Information processing method and system, electronic equipment, storage medium and program product

By extracting and generating summary comment content, the problem of users understanding a large number of and messy comments is solved, and fast and convenient comment understanding is achieved, and user experience is improved.

CN118940750APending Publication Date: 2024-11-12MIGU INTERACTIVE ENTERTAINMENT CO LTD +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411083236.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The prior art is difficult to effectively solve the problem of users understanding a large number of and messy comment content, which makes it take users a long time to read and understand comments.

Method used

By obtaining comment information, extracting the comment content corresponding to the preset keywords, and generating the second comment content when the keyword content is empty, using the artificial intelligence content generation model (AIGC) to generate summary comments, reducing the time for users to understand comments and difficulty.

Benefits of technology

This enables users to understand comments without reading the full text, significantly reduces the length of comments comprehension, improves user experience, and simplifies the understanding process of comment content.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940750A_ABST
    Figure CN118940750A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of information processing, and provides an information processing method and system, electronic equipment, a storage medium and a program product, and the method comprises the steps: obtaining comment information; based on the comment information, obtaining first comment contents corresponding to the N preset keywords; and under the condition that the first comment content corresponding to the N preset keywords is empty, generating second comment content corresponding to each keyword in the N preset keywords based on the comment information. Therefore, even if the original comment information is disordered and has more contents, the user can realize comment understanding by understanding the second comment contents corresponding to the N preset keywords obtained by understanding the original comment information without performing full-text reading understanding on the original comment information, so that the comment understanding time can be shortened, and the user can understand the comment conveniently.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of information processing technology, and in particular to an information processing method, system, electronic device, storage medium and program product. Background Art

[0002] With the development of Internet technology, users can comment on the content posted by others or themselves at any time and anywhere (i.e., subjective or objective opinions or views on things, etc.). For the same content, there may be a large number of users commenting. At present, the comments corresponding to any content can be filtered or sorted to meet the viewing needs of users.

[0003] However, there may be messy and long comments, and users need to spend a long time to understand such comments, that is, it is not convenient for users to understand such comments. Summary of the invention

[0004] The embodiments of the present application provide an information processing method, system, electronic device, storage medium and program product to solve the existing problem that it is inconvenient for users to understand comments.

[0005] In order to solve the above technical problems, this application is implemented as follows:

[0006] In a first aspect, an embodiment of the present application provides an information processing method, which is applied to a server, including:

[0007] Get comment information;

[0008] Based on the comment information, obtain first comment content corresponding to N preset keywords;

[0009] In the case that the first comment content corresponding to the N preset keywords is empty, the second comment content corresponding to each keyword in the N preset keywords is generated based on the comment information, where N is a positive integer.

[0010] In a second aspect, an embodiment of the present application provides an information processing method, which is applied to a terminal, including:

[0011] The second comment content corresponding to each of the N preset keywords sent by the receiving server is the comment content generated based on the comment information when there is empty content in the first comment content corresponding to the N preset keywords obtained based on the comment information.

[0012] In a third aspect, an embodiment of the present application provides an information processing device, which is applied to a server, including:

[0013] The first acquisition module is used to acquire comment information;

[0014] A second acquisition module is used to obtain first comment content corresponding to N preset keywords based on the comment information;

[0015] The content generation module is used to generate second comment content corresponding to each of the N preset keywords based on the comment information when there is no first comment content corresponding to the N preset keywords, where N is a positive integer.

[0016] In a fourth aspect, an embodiment of the present application provides an information processing device, applied to a terminal, including:

[0017] The first receiving module is used to receive second comment content corresponding to each of N preset keywords sent by the server, and the second comment content of the N preset keywords is comment content generated based on the comment information when there is empty content in the first comment content corresponding to the N preset keywords obtained based on the comment information.

[0018] In a fifth aspect, an embodiment of the present application further provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the information processing method described above when executing the computer program.

[0019] In a sixth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the information processing method as described above are implemented.

[0020] In a seventh aspect, an embodiment of the present application further provides a computer program product, comprising a computer program, which implements the steps of the above method when executed by a processor.

[0021] In this embodiment, the first comment content corresponding to N preset keywords can be extracted based on the comment information. When the first comment content corresponding to the N preset keywords is empty, it means that there is a keyword with no result among the N preset keywords, that is, the comment result corresponding to the keyword is not extracted in the comment information. In this case, the second comment content corresponding to each keyword of the N preset keywords can be generated based on the comment information, that is, the comment content corresponding to the N preset keywords in the comment information is obtained. In this way, even if the original comment information is messy and has a lot of content (that is, it is long), there is no need to read and understand the entire original comment information. Users can understand the comments by understanding the second comment content corresponding to the N preset keywords, which can reduce the time for understanding the comments and facilitate user understanding. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0023] Figure 1 It is one of the flow charts of the information processing method provided in the embodiment of the present application;

[0024] Figure 2 This is the second flowchart of the information processing method provided in the embodiment of the present application;

[0025] Figure 3 is a schematic diagram of an information processing method provided in an embodiment of the present application;

[0026] Figure 4 This is one of the application scenario diagrams of the information processing method provided in the embodiment of the present application;

[0027] Figure 5 This is the second application scenario diagram of the information processing method provided in the embodiment of the present application;

[0028] Figure 6 This is the third application scenario diagram of the information processing method provided in the embodiment of the present application;

[0029] Figure 7 This is the third flowchart of an information processing method provided in an embodiment of the present application;

[0030] Figure 8 This is the fourth application scenario diagram of the information processing method provided in the embodiment of the present application;

[0031] Fig. 9 It is one of the module schematic diagrams of the information processing device provided in the embodiment of the present application;

[0032] Fig.10 This is the second module schematic diagram of the information processing device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0034] See also Figure 1 , Figure 1is a flowchart of the information processing method provided in the embodiment of the present application, which can be applied to the server, such as Figure 1 As shown, the following steps are included:

[0035] Step 101: Obtain comment information;

[0036] Users can comment on any object with comment permission (for example, including but not limited to news, videos, text content, audio, products, commodities, etc.). In this embodiment, comment information of the target object can be obtained. The target object can be any object, for example, it can be an object published in a target application, and the target application can be any application, for example, including but not limited to news applications, video applications, social applications, audio applications, shopping applications, etc.

[0037] Step 102: Based on the comment information, obtain first comment content corresponding to N preset keywords.

[0038] It can be understood that the comment content (also called comment field) corresponding to the preset keyword in the comment information is obtained. It should be noted that there are many ways to obtain the first comment content corresponding to the preset keyword. This embodiment does not limit the way to obtain the first comment content corresponding to the preset keyword. For example, based on the comment information, the first comment content corresponding to N preset keywords can be obtained through a model, and based on the comment information, the first comment content corresponding to N preset keywords can be obtained through character matching, etc.

[0039] Step 103: when there is no first comment content corresponding to the N preset keywords, generate second comment content corresponding to each keyword in the N preset keywords based on the comment information, where N is a positive integer.

[0040] There may be many reasons why the first comment content of a keyword is empty. For example, some comments corresponding to some keywords may not exist in the comment information, so the first comment content of these keywords is empty. It is also possible that the comment content corresponding to some keywords cannot be extracted due to the way of obtaining the first comment content, which results in the first comment content being less accurate. In this embodiment, when the first comment content corresponding to N preset keywords is empty, the second comment content corresponding to each keyword in the N preset keywords can be further generated based on the comment information. It should be understood that the second comment content corresponding to each keyword in the N preset keywords is not empty. In addition, the first comment content corresponding to the N preset keywords may be empty if at least one of the N preset keywords has an empty first comment content. In addition, it should be noted that when the first comment content corresponding to the N preset keywords does not exist in a blank, there is no need to further generate the second comment content to reduce the amount of calculation. In this case, the first comment content of the N preset keywords can be sent to the terminal so that the terminal displays the first comment content of the N preset keywords. In this way, the user views the first comment content of the N preset keywords and can understand the comment information by understanding the first comment content of the N preset keywords, which can reduce the time for understanding the comments and facilitate user understanding.

[0041] In this embodiment, the first comment content corresponding to N preset keywords can be extracted based on the comment information. When the first comment content corresponding to the N preset keywords is empty, it means that there is a keyword with no result among the N preset keywords, that is, the comment result corresponding to the keyword is not extracted in the comment information. In this case, the second comment content corresponding to each keyword of the N preset keywords can be generated based on the comment information, that is, the comment content corresponding to the N preset keywords in the comment information is obtained. In this way, even if the original comment information is messy and has a lot of content (that is, it is long), there is no need to read and understand the entire original comment information. Users can understand the comments by understanding the second comment content corresponding to the N preset keywords, which can reduce the time for understanding the comments and facilitate user understanding.

[0042] In some embodiments, the method further comprises:

[0043] Receiving a comment query request sent by a terminal, wherein the comment query request includes a comment template identifier, a target object identifier, and a comment identifier list;

[0044] The obtaining, based on the comment information, first comment contents corresponding to N preset keywords, includes: obtaining, based on the comment information, first comment contents corresponding to N preset keywords when the comment template identifier is located in a preset template identifier list;

[0045] The comment information is a comment associated with the target object corresponding to the target object identifier, the comment information includes comments corresponding to the comment identifier in the comment identifier list, and the N preset keywords are keywords in the comment template corresponding to the comment template identifier.

[0046] That is, when a user views comments through a terminal, the terminal may send a comment query request to the server, and the N preset keywords may be keywords corresponding to the comment template (also referred to as comment format or comment format template or comment display format template, etc.) corresponding to the comment template identifier in the comment query request. The target object may be understood as the comment information corresponding to the target object specified in the comment query request, for example, the target object may include but is not limited to news, video, text content, audio, product, commodity, etc.

[0047] The preset template identification list may include at least one template identification, and the comment template corresponding to the template identification in the list may be understood as the identification of a custom template, and may include the identification of at least one custom comment template set by the user. The terminal may pre-send at least one template pre-set by the user to the server, and the server stores it, and the server maintains the identification of each template. After receiving the comment query request sent by the terminal, it may be searched in the preset template identification list. In the case where the preset template identification list includes the comment template identification in the query request message (i.e., the comment template identification is located in the preset template identification list), the first comment content corresponding to the keyword in the comment template corresponding to the comment template identification may be obtained based on the comment information. For example, the comment template corresponding to the comment template identification is an electronic product format template, and its keywords include model, performance, and cost performance. Based on the comment information, the first comment content corresponding to these keywords may be obtained, that is, the content corresponding to the model, the content corresponding to the performance, and the content corresponding to the cost performance may be obtained. For example, the content corresponding to the model is S****, and the content corresponding to the performance is: supporting lifting and rotation functions, with wire storage on the back, indirect and generous, but high power consumption, and the content corresponding to the cost performance is empty, and the price and cost performance information are not mentioned.

[0048] It should be noted that, when the comment template identifier is not included in the preset template identifier list (that is, the comment template identifier is not in the preset template identifier list), it can mean that the user does not need to use the template in the preset template representation list to display the comment, but can simply sample the original comment information for display, that is, send the above comment information to the terminal, and after the terminal receives the comment information, it can display the comment information, and the user can view the comment information associated with the target object.

[0049] In this implementation, based on the comment query request sent by the terminal, the comment information associated with the target object corresponding to the target object identifier in the comment query request can be obtained, and when the comment template identifier is in the preset template identifier list, the first comment content corresponding to the N preset keywords of the comment template corresponding to the comment template identifier in the comment query request can be obtained to meet the user's query requirements.

[0050] In some embodiments, based on the comment information, obtaining first comment content corresponding to N preset keywords includes: based on the comment information, obtaining first comment content corresponding to N preset keywords through a preprocessing model;

[0051] Based on the comment information, generate second comment content corresponding to each of the N preset keywords, including: based on the comment information, generate second comment content corresponding to each of the N preset keywords through an artificial intelligence content generation (AIGC) model.

[0052] That is, considering the large amount of comment information, relying on the AIGC model to generate a large amount of summary comment content has high performance requirements, and serial processing takes a long time. Therefore, the preprocessing model can be used as a predecessor model of the AIGC model, that is, the preprocessing model is used first to obtain the first comment content of the keyword. If the first comment content of the N preset keywords obtained is empty, the AIGC model is used to generate the second comment content corresponding to the N preset keywords to improve the efficiency of comment content generation.

[0053] In some embodiments, the method further comprises:

[0054] The preprocessing model is updated based on the comment information, the N preset keywords and second comment content of the N preset keywords.

[0055] That is, the preprocessing model can be retrained based on the comment information, the N preset keywords, and the second comment content with N preset keywords generated by the AIGC model to update the preprocessing model and improve the performance of the preprocessing model so that the accuracy of the first comment content can be obtained subsequently through the preprocessing model.

[0056] In some embodiments, the method further comprises:

[0057] Get sample reviews;

[0058] For the sample comments, generate third comment content of M sample keywords through the AIGC model, where M is a positive integer;

[0059] Model training is performed based on the sample comments, the M sample keywords, and third comment contents of the M sample keywords of the sample comments to obtain the preprocessing model.

[0060] It should be noted that the number of sample comments can be one or more. If there are more than one, then for each sample comment, the third comment content of M sample keywords is generated by the AIGC model, that is, each sample comment has corresponding third comment content of M sample keywords. In this embodiment, before obtaining the first comment content corresponding to N preset keywords through the preprocessing model, model training can be performed in advance to obtain the preprocessing model. Specifically, the sample comments, M sample keywords, and the third comment content of M sample keywords generated by the AIGC model can be used to perform model training to obtain the preprocessing model, so as to improve the accuracy of the first comment content obtained by the trained preprocessing model later.

[0061] In some embodiments, the method further comprises:

[0062] The second comment content of the N preset keywords is sent to the terminal, so that the terminal displays the second comment content of the N keywords.

[0063] That is, after generating the second comment content corresponding to each of the N preset keywords, the second comment content of the N preset keywords can be sent to the terminal, so that the terminal can display the second comment content of the N keywords for users to view through the terminal.

[0064] In some embodiments, the method further comprises:

[0065] Determine a matching parameter value between each keyword in the N preset keywords and the corresponding second comment content, wherein the matching parameter value is used to represent the matching degree between the keyword and the corresponding second comment content;

[0066] When the matching parameter values ​​between the keywords in the N preset keywords and the corresponding second comment content are all less than the preset threshold, a prompt message is sent to the terminal, where the prompt message is used to prompt that there is no matching keyword content.

[0067] In this embodiment, after generating the second comment content corresponding to each of the N preset keywords, the matching parameter value between each of the N preset keywords and the corresponding second comment content can also be determined. When the matching parameter value between the keyword and the corresponding second comment content is less than the preset threshold, it means that the degree of match between the keyword and the corresponding second comment content is small, or the second comment content corresponding to the keyword has no obvious intersection with the keyword. To avoid the loss of other content, comment information can be sent to the terminal, and the terminal can receive the comment information and display it, that is, the terminal will display the comment information as it is. In addition, a prompt message can be sent to the terminal, and the prompt message can prompt that there is no matching keyword content, and can also prompt suggested keywords. The suggested keywords can be keywords determined based on the comment information.

[0068] In some embodiments, sending the second comment content of the N preset keywords to the terminal may include: when the N preset keywords include a target keyword whose matching parameter value with the corresponding second comment content is greater than or equal to a preset threshold, sending the second comment content of the N preset keywords to the terminal. When the matching parameter value between the keyword and the corresponding second comment content is greater than or equal to the preset threshold, it means that the degree of matching between the keyword and the corresponding second comment content is relatively large. In order to reduce the difficulty of user understanding, the second comment content of the N preset keywords is sent to the terminal, so that the user can view the second comment content of the N preset keywords through the terminal to understand the comment information.

[0069] See also Figure 2 , Figure 2 is a flowchart of an information processing method provided in an embodiment of the present application, which can be applied to a terminal, such as Figure 2 As shown, the following steps are included:

[0070] Step 201: Receive second comment content corresponding to each of N preset keywords sent by the server, where the second comment content of the N preset keywords is comment content generated based on comment information when there is empty content in the first comment content corresponding to the N preset keywords obtained based on the comment information.

[0071] In some embodiments, the method further comprises:

[0072] A comment query request is sent to the server, wherein the comment query request includes a comment template identifier, a target object identifier and a comment identifier list, the comment information is a comment associated with a target object corresponding to the target object identifier, the comment information includes a comment corresponding to a comment identifier in the comment identifier list, and the N preset keywords are keywords in a comment template corresponding to the comment template identifier.

[0073] It should be noted that the method applied to the terminal is a method applied on a different side from the method applied to the server, and the technical features correspond one to one, which will not be repeated here.

[0074] The process of the above method is described in detail below with reference to some specific embodiments.

[0075] Description of related technologies:

[0076] In the related technology, the comment content is identified and the comment filtering and sorting capabilities are provided, but the comment content will not be modified or changed. Although this method can filter out the comments that users want to view to a certain extent, considering the limited filtering conditions, it is also possible to filter out some comments with important viewpoints. It is of little help in solving the problem of how to extract the key content of the comments and display them in a suitable format. For some long comments, users need to spend a long time reading and understanding, which is not convenient for users to understand such comments, resulting in user reading fatigue, and thus failing to reflect the true effect of the comments.

[0077] The present application embodiment proposes a method for processing information, taking the server end including a server (application (APP) server) and an AI server as an example, the principle is as follows: Figure 3 shown.

[0078] The user sets the comment display format (i.e. comment format template or comment display format template):

[0079] like Figure 4 As shown, the user end (terminal) accesses the APP comment format setting page to edit the custom comment format template. The page format is as follows Figure 4 As shown: First edit the keyword information of interest, such as quality, cost performance, place of use, how far it can run, defects, models, performance, etc. Then edit the review format template, such as Figure 4 As shown, for example, add a review format template for electronic products, select the added keyword information, and edit the text to form the final review format template for saving. You can use the radio button on this page to select the default review format template. The user end sends the set review display format to the server on the server end, and the server interface service receives the review display format and stores it.

[0080] Comment list page select format:

[0081] like Figure 5 As shown in the figure, take the online shopping app as an example, enter the product details page, select the evaluation tag, and the original comment information is displayed by default. The comment information contains a lot of content, and the key content can be extracted. Select the comment display format template you defined, and it will automatically refresh and display in the form of template specifications, such as Figure 6As shown, the keywords in the review format template for electronic products may include model, performance, and cost performance, etc. Figure 6 The comment information is displayed in the form of keywords and corresponding comment content.

[0082] AI review content generation:

[0083] When the user selects a custom comment format template, the APP sends the template ID, product ID, and comment ID list to the server. The server queries the specific content of the template and the specific content of the comment, and combines them into the following prompt words:

[0084] According to the following text: "It was really a good experience when I bought a monitor for the first time. I mainly use it for drawing. It is really good. This SL1** is slightly inferior to SL2** in resolution support and refresh rate, but the price is half the price. I recommend it.", the following content is summarized: "Model: (), Performance: (), Cost-effectiveness: (). Send the prompt word to the AI ​​server, and use AIGC technology to get the return content as follows:

[0085] Model: SL1**

[0086] Performance: Suitable for drawing, the resolution and refresh rate are slightly lower than SL2** but the performance is still good

[0087] Cost-effectiveness: half the price, recommended"

[0088] The content generated by AIGC can be combined into a message and returned to the APP for display.

[0089] Comments on secondary processing technology:

[0090] like Figure 7As shown, considering that there are a lot of comment data per page, relying on AIGC to generate summary texts in batches has high performance requirements, and the serial processing requires a long waiting time for users. Therefore, in this embodiment, the secondary processing technology of comments can be used to perform secondary training on the generated content. For example, at a preset time every day (for example, early morning), the original text (original comment) and keywords used by the AIGC technology on that day are used as input, and the results of the keywords are used as output. The neural network technology is used for training to generate a preprocessing model (also called a comment preprocessing model or a comment preset model), that is, the current data set is retrained with the preprocessing model. The preprocessing model is used as a pre-model of the AIGC model, that is, the first comment content corresponding to the keyword (the result of the keyword) is obtained first through the preprocessing model. If the first comment content obtained is empty, that is, there is an empty result in the obtained result, that is, there is a keyword but no result, then the second comment content corresponding to the keyword (the result of the keyword) is generated through the AIGC model. In addition, if the user is not satisfied with the comment content obtained by the preprocessing model, it can also be manually triggered to regenerate through the AIGC model.

[0091] Abnormal situation handling:

[0092] When the review content (e.g., the generated second review content) has no obvious intersection with the keywords defined by the user, that is, the matching parameter value is small, in order to avoid the loss of other information, the original review information will be displayed and suggested keywords will be provided;

[0093] For example:

[0094] According to the following text: "Thank you, the shopkeeper, the shopkeeper is very professional and enthusiastic, and he answered all questions very quickly.", the following prompt words are summarized: "Model: (), Performance: (), Cost-effectiveness: (");

[0095] AI server returns data:

[0096] According to the text provided, this text seems to be expressing an evaluation of the service attitude of the merchant or store, and does not mention any product model, performance or cost-effective information, so it is impossible to summarize the corresponding content. You can use the text NLP technology to extract the keyword "merchant or store service attitude" and return it to the APP end (terminal), and the APP end will display it like this Figure 8 As shown, in this example, the server can send a prompt message to the APP to prompt "No matching keywords, it is recommended to add: the service attitude of the merchant or store". The user can add keywords to the template according to the prompt.

[0097] The embodiment of the present application provides a complete set of dynamic comment generation technology, which optimizes the problems of existing messy comment content and reading fatigue. And the AIGC technology is reasonably used to extract targeted keywords from the comment content, without affecting the original content, and standardizing the display format. Moreover, correction suggestions are provided for situations where it cannot be fully matched, which optimizes the user experience. In addition, considering the performance problem of AIGC, a lightweight model (i.e., a preprocessing model) is generated through secondary processing technology to improve generation efficiency and optimize user experience. In other words, the AIGC technology solves the problem of messy display and excessive content of existing comments. And a complete format configuration and fault-tolerant solution are provided to avoid the problem of inability to match. And a two-layer model query method is provided to save computing resources and optimize the user experience. Through the solution of this embodiment, in terms of product experience, it can meet the user's rapid understanding of comments, improve the speed of content understanding, and have high commercial value.

[0098] See also Fig. 9 , Fig. 9 is a schematic diagram of a module of an information processing device 900 provided in an embodiment of the present application, which can be applied to a server, such as Fig. 9 As shown, including:

[0099] The first acquisition module 901 is used to acquire comment information;

[0100] A second acquisition module 902 is used to obtain first comment content corresponding to N preset keywords based on the comment information;

[0101] The content generation module 903 is used to generate second comment content corresponding to each of the N preset keywords based on the comment information when there is no first comment content corresponding to the N preset keywords, where N is a positive integer.

[0102] In some embodiments, the apparatus 900 further includes:

[0103] A second receiving module, configured to receive a comment query request sent by a terminal, wherein the comment query request includes a comment template identifier, a target object identifier, and a comment identifier list;

[0104] The obtaining, based on the comment information, first comment contents corresponding to N preset keywords, includes: obtaining, based on the comment information, first comment contents corresponding to N preset keywords when the comment template identifier is located in a preset template identifier list;

[0105] The comment information is a comment associated with the target object corresponding to the target object identifier, the comment information includes comments corresponding to the comment identifier in the comment identifier list, and the N preset keywords are keywords in the comment template corresponding to the comment template identifier.

[0106] In some embodiments, based on the comment information, obtaining first comment content corresponding to N preset keywords includes: based on the comment information, obtaining first comment content corresponding to N preset keywords through a preprocessing model;

[0107] Based on the comment information, generate the second comment content corresponding to each of the N preset keywords, including: based on the comment information, generate the second comment content corresponding to each of the N preset keywords through an artificial intelligence content generation AIGC model.

[0108] In some embodiments, the apparatus further comprises:

[0109] An updating module is used to update the preprocessing model based on the comment information, the N preset keywords and the second comment content of the N preset keywords.

[0110] In some embodiments, the apparatus further comprises:

[0111] The third acquisition module is used to obtain sample comments;

[0112] A content generation module, configured to generate third comment content of M sample keywords for the sample comment by using the AIGC model;

[0113] A training module is used to perform model training based on the sample comments, the M sample keywords and third comment contents of the M sample keywords of the sample comments to obtain the preprocessing model.

[0114] In some embodiments, the apparatus further comprises:

[0115] The first sending module is used to send the second comment content of the N preset keywords to the terminal, so that the terminal displays the second comment content of the N keywords.

[0116] In some embodiments, the apparatus further comprises:

[0117] A matching parameter determination module, used to determine a matching parameter value between each of the N preset keywords and the corresponding second comment content, wherein the matching parameter value is used to represent the degree of matching between the keyword and the corresponding second comment content;

[0118] The second sending module is used to send a prompt message to the terminal when the matching parameter values ​​between the keywords in the N preset keywords and the corresponding second comment content are all less than a preset threshold, and the prompt message is used to prompt that there is no matching keyword content.

[0119] The technical features of the above-mentioned information processing device 900 correspond to the technical features of the above-mentioned information processing method applied to the server. The various processes of the above-mentioned information processing method are implemented through the information processing device 900, and the same effect can be obtained. To avoid repetition, they will not be repeated here.

[0120] See also Fig.10 , Fig.10 1 is a schematic diagram of a module of an information processing device 1000 provided in an embodiment of the present application, which can be applied to a terminal, such as Fig.10 As shown, including:

[0121] The first receiving module 1001 is used to receive the second comment content corresponding to each of the N preset keywords sent by the server, and the second comment content of the N preset keywords is the comment content generated based on the comment information when there is empty content in the first comment content corresponding to the N preset keywords obtained based on the comment information.

[0122] In some embodiments, the apparatus further comprises:

[0123] The third sending module is used to send a comment query request to the server, wherein the comment query request includes a comment template identifier, a target object identifier and a comment identifier list, the comment information is a comment associated with the target object corresponding to the target object identifier, the comment information includes a comment corresponding to the comment identifier in the comment identifier list, and the N preset keywords are keywords in the comment template corresponding to the comment template identifier.

[0124] The technical features of the above-mentioned information processing device 1000 correspond to the technical features of the above-mentioned information processing method applied to the terminal. The various processes of the above-mentioned information processing method are implemented through the information processing device 1000, and the same effect can be obtained. To avoid repetition, they will not be repeated here.

[0125] An embodiment of the present application also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program that can be run on the processor. When the computer program is executed by the processor, the various processes in the above method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0126] The embodiment of the present application also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the various processes of the above-mentioned method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0127] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0128] The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0129] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0130] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for an electronic device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0131] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. An information processing method, characterized in that: Applied to the server, including: Get comment information; Based on the comment information, obtain first comment content corresponding to N preset keywords; In the case that the first comment content corresponding to the N preset keywords is empty, the second comment content corresponding to each keyword in the N preset keywords is generated based on the comment information, where N is a positive integer.

2. The method according to claim 1, characterized in that The method further comprises: Receiving a comment query request sent by a terminal, wherein the comment query request includes a comment template identifier, a target object identifier, and a comment identifier list; The obtaining, based on the comment information, first comment contents corresponding to N preset keywords, includes: obtaining, based on the comment information, first comment contents corresponding to N preset keywords when the comment template identifier is located in a preset template identifier list; The comment information is a comment associated with the target object corresponding to the target object identifier, the comment information includes comments corresponding to the comment identifier in the comment identifier list, and the N preset keywords are keywords in the comment template corresponding to the comment template identifier.

3. The method according to claim 1, characterized in that Based on the comment information, obtaining first comment content corresponding to N preset keywords includes: based on the comment information, obtaining first comment content corresponding to N preset keywords through a preprocessing model; Based on the comment information, generate the second comment content corresponding to each of the N preset keywords, including: based on the comment information, generate the second comment content corresponding to each of the N preset keywords through an artificial intelligence content generation AIGC model.

4. The method according to claim 3, characterized in that The method further comprises: The preprocessing model is updated based on the comment information, the N preset keywords and second comment content of the N preset keywords.

5. The method according to claim 3 or 4, characterized in that: The method further comprises: Get sample reviews; For the sample comments, generate third comment content of M sample keywords through the AIGC model, where M is a positive integer; Model training is performed based on the sample comments, the M sample keywords, and third comment contents of the M sample keywords of the sample comments to obtain the preprocessing model.

6. The method according to claim 1, characterized in that The method further comprises: The second comment content of the N preset keywords is sent to the terminal, so that the terminal displays the second comment content of the N keywords.

7. The method according to claim 1, characterized in that The method further comprises: Determine a matching parameter value between each keyword in the N preset keywords and the corresponding second comment content, wherein the matching parameter value is used to represent the matching degree between the keyword and the corresponding second comment content; When the matching parameter values ​​between the keywords in the N preset keywords and the corresponding second comment content are all less than the preset threshold, a prompt message is sent to the terminal, where the prompt message is used to prompt that there is no matching keyword content.

8. An information processing method, characterized in that: Applied to terminals, including: The second comment content corresponding to each of the N preset keywords sent by the receiving server is the comment content generated based on the comment information when there is empty content in the first comment content corresponding to the N preset keywords obtained based on the comment information.

9. The method according to claim 8, characterized in that The method further comprises: A comment query request is sent to the server, wherein the comment query request includes a comment template identifier, a target object identifier and a comment identifier list, the comment information is a comment associated with a target object corresponding to the target object identifier, the comment information includes a comment corresponding to a comment identifier in the comment identifier list, and the N preset keywords are keywords in a comment template corresponding to the comment template identifier.

10. An electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the processor implements the steps of the information processing method described in any one of claims 1 to 7, or implements the steps of the information processing method described in any one of claims 8 to 9.

11. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the information processing method described in any one of claims 1 to 7 are implemented, or the steps of the information processing method described in any one of claims 8 to 9 are implemented.

12. A computer program product, comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 9.