Comment information processing method and device, electronic equipment and storage medium

Abstract

The application discloses a comment information processing method and device, electronic equipment and storage medium. When a target object comments on a target product, a comment initiation operation is triggered, and a terminal device displays a comment editing interface. The comment editing interface comprises a main input area and at least one auxiliary input area, wherein the auxiliary input area displays title information. After the target object completes editing operations in the main input area and the auxiliary input area, a comment ending operation is triggered, all contents in the main input area and the edited auxiliary input area are integrated, target comment information is obtained and displayed, and the unedited auxiliary input area is discarded. The application can provide a flexible and convenient comment editing method for the target object, can reduce the editing difficulty of the comment information, and is beneficial to improving the reference value of the comment information. The technical scheme of the application can be widely applied to the technical field of information processing.

Description

Technical Field

[0001] This application relates to the field of information processing technology, and in particular to a method, apparatus, electronic device, and storage medium for processing comment information. Background Technology

[0002] Currently, with the popularization and rapid development of the internet, online applications have gradually integrated into people's lives, providing a wide variety of services. For example, video playback applications can play various movies and TV shows, bringing viewers a rich audio-visual entertainment experience; online shopping applications can support online shopping, improving the convenience and efficiency of shopping.

[0003] In related technologies, applications often include review functions to help more people understand product features. When browsing products, users can refer to reviews from other users to decide whether to use or purchase the product. However, current in-app review functions generally rely entirely on reviewers manually inputting information, supporting only basic text editing software such as copy and paste and keyword suggestions. This makes editing reviews difficult and time-consuming, impacting efficiency and user interest. Furthermore, some reviewers lack sufficient product knowledge and summarizing skills, making it difficult to provide valuable reviews, thus reducing the objectivity and usability of the reviews.

[0004] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0005] This application provides a method, apparatus, electronic device, and storage medium for processing comment information, which can effectively reduce the difficulty of editing comment information and improve the reference value of comment information.

[0006] One aspect of this application provides a method for processing comment information, including the following steps:

[0007] In response to a target user's comment on a target product, a comment editing interface is displayed; the comment editing interface includes a main input area and at least one sub-input area, and each sub-input area displays title information.

[0008] When the input focus is on the main input area, in response to the first input operation of the target object, first content information is displayed in the main input area;

[0009] When the input focus is on the secondary input area, in response to a second input operation of the target object, second content information is displayed in the secondary input area, wherein the second content information includes the title information and the information input by the second input operation;

[0010] In response to the comment ending operation of the target object, target comment information is displayed, wherein the target comment information includes the first content information and the second content information.

[0011] On the other hand, embodiments of this application provide a comment information processing apparatus, including:

[0012] The initiating unit is used to respond to the target object's initiation of a review for the target product and display the review editing interface; the review editing interface includes a main input area and at least one sub-input area, and each sub-input area displays title information;

[0013] The first input unit is configured to display first content information in the main input area in response to a first input operation of the target object when the input focus is on the main input area;

[0014] The second input unit is configured to, in response to a second input operation of the target object, display second content information in the secondary input area when the input focus is on the secondary input area, wherein the second content information includes the title information and the information input by the second input operation;

[0015] The display unit is configured to display target comment information in response to the comment end operation of the target object, wherein the target comment information includes the first content information and the second content information.

[0016] Optionally, the review editing interface further includes a rating area, which displays a rating selector. The review information processing device in this embodiment further includes:

[0017] The selection unit is used to display first rating information in the rating area in response to the target object's selection operation on the rating selector.

[0018] Optionally, in the comment information processing apparatus of this application embodiment, the rating selector includes at least one of a sliding rating selector, a star rating selector, a drop-down menu rating selector, or a scrolling rating selector. The sliding rating selector displays a rating progress bar, which is used to represent at least two rating information. The star rating selector displays at least two rating stars. The drop-down menu rating selector displays drop-down menu options for at least two rating information. The scrolling rating selector displays a list option for at least two rating information.

[0019] The selection unit includes at least one of the following:

[0020] A sliding selection subunit is used to display the selected rating information as the first rating information in the rating area in response to the sliding selection operation of the target object on the rating progress bar.

[0021] Alternatively, clicking the selected sub-unit is used to respond to the target object's click selection operation on the rating star, and to display the selected rating star as the first rating information in the rating area;

[0022] Alternatively, a drop-down selection subunit is used to display the selected rating information as the first rating information in the rating area in response to the target object's drop-down selection operation on the drop-down menu option;

[0023] Alternatively, a scroll selection subunit is configured to display the selected rating information as the first rating information in the rating area in response to a scroll selection operation of the target object on the list options.

[0024] Optionally, the comment information processing apparatus in this embodiment of the application further includes:

[0025] An initialization unit is configured to display initialized first content information in the main input area, wherein the initialized first content information is associated with the first rating information.

[0026] Optionally, in the comment information processing apparatus of this application embodiment, the second input unit is specifically used for:

[0027] When the input focus is on the secondary input area, the cursor position is anchored behind the title information displayed in the secondary input area;

[0028] In response to a second input operation from the target object, the input information is displayed after the title information in the sub-input area where the cursor is located.

[0029] Optionally, the comment information processing apparatus in this embodiment further includes:

[0030] The adjustment unit is used to adjust the sub-input area where the cursor is located to be above the sub-input area that has not been edited when the cursor position is in the sub-input area.

[0031] Optionally, the comment information processing apparatus in this embodiment further includes:

[0032] The generation unit is used to obtain the first rating information of the target object on the target product; and generate the title information based on the first rating information.

[0033] Optionally, in the comment information processing apparatus of this application embodiment, the generation unit is specifically used for:

[0034] Obtain fourth content information for each dimension of the target product;

[0035] Based on the sub-score information, determine the evaluation feature words corresponding to each dimension;

[0036] For each dimension, the title information is generated based on the corresponding evaluation feature words and the fourth content information.

[0037] Optionally, the comment information processing apparatus in this embodiment further includes:

[0038] The updating unit is configured to determine the second rating information of the target object for the target product based on the first content information and the second content information; and update the title information displayed in at least one unedited sub-input area, or add a sub-input area displaying new title information, based on the second rating information.

[0039] Optionally, in the comment information processing apparatus of this application embodiment, the updating unit is specifically used for:

[0040] The first content information is input into the trained scoring model to extract the first feature data of the first content information, and the second content information is input into the scoring model to extract the second feature data of the second content information.

[0041] The first feature data and the second feature data are fused to obtain the third feature data;

[0042] Based on the third feature data, the second rating information of the target object for the target product is determined by predicting using the rating model.

[0043] Optionally, the comment information processing apparatus in this embodiment further includes:

[0044] The training unit is used to acquire third-party content information and corresponding evaluation tags of batch sample objects' reviews of sample products, wherein the evaluation tags are used to characterize the true results of the sample objects' rating information of the sample products; input the third-party content information into an initialized rating model, process the third-party content information through the rating model to obtain third-party rating information; determine the training loss value based on the third-party rating information and the evaluation tags; update the parameters of the rating model based on the loss value to obtain a trained rating model.

[0045] Optionally, the comment information processing apparatus in this embodiment further includes:

[0046] The merging unit is configured to, in response to the comment end operation of the target object, add a newline character after the first content information and each of the second content information; and concatenate the first content information and each of the second content information after adding the newline character to obtain the target comment information.

[0047] Optionally, the comment information processing apparatus in this embodiment further includes:

[0048] The display unit is used to respond to the comment submission operation of the target object, perform permission verification on the target object, and perform content verification on the target comment information; when the permission verification of the target object passes and the content verification of the target comment information passes, the target comment information is published to the comment area corresponding to the target product for display.

[0049] On the other hand, embodiments of this application provide an electronic device, including a processor and a memory;

[0050] The memory is used to store computer programs;

[0051] The processor executes the computer program to implement the aforementioned method for processing comment information.

[0052] On the other hand, embodiments of this application provide a computer-readable storage medium storing a computer program that is executed by a processor to implement the aforementioned method for processing comment information.

[0053] This application also discloses a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method for processing comment information.

[0054] The embodiments of this application include at least the following beneficial effects: This application provides a method, apparatus, electronic device, and storage medium for processing comment information. When a target object comments on a target product, a comment initiation operation is triggered, and the terminal device displays a comment editing interface. The comment editing interface includes a main input area and at least one secondary input area. The secondary input area displays title information, which provides guidance for the target object to comment on the target product, helps the target object to summarize the comment content, effectively reduces the difficulty of editing comment information, and improves the efficiency of comment information editing. After the target object completes the editing operation in the main input area and the secondary input area, a comment end operation is triggered, which integrates all the content in the main input area and the edited secondary input area to obtain and display the target comment information. The unedited secondary input areas are discarded. This application can provide a flexible and convenient comment editing method for the target object, which not only reduces the difficulty of editing comment information but also helps to improve the reference value of comment information. Attached Figure Description

[0055] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0056] Figure 1 A schematic diagram of the interface of an online shopping application provided in an embodiment of this application;

[0057] Figure 2 This is a schematic diagram of the interface of a video playback application in related technologies;

[0058] Figure 3 A schematic diagram illustrating the implementation environment of a comment information processing method provided in this application embodiment;

[0059] Figure 4 A flowchart illustrating a method for processing comment information provided in an embodiment of this application;

[0060] Figure 5 This is a schematic diagram of a comment editing interface provided in an embodiment of this application;

[0061] Figure 6 A schematic diagram illustrating the display of first content information in the main input area, as provided in this application embodiment;

[0062] Figure 7 This application provides a schematic diagram of displaying target comment information as an embodiment of the present application.

[0063] Figure 8 A schematic diagram of a rating selector supporting multi-dimensional evaluation provided in an embodiment of this application;

[0064] Figure 9 A schematic diagram illustrating the anchoring of the cursor position in a secondary input area, as provided in an embodiment of this application;

[0065] Figure 10 A schematic diagram illustrating the position of a transformed sub-input region provided in an embodiment of this application;

[0066] Figure 11 This is a schematic diagram of the comment section of a video playback application provided in an embodiment of this application;

[0067] Figure 12 This application provides a schematic diagram of the architecture of a review publisher.

[0068] Figure 13 A flowchart illustrating data processing logic on the terminal device side, provided as an embodiment of this application;

[0069] Figure 14 A schematic diagram of data flow provided in an embodiment of this application;

[0070] Figure 15 A structural block diagram of a comment information processing device provided in an embodiment of this application;

[0071] Figure 16 This is a structural block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0073] It is understood that the terms “first,” “second,” etc., used in this application may be used to describe various concepts herein, but unless otherwise stated, these concepts are not limited by these terms. These terms are used only to distinguish one concept from another.

[0074] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0075] Before providing a detailed description of the embodiments of this application, necessary explanations will be given for the technical terms or characteristic terms that may be involved in the embodiments of this application:

[0076] 1) O2O, short for Online To Offline, refers to combining offline business opportunities with the Internet, making the Internet a platform for offline transactions.

[0077] 2) Feature data: also called embedding representation, is mainly a distributed representation of the input "object" generated by the neural network model. The main function of embedding representation is to transform the high-dimensional sparse vectors of the original data into low-dimensional, dense vectors. At the same time, the distance between vectors can reflect the similarity between data, thus facilitating the processing of downstream models, especially deep learning models.

[0078] 3) Artificial Intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to have perception, reasoning, and decision-making capabilities. AI technology is a comprehensive discipline involving a wide range of fields, encompassing both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operating / interactive systems, and mechatronics. AI software technologies mainly include computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0079] 4) Machine Learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence. Its applications span all areas of artificial intelligence. Machine learning (deep learning) typically includes techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and instructional learning.

[0080] 5) Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and cryptographic algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include the underlying blockchain platform, the platform's product and service layer, and the application service layer. Blockchains can include public blockchains, consortium blockchains, and private blockchains. A public blockchain is one where anyone can access the blockchain network at any time to read data, send data, or compete for ledger entries; a consortium blockchain is one jointly managed by several organizations or institutions; a private blockchain is one with a degree of centralized control, where the right to write to the ledger is controlled by a specific organization or institution, and data access and use are subject to strict permission management.

[0081] The comment information processing method provided in this application mainly relates to various network applications in the Internet field. Specifically, the comment information processing method provided in this application can use information processing technology and machine learning / deep learning technology to process relevant data. This comment information processing method can be executed in various application scenarios of network applications:

[0082] For example, in online shopping scenarios, after purchasing a product, consumers can leave a review; the review is then published in the product's corresponding review section. When other consumers are selecting products online, in addition to comparing the product information itself, they can also read the reviews in the product's review section to decide whether to purchase a particular product, thus reducing the likelihood of buying a poor-quality product or one that does not meet their needs.

[0083] For example, in food delivery applications, consumers can search through the platform to find information on restaurants, hotels, and other offline businesses near a specified location. They can also search for restaurant information within those businesses. When selecting a business or restaurant, consumers can refer to reviews from other consumers to determine if the food's taste and quality, the business's after-sales service, delivery speed, and other factors meet their needs, thus choosing a suitable food delivery service.

[0084] For example, in video playback applications, viewers can evaluate a film or television work and provide comments. Other viewers can then use these comments to determine if the work suits their preferences, making it easier to choose their preferred genre and enhancing the viewing experience.

[0085] Of course, it should be noted that the above application scenarios are merely illustrative and do not imply any limitation on the actual application of the comment information processing method in the embodiments of this application. Those skilled in the art will understand that the comment information processing method provided in the embodiments of this application can be used to perform specified tasks in different application scenarios.

[0086] Furthermore, it should be noted that in various specific embodiments of this application, when processing data related to the identity or characteristics of the target object, such as the target object's information, behavioral data, historical data, and location information, is required, the target object's permission or consent will be obtained first. Moreover, the collection, use, and processing of this data will comply with the relevant laws, regulations, and standards of the relevant countries and regions. Additionally, when embodiments of this application require obtaining sensitive information about the target object, separate permission or consent from the target object will be obtained through pop-ups or redirection to a confirmation page. Only after obtaining the target object's separate permission or consent will the necessary target object-related data for the normal operation of the embodiments of this application be obtained.

[0087] Currently, with the popularization and rapid development of the internet, the O2O model has been widely promoted, and online applications have gradually integrated into people's lives, providing a variety of services. For example, video playback applications can play various movies and TV shows, bringing viewers a rich audio-visual entertainment experience; online shopping applications can support online shopping, improving the convenience and efficiency of shopping.

[0088] In related technologies, to help more people understand the features of a product, applications often include review functions. Users who have purchased or used a product can post their reviews on the corresponding application platform. This allows other users browsing the product to refer to reviews from previous buyers or users to decide whether to use or purchase it. For example, referring to... Figure 1 , Figure 1 This is a schematic diagram of the interface of an online shopping application provided in this application embodiment. Figure 1 The online shopping application interface shown includes a product display area 110 and a review information display area 120. The product display area 110 can display image information of the product, such as a three-dimensional or two-dimensional diagram of the product's overall appearance or some components; this application does not limit this. The review information display area 120 can display reviews posted by those who have purchased or used the product, commenting on related services. Furthermore, in some possible implementations, the positive review rate and other data can be calculated by aggregating the positive or negative review categories of each review and simultaneously displayed in the review information display area 120 to help shoppers select products.

[0089] Understandably, in the current era of booming internet big data technology, product reviews serve as an informational representation of consumers' attitudes towards products, possessing significant value for data analysis. Analyzing these reviews provides consumers with detailed, authentic, and transparent product evaluations, enabling them to understand products from multiple perspectives and facilitate decision-making. For businesses, it provides insights into consumer preferences and market demands, thereby helping to improve service quality and increase customer loyalty.

[0090] However, current in-app comment functionality generally relies entirely on commenters entering information themselves, for example, referring to... Figure 2 , Figure 2 The diagram illustrates the interface of a video playback application in the related art. Figure 2 The online shopping application interface shown includes a comment editing interface 220, which is displayed below the video screen 210. While watching the video screen 210, commenters can enter their comments using the comment editing interface 220.

[0091] Specifically, in the comment editing interface 220, commenters can input text information via the keyboard or insert images, emoticons, etc., using application plugins to generate corresponding comment information. This editing method, for text information, only supports the copy, paste, and conjunction suggestions functions built into the terminal device's text editing software. Editing text information is cumbersome and inefficient, leading to high difficulty and time consumption in comment editing, which can easily generate aversion or resistance, affecting the efficiency and interest of commenting. This, in turn, results in low usage of the comment function provided in the application, impacting user experience and retention. Furthermore, from the perspective of comment content, because commenters have little interest in editing themselves, they generally tend to directly copy and paste large sections of content edited by others, leading to high homogeneity and low reference value. Some commenters lack sufficient understanding of the product and the ability to express and summarize, making it difficult to organize and provide highly valuable comment information, thus reducing the objectivity and usability of the comment information.

[0092] In view of this, this application provides a method for processing comment information. When a target object comments on a target product, a comment initiation operation is triggered, and the terminal device displays a comment editing interface. The comment editing interface includes a main input area and at least one secondary input area. The secondary input area displays title information, which provides guidance for the target object to comment on the target product, helps the target object to summarize the comment content, effectively reduces the difficulty of editing comment information, and improves the efficiency of comment information editing. After the target object completes the editing operation in the main input area and the secondary input area, a comment end operation is triggered. All content in the main input area and the edited secondary input area can be integrated to obtain and display the target comment information, while the unedited secondary input areas are discarded. This application can provide a flexible and convenient comment editing method for the target object, which not only reduces the difficulty of editing comment information but also helps to improve the reference value of comment information.

[0093] Reference Figure 3 , Figure 3 This illustration shows a schematic diagram of an implementation environment for the comment information processing method provided in this application embodiment. In this implementation environment, the main hardware and software components involved include a terminal device 310 and an application server 320, with a communication connection between the terminal device 310 and the application server 320.

[0094] Specifically, in Figure 3 In the illustrated implementation environment, the object editing the comment information can be referred to as the target object, and the terminal device 310 is the terminal device used by the target object. An application can be installed on the terminal device 310. This application can provide the target object with relevant product information or product services and provide comment functions for these products, or simply provide comment functions for products. In this embodiment, the type of application and the specific product information content and category provided are not limited. For example, in some embodiments, the application can be a video playback application, which can provide the target object with corresponding video product playback services and video comment functions; in other embodiments, the application can also be a financial management application, which can provide the target object with corresponding financial product information and financial product comment functions.

[0095] Of course, it's understandable that any application that can be configured with comment functionality can be based on... Figure 3 The given implementation environment is used to build the corresponding system architecture. The types of applications installed in the terminal device 310 are not limited to those given above, and can be flexibly configured according to actual needs. This application does not impose any restrictions on this.

[0096] The comment information processing method provided in this application embodiment can be executed on the terminal device 310 side, or through data interaction between the terminal device 310 and the application server 320. For example, in some embodiments, the terminal device 310 can respond to a target object's comment on a target product by obtaining relevant business data from the application server 320, and then displaying a comment editing interface locally. The target object can then input information in the comment editing interface displayed on the terminal device 310. Correspondingly, the terminal device 310 displays first content information and second content information in the comment editing interface, and sends the first content information and second content information to the application server 320. The application server 320 merges the first content information and second content information to obtain the target comment information. Then, the application server 320 sends the target comment information back to the terminal device 310 for display by the target object. Furthermore, after the target object selects to submit, the application server 320 can publish the target comment information to the comment section corresponding to the target product, so that other objects can also see the target comment information edited by the target object through the terminal device.

[0097] The terminal device 310 in the above embodiments may include smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and in-vehicle terminals, but is not limited to these.

[0098] Application server 320 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0099] In addition, application server 320 can also be a node server in a blockchain network.

[0100] The terminal device 310 and the application server 320 can establish a communication connection via a wireless network or a wired network. This wireless or wired network uses standard communication technologies and / or protocols. The network can be the Internet or any other network, including but not limited to any combination of Local Area Network (LAN), Metropolitan Area Network (MAN), Wide Area Network (WAN), mobile, wired or wireless networks, private networks, or virtual private networks. Furthermore, these hardware and software components can use the same or different communication connection methods; this application does not impose specific limitations in this regard.

[0101] Of course, this is understandable. Figure 3 The implementation environment described in this application is merely one of the optional application scenarios for the comment information processing method provided in this embodiment. The actual application is not fixed. Figure 3 The software and hardware environment shown is illustrated. The methods provided in this application can be applied to various technical fields, such as shopping, games, audio and video, social networking, cloud technology, artificial intelligence, smart transportation, and assisted driving, and this application does not impose specific limitations on them.

[0102] Below, in conjunction with the above description of the implementation environment, a method for processing comment information provided in the embodiments of this application will be described in detail.

[0103] First, such as Figure 4 As shown in the embodiments of this application, a method for processing comment information is provided, which can be applied to the above-mentioned... Figure 3 The terminal device 310 can also be implemented between the terminal device 310 and the application server 320, or it can be executed on any other device or equipment with data processing capabilities. (See reference...) Figure 4 The method specifically includes, but is not limited to, steps 410 to 440:

[0104] Step 410: In response to the target object's comment on the target product, display the comment editing interface; the comment editing interface includes a main input area and at least one sub-input area, and each sub-input area displays title information;

[0105] In this step, the object that needs to post comments is denoted as the target object. The target object generally refers to a person, but can also include other artificial intelligence devices or programs capable of performing interactive actions. The product corresponding to the comments posted by the target object is denoted as the target product. It is understood that in this embodiment, a product refers to a combination of goods and services intended for use or consumption, which may include tangible physical goods, intangible virtual goods, commercial services, or combinations thereof. For example, clothing and computers on shopping platforms; music and videos in audio and video software; props or paid services in online games; and stocks and funds on securities platforms all fall within the scope of products in this embodiment.

[0106] When a target user wants to comment on a target product, they can trigger a request operation on the terminal device. In this embodiment, this operation is referred to as a comment initiation operation. After receiving the comment initiation operation from the target user, the terminal device can respond to the comment initiation operation and display a corresponding interface, which is referred to as the review editing interface in this embodiment. In this embodiment, there are no restrictions on the size of the review editing interface, the proportion of the terminal device's display screen it occupies, or the display method of the interface. For example, in some embodiments, the review editing interface can fill the entire display screen of the terminal device and be displayed as a separate interface; in other embodiments, the review editing interface can be overlaid on other interfaces of the application and displayed as a floating window.

[0107] Specifically, please refer to Figure 5 , Figure 5 This illustration shows a schematic diagram of a comment editing interface provided in an embodiment of this application. Figure 5 The comment editing interface shown includes a main input area and a secondary input area. The main input area 510 displays the text "Say a few words, click below to try the subheading," which is an initial prompt. In some embodiments, this prompt can be omitted, or the specific content displayed can be flexibly adjusted as needed; this application does not impose any restrictions on this.

[0108] Figure 5 The comment editing interface shown includes three sub-input areas, which are referred to as the first sub-input area 520, the second sub-input area 530, and the third sub-input area 540 in this embodiment. The first sub-input area 520 displays the text "A scene that's hard to criticize"; the second sub-input area 530 displays the text "A confusing scene"; and the third sub-input area 540 displays the text "A character I don't like". The text displayed in these sub-input areas can serve as title information.

[0109] In particular, in the embodiments of the present application, before the title information in each secondary input area, a "|" symbol is also displayed. This identifier is used to mark the secondary input area to distinguish it from the previous primary input area. Of course, in some embodiments, this identifier can also be omitted or replaced with other identifiers, and the present application does not limit this.

[0110] In the embodiments of the present application, the primary input area 510, the first secondary input area 520, the second secondary input area 530, and the third secondary input area 540 can all be used to input comment information. Specifically, the primary input area 510 allows the target object to input various content that it wants to express, such as the overall feelings and experiences of the target product. Each secondary input area is pre-planned to display corresponding title information, and the title information can provide guidance for the target object to input comment information. The target object can input the content corresponding to the title information into the specified secondary input area. For example, as Figure 5 shown in the comment editing interface, if the target object watches a certain film or television work in a video playback application but is not particularly fond of it, then based on Figure 5 the title information displayed in the first secondary input area 520, the second secondary input area 530, and the third secondary input area 540, the target object can conveniently make targeted comments on the disliked clips, scenes, or characters in the film or television work. On the one hand, it can reduce the editing difficulty of the comment information and improve the efficiency of editing the comment information. On the other hand, it also facilitates the target object to summarize and organize the places it doesn't like, improves the content richness of the comment information, and makes the comment information have a higher reference value.

[0111] Of course, it can be understood that in the embodiments of the present application, the role of the title information is to provide guidance for the target object to edit the comment information rather than to limit the content of the comment information. In some embodiments, the information input by the target object in the secondary input area can match the title information or there may be no corresponding relationship.

[0112] In the embodiments of the present application, when the target object edits the comment information, it can click on the display screen of the terminal device to confirm the input focus. The input focus can help determine which area the target object wants to input content into currently, so as to facilitate putting the corresponding input content into the corresponding primary input area or secondary input area for display.

[0113] Step ④20: When the input focus is in the primary input area, in response to the first input operation of the target object, display the first content information in the primary input area;

[0114] In this step, as described above, the target object can confirm the input focus by clicking on the display screen of the terminal device. Therefore, for the terminal device executing the comment information processing method in this embodiment, the location of the input focus can be detected during the process of the target object editing comment information. In this embodiment, the location of the input focus refers to the location where the target object last clicked on the comment editing interface. Based on the foregoing description, it can be understood that when it is determined that the input focus is in the main input area, it means that the target object currently wants to input content in the main input area. At this time, in response to the input operation of the target object, the content information input by the target object can be received, and then the content information can be displayed in the main input area. In this embodiment, the operation of the target object inputting content information in the main input area is recorded as the first input operation, and the content information input by the target object in the main input area through the first input operation is recorded as the first content information.

[0115] It is understood that the first content information in the embodiments of this application may include, but is not limited to, information in any format such as text, characters, numbers, and emoticons. Furthermore, parameters such as the language type and data input limit of the first content information can be flexibly adjusted as needed, and this embodiment of the application does not impose any restrictions on this.

[0116] In this application, the specific form of the first input operation is not limited. For example, in some embodiments, the target object can input the first content information by keyboard editing; in some embodiments, the target object can input the first content information by handwriting recognition; in other embodiments, the target object can also input the first content information by voice recognition.

[0117] Specifically, it should be noted that the foregoing embodiments described the ability to initialize and display prompt information in the main input area, thereby prompting the target user to edit comment information. For example... Figure 5 As shown, the main input area 510 originally displayed the prompt message "Say a few words, click below to try the subheading." When the target object performs its first input operation, after receiving the content information input by the target object, a clearing operation can be performed in the main input area first, and then the information input by the target object can be displayed in the main input area as the first content information. For example, refer to... Figure 6 , Figure 6 This diagram illustrates the display of first content information in the main input area 510. Figure 6In the original text, the prompt message "Say something, click below to try the subheading" has been removed. In its place, the content message entered by the target user during the first input operation, "I think it's a bit boring, not very pretty," is displayed. Of course, to make the prompt message and the first content message more distinct, the fonts used for the two can be different. For example, the font color corresponding to the prompt message can be lighter, and the font color corresponding to the first content message can be slightly darker, thus making it easier to distinguish.

[0118] Step 430: When the input focus is on the secondary input area, in response to the second input operation of the target object, the second content information is displayed in the secondary input area, wherein the second content information includes title information and information input by the second input operation;

[0119] In this step, similar to step 420, when the input focus is determined to be in the secondary input area, it indicates that the target object currently wants to input content in the secondary input area. At this time, in response to the target object's input operation, the content information input by the target object can be received, and then the content information can be displayed in the secondary input area. The title information originally displayed in the secondary input area is retained. Thus, in an edited secondary input area, both the title information and the content information input by the target object will be displayed simultaneously. In this embodiment, the operation of the target object inputting content information in the secondary input area is recorded as the second input operation, and the content information input by the target object through the second input operation in the secondary input area, along with the title information originally displayed in the secondary input area, is recorded as the second content information.

[0120] It is understood that in this embodiment, the target object performs a second input operation in the sub-input area based on the title information, which can provide the target object with guidance on commenting on the target product and reduce the difficulty of editing comment information. The content information entered in the second input operation is displayed together with the original title information, which can improve the completeness and reference value of the comment information. Other objects can quickly locate the comment content they need to know based on the title information, thereby improving the usability of the comment information.

[0121] Similarly, in the embodiments of this application, the second content information may also include, but is not limited to, information in any format such as text, characters, numbers, or emoticons; the language type, data input limit, and other parameters of the second content information, as well as the specific form of the second input operation, can be flexibly adjusted as needed, and will not be elaborated here.

[0122] Step 440: In response to the comment end operation of the target object, display the target comment information, wherein the target comment information includes first content information and second content information.

[0123] In this step, after the target object completes the input operations in the main input area and each of the sub-input areas, it can trigger a request operation on the terminal device to end the comment editing task. In this embodiment, this is referred to as the comment end operation. After receiving the comment end operation from the target object, the terminal device can respond to the comment end operation by displaying the comment information currently edited by the target object. Specifically, the comment information includes first content information and second content information, which can be obtained by merging the first content information and the second content information. In this embodiment, this is referred to as the target comment information. Here, in addition to the first content information and the second content information, the target comment information may also contain other content information, such as the target object's identity information or information associated with the target object. This application does not impose any restrictions on this.

[0124] In this step, the target object can interact with the terminal device, thereby triggering a comment end operation for the comment information editing task in the application. The terminal device can respond to this comment end operation by triggering the execution of the application's business logic and displaying the target comment information. This application does not limit the specific implementation of the interaction between the target object and the terminal device to trigger the comment end operation.

[0125] Specifically, for example, the terminal device can receive interactive operations from the target object on relevant components, and regard such interactive operations as comment termination operations. Here, the relevant components can be components configured in the terminal device's underlying system or components configured in the application, and their implementation can be flexibly set as needed. For example, in some embodiments, the component can be a component with voice interaction capabilities in the terminal device, which can recognize the target object's voice data through voice recognition technology. The target object can interact with the terminal device through voice, such as saying "End editing" to the terminal device, thereby triggering the comment termination operation. In some embodiments, the component can also be a touch component, for example, referring to... Figure 5 On the comment editing interface, a "Publish" icon button 550 can be set. When the target user wants to end the comment editing task, they can click the icon button 550 displayed on the terminal device screen. At this time, the target user's click operation can be regarded as the comment ending operation. Of course, it is understood that in some embodiments, the received touch, swipe, or other operations of the target user in a designated screen area of ​​the terminal device can also be regarded as the comment ending operation. This application does not make specific limitations on this. Furthermore, similarly, other operations in the embodiments of this application can also be set with reference to the implementation method of the comment ending operation, which will not be described in detail here.

[0126] Specifically, in one possible implementation, the target comment information can be obtained through the following steps:

[0127] In response to the end of the comment for the target object, add a newline character after the first content information and each of the second content information;

[0128] The target comment information is obtained by concatenating the first content information (after adding a line break) and each of the second content information.

[0129] In the embodiments of this application, reference is made to Figure 7 , Figure 7 This diagram illustrates how to display target comment information. Figure 7 In this process, the target comment information is obtained by merging the first content information and the three parts of the second content information. During the merging process, in response to the target object's comment completion action, a newline character is added to the end of the first content information and each part of the second content information edited by the target object. Then, the content information after adding the newline character can be concatenated to obtain the target comment information. This way, the target comment information maintains its segmented format, preventing it from being mixed into one paragraph, making it easier to browse.

[0130] It should be noted that in this embodiment, the target comment information is obtained by merging the first content information and the second content information. For the sub-input area that has not been edited by the target object, the title information originally displayed therein is not included in the final target comment information. In this way, when editing comment information, the target object can flexibly select the comment topic they want to edit according to their actual experience and editing needs. For title information that they are not interested in, they can simply skip it without any extra operations, which can greatly improve the flexibility and efficiency of comment information editing.

[0131] It is understood that in the comment information processing method provided in this application embodiment, when a target object comments on a target product, a comment initiation operation is triggered, and the terminal device displays a comment editing interface. The comment editing interface includes a main input area and at least one secondary input area. The secondary input area displays title information, which provides guidance for the target object's comment on the target product, helping the target object summarize the comment content, effectively reducing the difficulty of editing comment information and improving the efficiency of comment information editing. After the target object completes the editing operations in the main input area and the secondary input area, a comment end operation is triggered, which integrates all content in the main input area and the edited secondary input area to obtain and display the target comment information. The unedited secondary input areas are discarded. This application provides a flexible and convenient comment editing method for the target object, not only reducing the difficulty of editing comment information but also improving the reference value of the comment information.

[0132] It should be noted that this application does not limit the existence or specific type of association between the target object and the target product in its embodiments. For example, in some possible implementations, there may be no association between the target object and the target product. That is, whether the target object has purchased or used the target product has no impact on whether the target object has the right to comment on the target product. In other words, each object can freely edit comment information about the target product.

[0133] In other possible implementations, a relationship between the target object and the target product can be required. For example, in some embodiments, for online shopping applications, it can be pre-defined that a user must have purchased or used the target product before they can comment on it. In this case, a purchase or usage relationship exists between the target object and the target product. Before the target object comments on the target product, the application's backend server needs to verify the target object's identity information to determine if they have the authority to post comments. Verification can be performed through channels such as the target object's account purchase history; this application does not impose such restrictions. In some embodiments, for video playback applications, it can be pre-defined that a user must have fully or partially viewed the target product before they can comment on it. In this case, before the target object comments on the target product, the application's backend server can determine if they have the authority to post comments based on the target object's browsing history. This raises the bar for commenting and reduces the probability of malicious comments such as fake reviews or spam.

[0134] In this embodiment, the title information displayed in each sub-input area can be pre-set. For example, for online shopping applications, titles such as "Good Quality," "Satisfactory Logistics Speed," "Average Cost-Effectiveness," and "Poor Quality" can be fixed; for video playback applications, titles such as "Exciting Plot," "Star-Studded," and "Worth Watching" can be fixed. These titles can be randomly displayed in conjunction with the corresponding sub-input areas when the target user edits their review. Of course, it is understandable that when these titles are randomly displayed, there may be a problem of poor correspondence between the titles and the actual target product, causing some interference to the target user's review editing. Therefore, this embodiment also provides other methods for generating title information to improve the matching between title information and target products.

[0135] In the technical field related to this application, besides text-based methods like reviews, a commonly used method for product evaluation is rating. This allows users to rate a product to express their attitude towards it. Ratings can reflect a product's popularity in various aspects and are a simpler form of evaluation compared to reviews. In some embodiments, rating and review functions can be used together, providing users with both rating and review channels for the target product.

[0136] Specifically, in one possible implementation, the review editing interface further includes a rating area displaying a rating selector. The review information processing method provided in this embodiment also includes:

[0137] In response to the target object's selection operation on the rating selector, the first rating information is displayed in the rating area.

[0138] In this embodiment of the application, when a target object evaluates a target product, the review editing interface can also include a rating area. This rating area can display a rating selector, which the target object can use to select a rating for the target product.

[0139] In this embodiment, the target object's rating of the target product is recorded as the first rating information. Specifically, the first rating information can take various forms. For example, in some embodiments, the first rating information can be a specific numerical value, such as setting the rating method to a percentage or a ten-point scale. The magnitude of the first rating information can characterize the target object's evaluation of the target product. For instance, when the rating method is a ten-point scale, the value of the first rating information is between 0 and 10. The closer the value of the first rating information is to 10, the more satisfied the target object is with the target product, and the better the evaluation; conversely, the closer the value of the first rating information is to 0, the more dissatisfied the target object is with the target product, and the worse the evaluation. In other embodiments, the first rating information can be text information with pre-defined levels. For example, the first rating information can include four levels, with corresponding text information such as "Excellent," "Good," "Pass," and "Poor," and these levels are ranked from best to worst as Excellent, Good, Pass, and Poor. When the target audience's first evaluation of the target product is "excellent," it means that the target audience is more satisfied with the target product, and the evaluation is better; conversely, when the target audience's first evaluation of the target product is "poor," it means that the target audience is more dissatisfied with the target product, and the evaluation is worse.

[0140] Of course, it is understood that the above embodiments are only used to illustrate the implementation of the first scoring information in the embodiments of this application, and do not mean that the specific setting method is limited.

[0141] In this application embodiment, the rating selector can be implemented in multiple ways. For example, in some embodiments, the rating selector can be a sliding rating selector, for example, referring to... Figure 5 , Figure 5 The review editing interface shown also includes a sliding rating selector 560. The sliding rating selector 560 displays a rating progress bar, which can be slid horizontally. Different rating information is displayed behind the selected progress bar as it is slid to different positions. For example, Figure 5 In this system, when the rating progress bar is positioned to the left, the rating information can be negative, such as "indifferent"; conversely, when the rating progress bar is positioned to the right, the rating information can be positive, such as "highly recommended". It is understood that this rating progress bar can represent at least two different rating information. In some embodiments, the rating information displayed by the rating progress bar can also be numerical scores, and this application does not limit this. For the sliding rating selector, the target object can select the corresponding rating information as the first rating information through a sliding selection operation.

[0142] In some embodiments, the rating selector can also be a star rating selector, which can display at least two rating stars. For example, the star rating selector can display five star icons. When a target object clicks on one of the star icons, the total number of star icons, including that star icon and the star icons to its left, will be determined as the first rating information. If the target object clicks on the rightmost star icon, the corresponding first rating information is 5 stars. Therefore, the rating star determined by the target object through the click selection operation can be used as the first rating information and displayed in the rating area. Similarly, the rating selector can also be a drop-down menu rating selector or a scrolling rating selector. The drop-down menu rating selector can display a drop-down menu option with at least two rating information, and the scrolling rating selector can display a list option with at least two rating information. The target object can perform a drop-down selection operation on the drop-down menu option or a scrolling selection operation on the list option to determine the first rating information. The specific implementation details of these methods are not elaborated in this application.

[0143] Specifically, in one possible implementation, the title information is generated through the following steps:

[0144] Obtain the target audience's initial rating information for the target product;

[0145] Generate title information based on the first rating information.

[0146] As mentioned earlier, rating information can, to a certain extent, reflect the popularity of a product in various aspects. Therefore, in this embodiment, when generating title information, the first rating information can be referenced to improve the correspondence between the title information and the actual target product, and reduce interference caused by editing the target object's review information. Specifically, the target object's first rating information for the target product can be obtained. If the first rating information is relatively positive, such as a high score or a positive attitude, then the generated title information can also be biased towards a positive one. For example, when it is found that the target object has given a five-star rating to a product purchased online, then positive title information such as "Great quality" or "Satisfactory logistics" can be generated. In this way, the reliability of the title information guidance can be improved, which helps to reduce the difficulty of editing review information and improve the efficiency of editing review information.

[0147] It is understood that in this embodiment, different first rating information may result in different title information generated in the sub-input area. Of course, for some first rating information with relatively small differences, the corresponding title information can also be the same. For example, for film and television works, when the first rating information includes multiple different recommendation levels, such as "disappointing," "indifferent," "okay," "recommended," "highly recommended," etc., if the target audience gives a "recommended" or "highly recommended" level first rating information, the generated title information can be positive, such as "exciting plot" or "realistic acting"; if the target audience gives a "disappointing" or "indifferent" level first rating information, the generated title information can be negative, such as "cringeworthy character" or "awkward background music."

[0148] Specifically, in one possible implementation, the first rating information includes sub-rating information for multiple dimensions of the target product. Based on the first rating information, title information is generated, including:

[0149] Obtain fourth-party information about the target product from various dimensions;

[0150] Based on the sub-score information, determine the evaluation feature words corresponding to each dimension;

[0151] For each dimension, title information is generated based on the corresponding evaluation feature words and fourth content information.

[0152] In some embodiments of this application, the first rating information may contain more detailed content. For example, the first rating information can allow the target audience to evaluate the target product from different dimensions, and for each dimension, there is a corresponding sub-rating information. For instance, for an online shopping application, in addition to overall product reviews, it may also provide evaluation channels for various sub-stages of the shopping service, such as evaluation from multiple dimensions like product packaging, logistics speed, and merchant attitude. In this case, the obtained first rating information may include these sub-rating information, and parsing the sub-rating information can help generate more realistic title information.

[0153] Specifically, please refer to Figure 8 , Figure 8 A schematic diagram of a rating selector that supports multi-dimensional evaluation is shown, which can be displayed in the review editing interface. Figure 8 The rating selector in the application not only allows users to evaluate whether to recommend a film or television work as a whole, but also provides rating progress bars for evaluating the work in three dimensions: plot, acting, and production. Users can slide the corresponding rating progress bar to rate the film or television work in each dimension, thereby obtaining the corresponding sub-rating information for that dimension. In this embodiment, when generating title information based on the first rating information, the fourth content information of the target product in each dimension can be obtained. Here, the fourth content information refers to the related content information included in the target product in the corresponding dimension. For example, for film and television works, the acting dimension may include content information such as actors, lead actors, roles, and supporting roles; the plot dimension may include content information such as scenes, plot twists, and endings. For online shopping products, the merchant attitude dimension may include content information such as feedback time, communication attitude, and after-sales service. After obtaining the fourth content information in each dimension, the evaluation feature words corresponding to each dimension of the target product can be determined based on the sub-rating information. It is understood that there should be a high degree of matching between sub-rating information and evaluation feature words. For example, if the sub-rating information is positive, the corresponding evaluation feature words can also be positive, such as "satisfactory," "liked," or "good." Conversely, if the sub-rating information is negative, the corresponding evaluation feature words should also be negative, such as "terrible," "disgusting," or "disgusting." Of course, in the embodiments of this application, the specific content of the evaluation feature words can be flexibly adjusted and configured as needed, and this application does not impose any restrictions on this.

[0154] After identifying the fourth content information and corresponding evaluation keywords for each dimension of the target product, the evaluation keywords and fourth content information can be combined according to the dimension to generate title information. For example, for film and television works, title information such as "favorite character", "criticized scene", "satisfactory background music", and "terrible plot" can be obtained.

[0155] Understandably, for one dimension of the target product, the fourth content information involves the related content under that dimension, while the sub-rating information represents whether the target audience is satisfied with the services involved in the target product under that dimension. Therefore, the generated title information largely matches the target audience's orientation, reduces the probability of conflict with the actual situation, and can effectively improve the convenience of editing review information.

[0156] It should be noted that, in the embodiments of this application, since the first rating information can be associated with the title information, in some embodiments, when the target object performs a selection operation on the rating selector, not only can the first rating information given by the target object be displayed in the rating area, but the title information displayed in the sub-input area can also be updated. Here, it can be understood that when the target object changes the first rating information, the updated title information can be partial (at least one) or complete; this application does not impose any restrictions on this. Thus, the rating selector not only supports the target object in quickly adjusting the first rating information but also efficiently refreshes the displayed title information, improving the efficiency of the target object in determining appropriate title information, thereby improving the efficiency of editing comment information.

[0157] In some embodiments, the first rating information can also be used to initialize the first content information to further improve the convenience of editing comment information. Specifically, it can be understood that the first rating information is a simplified form of evaluation information, which provides an overall evaluation of the target product, and is similar to the comment approach of the first content information. Therefore, in this embodiment, some short comment information can be adaptively generated based on the first rating information, and this comment information can be used as the initialized first content information and displayed in the main input area. The target object can modify and add its own content based on the initialized first content information, thereby generating comment information faster and better.

[0158] Specifically, in this embodiment of the application, when the target object performs an input operation, it can click on the screen of the terminal device to determine the input focus. When the input focus is on the main input area, the cursor position can be anchored to the main input area, so that the target object can input the edited content information into the main input area based on the first input operation, and display it as the first content information.

[0159] Similarly, in this embodiment, when the input focus of the target object is in the secondary input area, the cursor is also anchored to the secondary input area. Specifically, for example, referring to... Figure 9 Since the title information is pre-displayed in the secondary input area, to facilitate editing of the target object, the cursor 910 can be directly anchored after the title information displayed in the secondary input area. In this way, the target object can input the edited content information into the title information in the secondary input area based on the second input operation, without needing to adjust the cursor position again, thus improving the efficiency of editing the second content information.

[0160] Specifically, in one possible implementation, the method further includes:

[0161] When the cursor is in a secondary input area, adjust the secondary input area where the cursor is located to be above the secondary input area that has not been edited.

[0162] In the embodiments of this application, reference is made to Figure 10 During the editing of a sub-input area by the target user, there may be instances where the title information of some sub-input areas is not of interest, and the user skips editing those sub-input areas. Displaying these unedited sub-input areas in their original positions would significantly hinder the target user's ability to browse the content they have edited. Therefore, in this embodiment, the currently edited sub-input area can be moved upwards. Specifically, for example, the cursor position on the terminal device screen can be detected. When the cursor is positioned within a sub-input area, the sub-input area where the cursor is located can be moved above the unedited sub-input areas. Please refer to... Figure 5 and Figure 10 For example, for Figure 5 In the comment editing interface shown, the second input area 530 was originally above the third input area 540. However, when the user edits, after completing the content in the first input area 520, they are not interested in the title information "Confusing Scene" and believe there is nothing they want to input. Therefore, they skip the second input area 530 and directly edit the third input area 540. At this time, the cursor is located within the third input area 540, so the third input area 540 will be moved above the second input area 530, right next to the already edited content. This improves the user's convenience in browsing the edited comment content, especially for longer comments. It helps the user quickly check which content has been edited and whether there are any omissions, without having to scroll up and down to find it, thus improving the user's experience in editing comment information.

[0163] Specifically, in one possible implementation, the method for processing comment information provided in this application embodiment further includes:

[0164] Based on the first content information and the second content information, determine the target object's second rating information for the target product;

[0165] Based on the second scoring information, update the title information displayed in at least one unedited sub-input area, or add a sub-input area displaying the new title information.

[0166] In this embodiment, in addition to the aforementioned method of generating title information based on the first rating information provided by the target object, the title information can also be generated or updated based on the content entered by the target object during the editing of the review information. Specifically, it can be understood that, compared to the first rating information, the first and second content information edited by the target object can more accurately reflect the target object's attitude towards the target product. Therefore, in this embodiment, another rating information of the target object for the target product can be determined based on the first and second content information, and denoted as the second rating information. Then, based on the second rating information, the title information displayed in at least one unedited sub-input area can be updated, or new title information can be generated, and a sub-input area displaying the new title information can be added to the review editing interface.

[0167] It should be noted that, in this embodiment, the meaning and data format of the second rating information can be set with reference to the aforementioned first rating information. It can be either a rating value or text information with a pre-set level. Furthermore, the process for generating the title information updated or newly generated based on the second rating information is the same as the process for generating the title information based on the first rating information, and will not be repeated here.

[0168] Specifically, in one possible implementation, determining the target object's second rating information for the target product based on the first content information and the second content information includes:

[0169] The first content information is input into the trained scoring model to extract the first feature data of the first content information, and the second content information is input into the scoring model to extract the second feature data of the second content information.

[0170] The first and second feature data are fused to obtain the third feature data;

[0171] Based on the third feature data, a scoring model is used to predict and determine the second rating information of the target object for the target product.

[0172] In this embodiment, a second rating for the target product by the target object is determined based on the first content information and the second content information. This process can be implemented based on machine learning / deep learning algorithms. Specifically, a rating model can be used to predict the second rating information based on the first and second content information. Here, the rating model can be built using any model architecture from machine learning / deep learning algorithms, and this application does not impose any restrictions on it.

[0173] For the scoring model, it can extract feature data of the first content information and feature data of the second content information, recording the feature data of the first content information as the first feature data and the feature data of the second content information as the second feature data. The data format of the first feature data can include at least one of numbers, vectors, matrices, or tensors, and this application does not limit it. Furthermore, it is understood that in the embodiments of this application, the feature data described below are all applicable to the implementation methods of the above data formats.

[0174] Specifically, when extracting feature data, a sequence-to-sequence (Seq2seq) model architecture can be adopted. Sequence-to-sequence models are among the most commonly used models for sequence tasks today, excelling at processing continuous data such as speech, text, and video. Specifically, the framework of a sequence-to-sequence model generally includes two parts: an encoder and a decoder. In this embodiment, the encoder is primarily used to extract feature data. The encoder in the sequence-to-sequence model can be implemented using a recurrent neural network (RNN), such as the commonly used Long Short-Term Memory (LSTM) network. Of course, it is understood that the type of encoder described above can be flexibly adjusted as needed, and this application does not impose any restrictions on it.

[0175] After obtaining the first and second feature data, feature fusion can be performed on the first and second feature data to obtain the third feature data. Here, the method of feature fusion can include data concatenation, data weighting, etc., and this application does not impose specific limitations on this. Based on the third feature data, a second scoring information can be obtained through a scoring model.

[0176] It should be noted that in the field of machine learning, models generally need to be trained and tuned before being put into use. Therefore, this application embodiment also provides a method for training a scoring model. Specifically, in one possible implementation, the scoring model is trained through the following steps:

[0177] Obtain the third content information of the batch of sample objects' reviews of the sample products and the corresponding evaluation tags, whereby the evaluation tags are used to characterize the true results of the sample objects' rating information of the sample products;

[0178] The third content information is input into the initialized scoring model, and the scoring model processes the third content information to obtain the third scoring information.

[0179] The training loss value is determined based on the third-party scoring information and evaluation labels;

[0180] The parameters of the scoring model are updated based on the loss value to obtain the trained scoring model.

[0181] In this embodiment, when training the rating model, the content information of multiple sample objects' comments on the sample product can be obtained, denoted as the third content information. The actual rating results of the sample objects on the sample product are also obtained, serving as the evaluation label corresponding to the third content information. Next, the third content information is input into the initialized rating model. Then, referring to the aforementioned method for extracting the first feature data, the feature data corresponding to the sample object is extracted through the initialized rating model based on the third content information, denoted as the fourth feature data. Based on the fourth feature data, the rating information of the sample object on the sample product can be predicted, denoted as the third rating information.

[0182] Under normal circumstances, it is desirable for the rating model to predict rating information as closely as possible to the actual situation, that is, the true result of the sample object's rating information for the sample product. In other words, the predicted third rating information should be close to or the same as the evaluation label. If the two differ significantly, it indicates that the rating model is ineffective. Therefore, in this embodiment, the accuracy of the rating model can be determined based on the evaluation label and the third rating information. Specifically, the deviation between the evaluation label and the third rating information can be determined to obtain the loss value corresponding to the rating model. After obtaining the loss value, the accuracy of the rating model can be evaluated based on the magnitude of the loss value, so as to perform backpropagation training on the rating model and update its internal parameters.

[0183] Specifically, for machine learning models, the accuracy of their predictions can be measured by a loss function. The loss function is defined on a single training data point and measures the prediction error of that data point. Specifically, the loss value is determined by the label of that single training data point and the model's prediction result for that data. However, in actual training, a training dataset contains many data points. Therefore, a cost function is generally used to measure the overall error of the training dataset. The cost function is defined on the entire training dataset and calculates the average prediction error of all training data points, providing a better measure of the model's prediction performance. For general machine learning models, the aforementioned cost function, plus a regularization term to measure model complexity, forms the training objective function. Based on this objective function, the loss value of the entire training dataset can be calculated. Many types of loss functions are commonly used, such as 0-1 loss, squared loss, absolute loss, logarithmic loss, and cross-entropy loss, which will not be elaborated upon here.

[0184] Specifically, in one possible implementation, the method for processing comment information provided in this application embodiment further includes:

[0185] In response to the comment submission operation of the target object, perform permission verification on the target object and content verification on the target comment information;

[0186] Once the target object's permissions are verified and the target comment information's content is verified, the target comment information will be published to the comment section corresponding to the target product for display.

[0187] In this embodiment, after the target object completes editing the target comment information, an operation request can be triggered to display the target comment information in the comment section corresponding to the target product. Specifically, this operation request is denoted as a comment submission operation. The terminal device can respond to the target object's comment submission operation by verifying whether the target object has the permission to publish comment information, and can simultaneously perform content verification on the target comment information. When it is determined that the target object's permission verification passes and the content verification of the target comment information passes, the target comment information can be published to the comment section corresponding to the target product for display. For example, refer to... Figure 7When displaying the target comment information, corresponding operation buttons can be provided to the target audience, such as the edit button 710, the save button 720, or the submit button 730. When the target audience clicks the edit button 710, they can return to the comment editing interface; when they click the save button 720, the comment information can be saved, awaiting modification or submission by the target audience; when they click the submit button 730, the comment submission operation is triggered, and the verification and publication process begins. (Refer to...) Figure 11 , Figure 11 The illustration shows a comment section of a video playback application. When displaying comment information, it can show information about the editor of the comment, such as account ID, avatar, etc., and can also show the time when the comment was published.

[0188] It should be noted that since some comments may be quite long and take up a lot of display space in the comment section, the display length of each comment can be limited. Comments that exceed the fixed length can be collapsed, and other users can click on the comment to view the complete content.

[0189] As described above, the comment information processing method provided in this application embodiment can be applied to video playback applications. The following detailed description of the comment information processing method provided in this application, in conjunction with specific applications, provides a more comprehensive understanding.

[0190] With the rapid innovation of internet media technology, the film and entertainment industry, including cinemas and home entertainment, is booming. Movies have become a daily entertainment option for people, and the acceptance and welcome of movies have generated a large amount of commentary. Extracting subjective viewpoints from these comments and judging the public's positive or negative inclinations is an important issue in information extraction and mining within the field of natural language processing. Furthermore, film commentary demonstrates its value in terms of value transmission and shaping the film and television environment; analyzing it contributes to the in-depth development of film and television studies. Therefore, conducting bias analysis on film commentary is of great significance.

[0191] This application provides a method for processing comment information, which can be applied to video playback applications. It provides viewers with a flexible and convenient way to edit comments, which not only reduces the difficulty of editing comment information but also helps to improve the reference value of comment information.

[0192] In this embodiment of the application, a comment publisher can be developed using UI technology within a video playback application, for example, referring to... Figure 12 , Figure 12This diagram illustrates the architecture of a review publisher according to an embodiment of this application. The review publisher includes multiple modules. The title bar displays the reviewer's account and avatar information. The rating module is primarily for reviewers to rate content and can include a rating selector. Reviewers can use the rating selector to rate film and television works. Within the rating module, a main input box (main and secondary input areas) and multiple secondary input boxes (i.e., secondary input areas) can be set. Figure 12 The application includes three secondary input fields, and this application does not limit the specific number of these fields; they can be flexibly adjusted as needed. Below the secondary input fields is a function button area for selecting certain functions, such as inserting images and emoticons. Below the function button area is the keyboard area, where commenters can type in their comments to edit the information.

[0193] In this embodiment of the application, when editing comment information in a video playback application, the data processing logic on the terminal device side is as follows: Figure 13 As shown. (Refer to...) Figure 13 After a commenter enters the review publisher, the terminal device can retrieve the review information parsing function to analyze the rating information provided by the commenter. Furthermore, a rating selector can be activated, such as displaying a rating progress bar by dimension, allowing the commenter to provide rating information. Next, the terminal device can check whether the video playback application is configured with the method described in this embodiment and whether it has a function to provide guiding title information. If this function is not configured, the main editing box is used in the original way, allowing the commenter to edit the review information themselves and display it through the review publisher. If the title information function is configured, the rating information can be parsed in a loop after the commenter provides the review information to obtain the title information corresponding to each dimension. Of course, in some cases, the commenter may not have provided a rating, resulting in some title information not being parsed. In this case, the commenter still needs to edit the review information themselves and display it through the review publisher. If title information is parsed, it can be recorded and inserted into a sub-input box with the same number of title information entries in the review publisher, displaying the title information in the sub-input box for easy editing by the commenter.

[0194] Reference Figure 14 , Figure 14This diagram illustrates the data flow when editing comment information in a video playback application. Specifically, on the application server side, multi-dimensional comment information for film and television works can be pre-configured and associated with rating information. On the terminal device side, when a commenter rates a film or television work, the corresponding comment information can be retrieved, parsed, and a title generated. This title is then displayed in a secondary input box. The commenter can edit their comment information on the terminal device and submit it, concatenating the edited content. On the application server side, when the terminal device sends a submission request, the request is validated. If successful, a review post is generated, which can include the commenter's rating and comment information. Furthermore, this review post can be associated with pre-configured comment information for other viewers to refer to, enhancing the richness of the comment information.

[0195] The following is a portion of the code data corresponding to the interface section of the video playback application:

[0196] / / Get Review Information Service

[0197] service GradeService{

[0198] rpc getGradeInfo(GetGradeInfoRequest)returns(GetGradeInfoResponse);

[0199] / / Request to retrieve review information

[0200] message GetGradeInfoRequest{

[0201] string data_key = 1; / / Request datakey, including activity ID, rating ID, etc.

[0202] uint64 local_play_duration = 2; / / Local real-time viewing duration (last 2 days), unit: seconds

[0203] / / Get review information feedback packet

[0204] message GetGradeInfoResponse{

[0205] repeated GradeItem grade_item_list = 1; / / List of rating dimensions

[0206] string verify_msg = 2; / / Comment qualification verification prompt message

[0207] map<int32,Operation> operation_map = 3; / / operation_map, the action for review qualification verification.

[0208] uint64 total_play_duration = 4; / / Viewing duration, in seconds

[0209] / / Rating display style

[0210] enum GradeItemType{

[0211] GRADE_ITEM_TYPE_LABEL = 0; / / Tag style, directly displays text options (disappointing, indifferent, okay, recommended, highly recommended).

[0212] GRADE_ITEM_TYPE_STAR = 1; / / Rating style, displays the rating bar, and shows text options based on the star value (disappointing, indifferent, okay, recommended, highly recommended).

[0213] / / Rating Dimensions

[0214] message GradeItem{

[0215] string name = 1; / / Dimension ID

[0216] string name = 2; / / Dimensional text: your recommendation value, plot design, character performance, audiovisual effects, post-production editing

[0217] GradeItemType item_type = 3; / / Rating display style

[0218] repeated GradeOption grade_option_list = 4; / / Grading options

[0219] / / Rating options

[0220] message GradeOption{

[0221] string id = 1; / / Option id

[0222] string name = 2; / Option text: Disappointing, Indifferent, Okay, Recommended, Highly Recommended

[0223] string tips = 3; / / Option prompts

[0224] }

[0225] Reference Figure 15 , Figure 15 This is a schematic diagram of the structure of a comment information processing device provided in an embodiment of this application. The comment information processing device includes:

[0226] The initiating unit 1510 is used to respond to the target object's initiation of a review of the target product and display the review editing interface; the review editing interface includes a main input area and at least one sub-input area, and each sub-input area displays title information;

[0227] The first input unit 1520 is used to display first content information in the main input area in response to the first input operation of the target object when the input focus is on the main input area;

[0228] The second input unit 1530 is used to display second content information in the sub-input area in response to a second input operation of the target object when the input focus is on the sub-input area. The second content information includes title information and information input by the second input operation.

[0229] The display unit 1540 is used to display target comment information in response to a comment end operation of the target object, wherein the target comment information includes first content information and second content information.

[0230] Optionally, the review editing interface also includes a rating area, which displays a rating selector. The review information processing device in this embodiment further includes:

[0231] The selection unit is used to display the first rating information in the rating area in response to the target object's selection operation on the rating selector.

[0232] Optionally, in the comment information processing apparatus of this application embodiment, the rating selector includes at least one of a sliding rating selector, a star rating selector, a drop-down menu rating selector, or a scrolling rating selector. The sliding rating selector displays a rating progress bar, which is used to represent at least two rating information; the star rating selector displays at least two rating stars; the drop-down menu rating selector displays drop-down menu options for at least two rating information; and the scrolling rating selector displays a list option for at least two rating information.

[0233] The selection unit includes at least one of the following:

[0234] The sliding selection sub-unit is used to respond to the sliding selection operation of the target object on the rating progress bar and display the selected rating information as the first rating information in the rating area;

[0235] Alternatively, clicking to select a sub-cell is used in response to the target object's click to select a rating star, and the selected rating star is displayed in the rating area as the first rating information;

[0236] Alternatively, a drop-down selection sub-unit is used to display the selected rating information as the first rating information in the rating area in response to the target object's drop-down selection operation on the drop-down menu option;

[0237] Alternatively, a scroll selection sub-unit can be used to display the selected rating information as the first rating information in the rating area in response to the target object's scroll selection operation on the list options.

[0238] Optionally, the comment information processing apparatus in this embodiment further includes:

[0239] An initialization unit is used to display initialized first content information in the main input area, wherein the initialized first content information is associated with first rating information.

[0240] Optionally, in the comment information processing apparatus of this application embodiment, the second input unit is specifically used for:

[0241] When the input focus is on the secondary input area, the cursor position will be anchored after the title information displayed in the secondary input area;

[0242] In response to a second input operation on the target object, the input information is displayed after the title information in the sub-input area where the cursor is located.

[0243] Optionally, the comment information processing apparatus in this embodiment further includes:

[0244] The adjustment unit is used to move the sub-input area where the cursor is located above the sub-input area that has not been edited when the cursor is in the sub-input area.

[0245] Optionally, the comment information processing apparatus in this embodiment further includes:

[0246] The generation unit is used to obtain the target object's first rating information for the target product; and generate title information based on the first rating information.

[0247] Optionally, in the comment information processing apparatus of this application embodiment, the generation unit is specifically used for:

[0248] Obtain fourth-party information about the target product from various dimensions;

[0249] Based on the sub-score information, determine the evaluation feature words corresponding to each dimension;

[0250] For each dimension, title information is generated based on the corresponding evaluation feature words and fourth content information.

[0251] Optionally, the comment information processing apparatus in this embodiment further includes:

[0252] The updating unit is configured to determine the second rating information of the target object for the target product based on the first content information and the second content information; and update the title information displayed in at least one unedited sub-input area, or add a sub-input area displaying new title information, based on the second rating information.

[0253] Optionally, in the comment information processing apparatus of this application embodiment, the updating unit is specifically used for:

[0254] The first content information is input into the trained scoring model to extract the first feature data of the first content information, and the second content information is input into the scoring model to extract the second feature data of the second content information.

[0255] The first and second feature data are fused to obtain the third feature data;

[0256] Based on the third feature data, a scoring model is used to predict and determine the second rating information of the target object for the target product.

[0257] Optionally, the comment information processing apparatus in this embodiment further includes:

[0258] The training unit is used to acquire third-party content information and corresponding evaluation labels of batch sample objects' reviews of sample products. The evaluation labels are used to represent the true results of the sample objects' rating information of sample products. The third-party content information is input into the initialized rating model, and the rating model processes the third-party content information to obtain the third-party rating information. Based on the third-party rating information and evaluation labels, the training loss value is determined. The parameters of the rating model are updated based on the loss value to obtain the trained rating model.

[0259] Optionally, the comment information processing apparatus in this embodiment further includes:

[0260] The merging unit is used to respond to the comment end operation of the target object by adding a newline character after the first content information and each of the second content information; the first content information and each of the second content information after adding the newline character are concatenated to obtain the target comment information.

[0261] Optionally, the comment information processing apparatus in this embodiment further includes:

[0262] The display unit is used to respond to the comment submission operation of the target object, perform permission verification on the target object, and perform content verification on the target comment information; when the permission verification of the target object passes and the content verification of the target comment information passes, the target comment information is published to the comment section corresponding to the target product for display.

[0263] It is understandable that, such as Figure 4 The content of the comment information processing method embodiments shown is applicable to the comment information processing device embodiments, and the specific functions implemented by the comment information processing device embodiments are the same as those shown. Figure 4 The processing embodiment for comment information shown is the same, and the beneficial effects achieved are the same as those described above. Figure 4 The beneficial effects achieved by the illustrated example of processing comment information are also the same.

[0264] Reference Figure 16 This application also discloses an electronic device, including:

[0265] At least one processor 1610;

[0266] At least one memory 1620 stores a computer program, and the processor 1610 executes the computer program to implement, for example... Figure 4 The illustrated example shows a method for processing comment information.

[0267] It is understandable that, such as Figure 4 The content of the comment information processing method embodiments shown is applicable to the embodiments of this electronic device, and the specific functions implemented by the embodiments of this electronic device are the same as those shown. Figure 4 The method for processing comment information shown is the same as in the embodiment, and the beneficial effects achieved are the same as those described above. Figure 4 The beneficial effects achieved by the illustrated method embodiment for processing comment information are also the same.

[0268] This application also discloses a computer-readable storage medium storing a processor-executable computer program, which, when executed by a processor, is used to implement, for example... Figure 4 The illustrated example shows a method for processing comment information.

[0269] Understandable Figure 4 The content of the comment information processing method embodiments shown is applicable to the embodiments of this computer-readable storage medium. The specific functions implemented by the embodiments of this computer-readable storage medium are the same as those shown in the embodiments. Figure 4 The method for processing comment information shown in the embodiment is the same, and the beneficial effects achieved are the same. Figure 4 The beneficial effects achieved by the illustrated method embodiment for processing comment information are also the same.

[0270] This application also discloses a computer program product, including a computer program, which, when executed by a processor, is used to implement... Figure 4 The illustrated example shows a method for processing comment information.

[0271] Understandable Figure 4 The content of the comment information processing method embodiments shown is applicable to the embodiments of this computer program product. The specific functions implemented in the embodiments of this computer program product are the same as those in the embodiments of this computer program product. Figure 4 The method for processing comment information shown in the embodiment is the same, and the beneficial effects achieved are the same. Figure 4 The beneficial effects achieved by the illustrated method embodiment for processing comment information are also the same.

[0272] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this application are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0273] Furthermore, although this application is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding this application. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional technology for an engineer. Therefore, those skilled in the art can implement the application set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of this application, which is determined by the full scope of the appended claims and their equivalents.

[0274] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0275] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable storage medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable storage medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0276] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0277] In the foregoing description of this specification, the references to terms such as "one embodiment," "another embodiment," or "some embodiments," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0278] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

[0279] The above is a detailed description of the preferred embodiments of this application, but this application is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application, and these equivalent modifications or substitutions are all included within the scope defined by the claims of this application.

Claims

1. A method for processing comment information, characterized in that, The method includes: In response to a target user's comment on a target product, a comment editing interface is displayed; the comment editing interface includes a main input area and at least one sub-input area, and each sub-input area displays title information. When the input focus is on the main input area, in response to the first input operation of the target object, first content information is displayed in the main input area; When the input focus is on the secondary input area, in response to a second input operation of the target object, second content information is displayed in the secondary input area, wherein the second content information includes the title information and the information input by the second input operation; In response to the comment ending operation of the target object, target comment information is displayed, wherein the target comment information includes the first content information and the second content information; The review editing interface also includes a rating area, which displays a rating selector. The method further includes: In response to a selection operation by a target object on the rating selector, first rating information for the target product is displayed in the rating area; wherein, the title information is generated based on the first rating information; Based on the first content information and the second content information, determine the second rating information of the target object for the target product; Based on the second rating information, update the title information displayed in at least one unedited sub-input area, or add a sub-input area displaying new title information.

2. The method for processing comment information according to claim 1, characterized in that, The rating selector includes at least one of a sliding rating selector, a star rating selector, a drop-down menu rating selector, or a scrolling rating selector. The sliding rating selector displays a rating progress bar, which is used to represent at least two rating information. The star rating selector displays at least two rating stars. The drop-down menu rating selector displays drop-down menu options for at least two rating information. The scrolling rating selector displays a list option for at least two rating information. The first rating information displayed in the rating area in response to the target object's selection operation of the rating selector includes at least one of the following: In response to the target object's sliding selection operation on the rating progress bar, the selected rating information is displayed in the rating area as the first rating information; Alternatively, in response to the target object's click selection operation on the rating star, the selected rating star is displayed in the rating area as the first rating information; Alternatively, in response to the target object's drop-down selection operation of the drop-down menu option, the selected rating information is displayed in the rating area as the first rating information; Alternatively, in response to the target object's scrolling selection of the list options, the selected rating information is displayed in the rating area as the first rating information.

3. The method for processing comment information according to claim 1 or 2, characterized in that, After the step of displaying the first rating information, the method further includes: The initial content information is displayed in the main input area, wherein the initial content information is associated with the first rating information; Alternatively, update the title information displayed in at least one of the sub-input areas.

4. The method for processing comment information according to claim 1, characterized in that, When the input focus is on the secondary input area, in response to a second input operation of the target object, displaying second content information in the secondary input area includes: When the input focus is on the secondary input area, the cursor position is anchored behind the title information displayed in the secondary input area; In response to a second input operation from the target object, the input information is displayed after the title information in the sub-input area where the cursor is located.

5. The method for processing comment information according to claim 4, characterized in that, The method further includes: When the cursor position is in the sub-input area, adjust the sub-input area where the cursor position is located to be above the sub-input area that has not been edited.

6. The method for processing comment information according to claim 1, characterized in that, The first rating information includes sub-rating information for multiple dimensions of the target product. Generating the title information based on the first rating information includes: Obtain fourth content information for each dimension of the target product; Based on the sub-score information, determine the evaluation feature words corresponding to each dimension; For each dimension, the title information is generated based on the corresponding evaluation feature words and the fourth content information.

7. The method for processing comment information according to claim 1, characterized in that, The step of determining the second rating information of the target object for the target product based on the first content information and the second content information includes: The first content information is input into the trained scoring model to extract the first feature data of the first content information, and the second content information is input into the scoring model to extract the second feature data of the second content information. The first feature data and the second feature data are fused to obtain the third feature data; Based on the third feature data, the second rating information of the target object for the target product is determined by predicting using the rating model.

8. The method for processing comment information according to claim 7, characterized in that, The scoring model is trained through the following steps: Obtain third content information of batch sample objects' reviews of sample products and the corresponding evaluation tags of the third content information, wherein the evaluation tags are used to characterize the true result of the sample objects' rating information of the sample products; The third content information is input into the initialized scoring model, and the scoring model processes the third content information to obtain the third scoring information; The training loss value is determined based on the third scoring information and the evaluation label. The parameters of the scoring model are updated based on the loss value to obtain the trained scoring model.

9. The method for processing comment information according to claim 1, characterized in that, The target comment information is obtained through the following steps: In response to the comment end operation of the target object, a newline character is added after the first content information and each of the second content information; The first content information after adding the newline character and each of the second content information are concatenated to obtain the target comment information.

10. The method for processing comment information according to claim 1, characterized in that, The method further includes: In response to the comment submission operation of the target object, the system performs permission verification on the target object and content verification on the target comment information. Once the permission verification of the target object is passed and the content verification of the target comment information is passed, the target comment information will be published to the comment section corresponding to the target product for display.

11. A device for processing comment information, characterized in that, The device includes: The initiating unit is used to respond to the target object's initiation of a review for the target product and display the review editing interface; the review editing interface includes a main input area and at least one sub-input area, and each sub-input area displays title information; The first input unit is configured to display first content information in the main input area in response to a first input operation of the target object when the input focus is on the main input area; The second input unit is configured to, in response to a second input operation of the target object, display second content information in the secondary input area when the input focus is on the secondary input area, wherein the second content information includes the title information and the information input by the second input operation; The display unit is configured to display target comment information in response to the comment end operation of the target object, wherein the target comment information includes the first content information and the second content information; The review editing interface also includes a rating area, which displays a rating selector. The device also includes: The selection unit is configured to, in response to a selection operation by a target object on the rating selector, display first rating information for the target product in the rating area; wherein the title information is generated based on the first rating information; The updating unit is configured to determine the second rating information of the target object for the target product based on the first content information and the second content information; and update the title information displayed in at least one unedited sub-input area, or add a sub-input area displaying new title information, based on the second rating information.

12. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the comment information processing method according to any one of claims 1 to 10.

13. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method for processing comment information as described in any one of claims 1 to 10.

14. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method for processing comment information as described in any one of claims 1 to 10.

Images