selected multimedia, but also dynamically update its emotion tags and tag popularity as new comments are published by users, thereby maintaining the real-time and accuracy of the emotion tags and providing users with richer and more accurate emotion feedback.

[0199] In some embodiments, based on the consideration of comment interaction, in order to further improve the real-time nature of user participation and emotional feedback, in addition to updating the emotional tags and their popularity by posting new comments, the user's like behavior for existing comments can also be used as an important indicator to reflect the popularity of the comment content and emotions. The like operation not only expresses the recognition of the comment text, but also indicates the user's resonance with the emotional tags in the comment. In order to more comprehensively capture the user's emotional feedback on the selected multimedia and reflect it in real time on the emotional tags of the comment page, this embodiment proposes another way to maintain the real-time nature and accuracy of the emotional tags in the comment page, including:

[0200] In response to any user's like operation on any comment text of the selected multimedia, the liked comment text associated with the like operation is identified, and a third emotion tag obtained by semantic analysis of the liked comment text can be quickly determined based on the operation position of the like operation, and the tag heat value of the third emotion tag is set to increase by a second preset value to reflect the user's positive feedback on the third emotion tag, and the tag display area is updated to display the increased tag heat of the third emotion tag, ensuring that the user can instantly see the changes in emotional feedback, thereby providing a richer, more accurate and dynamic emotional experience.

[0201] In the disclosed embodiment, when a like operation occurs, the number of likes for the liked comment text will be increased by one accordingly, so as to accurately reflect the degree of recognition and support of the comment text by the user through the number of likes. At the same time, for the like operation, based on the fact that the like behavior represents the user's resonance and recognition of the emotional tendency expressed by the comment text, this embodiment also increases the label heat value of the third emotional label corresponding to the liked comment text by a second preset value, thereby comprehensively capturing and reflecting the user's emotional feedback on the selected multimedia in real time, and enriching the dimension and depth of emotional feedback without increasing the user's input burden.

[0202] In some embodiments, based on the social interaction characteristics of social media platforms, in order to enhance the emotional connection and sense of community belonging between users, when a user triggers the tag heat update of an existing emotion tag in the tag display area or adds a new emotion tag to the tag display area in the manner described in any of the above embodiments, it indicates that the user selects the existing emotion tag or the added new emotion tag, and the user account information of selecting each emotion tag can be synchronously recorded. The above comment method can also include:

[0203] In response to an interaction detail viewing operation for any emotion tag in the tag display area, other account information of a fourth emotion tag selected by the interaction detail viewing operation is displayed; the other account information supports viewing the account homepage indicated by the other account information.

[0204] The interactive detail viewing operation may include but is not limited to setting touch operations (double-clicking, long pressing) for any emotion tag in the tag display area, and specified input operations of external input devices (such as double-clicking the right button of a mouse, keyboard combination keys).

[0205] When displaying other account information that has historically selected the fourth emotion tag, if the current user also selects the fourth emotion tag, the current user's account information can be displayed in the front, and then other account information is displayed in sequence according to the time sequence of selecting the fourth emotion tag.

[0206] The other account information may be displayed in the form of a list, icon, etc., and the displayed information may include but is not limited to the account avatar, name, account ID, the number of times the selected multimedia is viewed or played, the operation time of selecting the fourth emotion tag, etc. The displayed other account information has the function of viewing the account homepage, that is, the other account information may contain a link or button, and the user can jump to the homepage interface of the account after clicking it.

[0207] Based on the other account information displayed above, in order to further promote emotional communication and interaction between users and enhance emotional resonance within the community, the method may also include the following emotional resonance interaction function, pre-setting a resonance control in the area where each other account information is located, and responding to a resonance interaction operation for any of the other account information, that is, a selection operation for the resonance control, sending a resonance prompt message to the other accounts indicated by the other account information; the resonance prompt message is used to prompt the other account that there is user emotional resonance with the emotional tag under the selected multimedia.

[0208] For example, Fig. 9 An example of an emotional resonance diagram is shown. The selected multimedia is song A. The current user clicks to view the details of the emotional label "Single Loop" as shown in the left figure, and displays the information of other accounts with the historical selection of "Single Loop" as shown in the middle figure. A resonance control is set on the display area where each other account information is located. When the user clicks the resonance control under any account information, it triggers the automatic sending of a resonance notification to the other accounts indicated by the other account information, such as the resonance notification shown in the right figure, "Someone resonates with you under XXX song! Click to view details", to prompt the other account. The resonance notification can be sent by the customer service assistant account of the application in the selected multimedia, or the current user's account, and can be set according to actual interaction needs.

[0209] In the disclosed embodiment, by introducing the emotional resonance interaction function and using the resonance control as a bridge for users to express emotional resonance, the user's emotional expression is closely connected with other members in the community, solving the problem of insufficient emotional communication between users in the traditional comment system. It also promotes deep interaction and emotional resonance between users through the instant resonance prompt information feedback mechanism, enriches the content ecology of the community, and enhances the interactivity and user stickiness of the community. At the same time, by allowing users to customize the sender of the resonance prompt information, the flexibility of the system and the user experience are further improved.

[0210] In some embodiments, in order to enhance the user's information acquisition efficiency and enable the user to more intuitively understand and participate in the interaction of selected multimedia content, when displaying the comment page of the selected multimedia, with respect to the display of emotion tags in the tag display area on the comment page, the emotion tags and their tag heat can be displayed in sequence in the tag display area in descending order of the tag heat corresponding to the emotion tags of the selected multimedia.

[0211] For example, if the emotional tags corresponding to the selected multimedia and their tag heats include (delicious: 100, nice to listen to: 520, liked: 888, exciting: 498, awesome: 380), they are arranged in descending order of tag heat and displayed in the tag display area in sequence (like: 888, nice to listen to: 520, exciting: 498, awesome: 380, delicious: 100).

[0212] In the disclosed embodiment, by displaying in descending order based on tag popularity, users can quickly focus on the most popular emotional tendencies for the selected multimedia, thereby more easily finding a group that resonates with their emotions or exploring new viewpoints and emotional perspectives.

[0213] In some embodiments, based on the tag display area being used to display a set number of emotion tags, in order to maintain the simplicity of the interface and the smoothness of the user experience, the first few emotion tags with the highest tag popularity are displayed by default. Based on the selected multimedia corresponding to including multiple emotion tags, when the user needs to have a more comprehensive understanding of all emotion tags, if the total number of emotion tags of the selected multimedia exceeds the set number, this embodiment provides an interactive method to implement the extended display of emotion tags, that is, if the total number of emotion tags of the selected multimedia exceeds the set number, in response to the emotion tag expansion operation, all emotion tags of the selected multimedia are loaded and displayed.

[0214] When the total number of emotion tags of the selected multimedia exceeds the set number, an interactive prompt control, such as a "view more" button or icon, can be provided near the tag display area. In response to a selection operation on the prompt control, all emotion tags of the selected multimedia are loaded and displayed, thereby ensuring the simplicity of the interface while taking into account the user's need to obtain complete information.

[0215] For example, a prompt icon representing extended display of emotion tags is set in the tag display area, and only two rows of emotion tags are displayed in the tag display area. When the user clicks the prompt icon, all emotion tags of the selected multimedia are loaded and displayed.

[0216] In some embodiments, the comment page also includes a text display area for displaying comment texts. When the comment page of the selected multimedia is displayed, the comment text of the selected multimedia can be displayed in the text display area. The emotion tags corresponding to the selected multimedia are obtained by analyzing the individual comment texts of the selected multimedia. Each comment text corresponds to an emotion tag. In order to further enhance the user's browsing experience and enable the user to more conveniently focus on comments on the emotional dimensions of interest, this embodiment innovatively proposes a comment text viewing method based on emotion tags, that is, in response to the comment filtering operation based on the emotion tags, all comment texts matching the fifth emotion tag selected by the comment filtering operation can be obtained according to the stored comment texts and their corresponding emotion tags, and displayed in the text display area.

[0217] Among them, a filter control can be set on the comment page, and the comment filtering operation based on the emotion tag can include a click operation on the filter control. In response to the click operation on the filter control, a filter condition setting page is displayed, and the filter condition setting page displays various emotion tags of the selected multimedia. When the user selects at least one emotion tag as a filter condition and performs a confirmation filter operation, all the comment texts that meet the filter condition are filtered out from the stored comment texts.

[0218] For example, a user is viewing a movie review page that displays various emotion tags. The user is interested in the public's comments on touching moments in the movie. The user can click the filter control set on the review page and select the emotion tag "touched" as the filter condition from the pop-up emotion tag list to perform the review filtering operation, triggering the movie playback program to quickly retrieve all the comment texts marked as "touching" based on the stored comment texts and their corresponding emotion tag information. These comment texts are then loaded into the text display area, and the user can directly read all the comments related to the "touching" emotion without having to flip through a large number of comments.

[0219] In the disclosed embodiment, by providing an emotion tag filtering function, users can quickly locate comments that resonate with their interests or emotions, avoiding the trouble of blindly searching through a large number of comments, and enabling users to more efficiently obtain information on specific emotion dimensions, thereby improving the user's browsing experience.

[0220] Next, in order to enable those skilled in the art to better understand the comment method provided by the present invention, the comment method is exemplarily described by taking a song comment page in a song playing scenario as an example.

[0221] Currently, in the song playback scenario, songs support users' red heart collection and the number of red hearts will be displayed on the song playback page. However, the red hearts of the song only represent the number of users who like the song, and cannot express the diverse attitudes of users towards the song. The song comment page supports users to post text comments to express their feelings about listening to the song, and displays the comment content and number of comments on the comment page. Users can browse the content of single text comments and pictures. It is difficult to quickly capture the overall emotional distribution of the public towards the current song in a large amount of comment text. Moreover, the current comment area can only support users to manually generate text and picture content. For users who are not good at or unwilling to manually edit comment text, there is a lack of fast and convenient way to express emotions, which affects the interactivity between users and songs.

[0222] Based on this, the present invention provides a comment method applicable to songs, which is applied to a song playing program. When a user clicks a control for viewing comments through any interface in the song playing program, the method may further include the following steps c1-c3:

[0223] c1. In response to a comment viewing operation on a selected song, a comment page of the selected song is displayed; wherein the comment page includes a label display area and a text display area. The label display area is used to display the emotion label corresponding to the selected song and its label popularity, while the text display area is used to display the comment text of the selected song. The comment text can be displayed in order according to the release time or the number of likes.

[0224] The emotion tag is obtained by semantically analyzing the historical comment text of the selected song. When a new comment posting operation is detected for the selected song, the new comment text corresponding to the new comment posting operation is semantically analyzed to obtain the corresponding emotion tag, and the emotion tag corresponding to the selected song is synchronously updated to ensure the real-time and accuracy of the emotion tag of the selected song.

[0225] The method of semantic analysis is to establish an emotion classification model based on the user's emotional feelings (such as happiness, wanting to cry, healing, etc.), music evaluation (such as the melody is addictive, the lyrics are touching, the rhythm is dynamic, etc.), and status (such as studying, exercising, etc.). By manually collecting content samples and training the algorithm model, the emotion label corresponding to each comment can be identified.

[0226] For the emotion label, the corresponding emoji expression is matched for each emotion label. This can be done by identifying a group of emojis related to emotions, expressions, and states, and defining the semantics corresponding to the emoji and matching the emotion label.

[0227] The tag display area on the song review page displays the emotion tags and tag popularity of the selected song, which can be displayed in descending order according to the tag popularity. The tag display area is set to display 2 rows of emotion tags by default. For other emotion tags exceeding 2 rows, the user can click the expansion control to trigger the display of all tags under this song.

[0228] c2, by setting an interactive operation, triggering the update of the emotion tags and their tag popularity displayed in the tag display area; the interactive operation includes at least a third selection operation for the displayed emotion tags, a fourth selection operation for adding controls to the emotion tags, and a fifth selection operation for customizing controls for the emotion tags.

[0229] The third selection operation for the displayed emotion tags is used to enable users to select and vote for the displayed emotion tags. That is, users can vote for the emotion tags that have been displayed under the song to express their same feelings and attitudes towards the song; after the user votes, the animation effect of voting for the emotion tag can be displayed to the user on the comment page to strengthen the user's positive feedback on the statement, and the tag heat value of the emotion tag will increase, such as the cumulative number +1, and the user's selection of each emotion tag for each song will be recorded at the same time, so that the user can display the historically selected emotion tags when entering the comment page of this song next time.

[0230] The fourth selection operation for the emotion tag adding control and the fifth selection operation for the emotion tag customization control are used to add new emotion tags to the tag display area. The song playback program provides a set of preset candidate emotion tags. Through the fourth selection operation, the user can select one or more candidate emotion tags from the candidate emotion tag panel and add them to the tag display area. In addition, the user can also customize the emotion tag and trigger the display of the custom editing interface through the fifth selection operation. By selecting emoji and entering text on the custom editing interface, the custom emotion tag is obtained and added to the tag display area.

[0231] In the disclosed embodiments, the emotions and attitudes of users under the song are intuitively displayed in the form of emotional tags on the comment page of the selected song, so that users who are listening to the song can feel a strong emotional resonance; at the same time, users can select or freely create emotional tags to express their emotional attitudes towards the selected song, thereby expressing listening emotions under the song at a lower cost, reducing the complexity of interactive operations, allowing more users to participate in the comment page interaction, improving the activity and scale of music users, and enhancing the music community attributes and brand tone.

[0232] Exemplary Devices

[0233] After introducing the method of the exemplary embodiment of the present invention, next, refer to Fig.10 A device according to an exemplary embodiment of the present invention is described.

[0234] The implementation process of the functions and effects of each module in the following device is specifically described in the implementation process of the corresponding steps in the above method, which will not be repeated here. For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can be referred to the partial description of the method embodiment.

[0235] Fig.10 A comment device according to an embodiment of the present invention is schematically shown, the device comprising:

[0236] The emotion tag viewing module 1001 is used to display the comment page of the selected multimedia; the comment page includes a tag display area for displaying emotion tags, and the tag display area displays emotion tags and the tag heat of each emotion tag; the emotion tags are obtained by semantically analyzing the comment text of the selected multimedia;

[0237] The emotion tag interaction module 1002 is used to update the emotion tags and / or tag heat displayed in the tag display area in response to the interaction operation on the tag display area.

[0238] Exemplary Media

[0239] After introducing the method and apparatus of the exemplary embodiment of the present invention, next, reference is made to Fig.11 A readable storage medium according to an exemplary embodiment of the present invention is described.

[0240] In this exemplary embodiment, the above method can be implemented by a program product, such as a portable compact disk read-only memory (CD-ROM) and including program code, and can be run on a device, such as a personal computer. However, the program product of the present invention is not limited to this. In this document, a readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, an apparatus or a device.

[0241] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0242] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0243] The program code contained on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RE, etc., or any suitable combination of the foregoing.

[0244] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user computing device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0245] Exemplary Computing Devices

[0246] After introducing the method, apparatus and medium of the exemplary embodiments of the present invention, next, reference is made to Fig.12 A computing device according to an exemplary embodiment of the present invention is described.

[0247] Fig.12 The computing device 1200 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0248] like Fig.12 As shown, the computing device 1200 is in the form of a general computing device. The components of the computing device 1200 may include but are not limited to: at least one processing unit 1201, at least one storage unit 1202, and a bus 1203 connecting different system components (including the processing unit 1201 and the storage unit 1202).

[0249] The bus 1203 includes a data bus, a control bus, and an address bus.

[0250] The storage unit 1202 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 12021 and / or a cache memory 12022 , and may further include a readable medium in the form of a non-volatile memory, such as a read-only memory (ROM) 12023 .

[0251] The storage unit 1202 may also include a program / utility 12025 having a set (at least one) of program modules 12024, such program modules 12024 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0252] The computing device 1200 may also communicate with one or more external devices 1204 (eg, a keyboard, pointing device, etc.).

[0253] Such communication may be performed via input / output (I / O) interface 1205. In addition, computing device 1200 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via network adapter 1206. Fig.12 As shown, the network adapter 1206 communicates with other modules of the computing device 1200 via the bus 1203. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the computing device 1200, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0254] Although the spirit and principle of the present invention have been described with reference to several specific embodiments, it should be understood that the present invention is not limited to the disclosed specific embodiments, and the division of various aspects does not mean that the features in these aspects cannot be combined to benefit, and such division is only for the convenience of expression. The present invention is intended to cover various modifications and equivalent arrangements included in the spirit and scope of the attached claims.

Claims

1. A comment method, characterized in that: The method comprises: Displaying a comment page for the selected multimedia; the comment page includes a tag display area for displaying emotion tags, the tag display area displays emotion tags and tag heat of each emotion tag; the emotion tags are obtained by semantically analyzing the comment text of the selected multimedia; In response to an interactive operation on the tag display area, the emotion tag and / or tag heat displayed in the tag display area is updated.

2. The method according to claim 1, characterized in that The method also includes a step of pre-acquiring emotion tags: Obtaining all comment texts of the selected multimedia; Using a pre-built emotion classification model to perform semantic analysis on the comment text, obtaining an emotion classification result corresponding to each comment text; the emotion classification result includes a confidence level corresponding to each emotion classification preset in the emotion classification model; The emotion label corresponding to the emotion classification with the highest confidence in the emotion classification results is selected as the emotion label corresponding to the comment text.

3. The method according to claim 2, characterized in that The method further comprises: For the emotion tag corresponding to each comment text of the selected multimedia, the number of comment texts corresponding to the same emotion tag is determined as the tag heat of the emotion tag.

4. The method according to claim 2, characterized in that: For different types of multimedia, the emotion classification model pre-sets emotion classification sets that match different multimedia content classification categories of the multimedia; The multimedia content category refers to the classification according to the content theme or style of the multimedia; The method further comprises: inputting the multimedia content category of the selected multimedia into the emotion classification model; The emotion classification model obtains an emotion classification set that matches the multimedia content category, and uses each emotion classification included in the emotion classification set as an emotion classification preset in the emotion classification model.

5. The method according to claim 1, characterized in that In response to the interactive operation on the tag display area, updating the emotion tag and / or tag heat displayed in the tag display area includes: In response to a first selection operation on any emotion tag displayed in the tag display area, the tag heat value of the first emotion tag selected by the first selection operation is set to increase by a first preset value, and the tag display area is updated to display the increased tag heat of the first emotion tag.

6. The method according to claim 1, characterized in that In response to the interactive operation on the tag display area, updating the emotion tag and / or tag heat displayed in the tag display area includes: In response to an emotion tag adding operation acting on the tag display area, a preset emotion tag library is displayed; the emotion tag library includes a plurality of predefined candidate emotion tags; In response to a confirmation adding operation on at least one candidate emotion tag, the tag area is updated to display the at least one newly added candidate emotion tag and its tag heat; the tag heat value of the at least one candidate emotion tag is an initialization value.

7. The method according to claim 1, characterized in that In response to the interactive operation on the tag display area, updating the emotion tag and / or tag heat displayed in the tag display area includes: In response to an emotion tag customization operation performed on the tag display area, displaying a customization editing interface; Receiving emotional text input by the user through the custom editing interface; Generate a corresponding custom emotion tag according to the emotion text, and update the tag display area to display the newly added custom emotion tag and its tag heat; the tag heat value of the custom emotion tag is taken as an initialization value.

8. A comment device, characterized in that: The device comprises: An emotion tag viewing module is used to display a comment page of the selected multimedia; the comment page includes a tag display area for displaying emotion tags, the tag display area displays emotion tags and the tag heat of each emotion tag; the emotion tags are obtained by semantically analyzing the comment text of the selected multimedia; The emotion tag interaction module is used to update the emotion tags and / or tag heat displayed in the tag display area in response to the interaction operation on the tag display area.

9. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. A computing device, characterized in that include: Processor, memory; The memory is used to store computer programs; The processor is configured to execute the method according to any one of claims 1 to 7 by calling the computer program.

Images