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

Abstract

This application provides an information processing method, apparatus, electronic device, and storage medium, belonging to the field of information processing technology, and is used to solve the problems of low information processing efficiency and low accuracy in related technologies. The method includes: acquiring comment information displayed to the current user; performing semantic analysis on the comment information to determine the commented object; and outputting response information associated with the commented object.

Description

Technical Field

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

[0002] With the development of the internet economy, more and more merchants and users are accustomed to conducting online transactions through platforms, such as buying and selling goods. During the actual transaction process, users can provide feedback on the transaction through the review section, for example, leaving comments about points of dissatisfaction or satisfaction. Merchants and platforms generally employ two methods for handling user comments: one is to use a fixed response for comments that can be replied to with standardized content; the other is to manually reply to each comment individually if a standardized response is not possible.

[0003] However, the first technical method mentioned above, because it uses fixed response content for a uniform reply, fails to consider the specific dissatisfaction users have with the comments, resulting in low accuracy and an inability to substantively resolve user issues. The second technical method, which relies on manual, one-by-one responses, suffers from high latency and excessive manpower investment due to the inherent limitations of human interaction. Consequently, users experience a negative experience due to the inability to address their issues promptly, while businesses or platforms face the disadvantage of low economic efficiency due to the significant investment of human resources. Summary of the Invention

[0004] To address the aforementioned problems, this application provides an information processing method, apparatus, electronic device, and storage medium. The information processing method provided by this application aims to provide highly accurate and effective solutions to user-generated questions, while simultaneously improving the efficiency of problem-solving.

[0005] According to a first aspect of the present disclosure, an information processing method is provided, the method comprising:

[0006] Get the comment information displayed to the current user;

[0007] Perform semantic analysis on the comment information to determine the object being commented on;

[0008] Output the response information associated with the commented object.

[0009] Optionally, the method further includes:

[0010] Perform semantic analysis on the comment information to determine its sentiment tendency;

[0011] Output response information associated with the commented object, including:

[0012] If the emotional tendency is positive, output the resources and / or information of the object being commented on;

[0013] If the emotional tendency is negative, output resources and / or information of other objects similar to the object being commented on.

[0014] Optionally, if the comment information is that no comment has been posted, the method further includes:

[0015] Perform semantic analysis on the uncommented information to determine its sentiment tendency;

[0016] Output response information associated with the commented object, including:

[0017] Detect requests to post comments on information for which no comments have been posted;

[0018] In response to the posting request, output response information associated with the commented object to guide the sentiment of the uncommented information to switch to a positive sentiment.

[0019] Optionally, output resources and / or information about other objects similar to the object being commented on, including:

[0020] The comment information is analyzed to determine the current user's intent;

[0021] If the current user's intent is to obfuscate the currently commented object, the system queries the historical order database for similar attribute objects within a preset geographical range of the currently commented object that have an attribute similarity greater than a preset similarity.

[0022] If multiple similar attribute objects exist within a preset geographical range of the currently commented object, then based on the comment information of each of the multiple similar attribute objects, determine the similar attribute target object whose positive review score is greater than the preset positive review score from among the multiple similar attribute objects;

[0023] Output the resources and / or information of the target object with the similar attributes.

[0024] Optionally, the method further includes:

[0025] If no similar attribute objects exist within the preset geographical range of the currently commented object, the object database is queried to find objects of the same category as the currently commented object within the preset geographical range of the currently commented object;

[0026] If there are multiple objects of the same category within the preset geographical range of the object being commented on, then, based on the comment information of each of the multiple objects of the same category, determine the target object of the same category whose positive review score is greater than the preset positive review score from among the multiple objects of the same category;

[0027] Output resources and / or information about other objects similar to the object being commented on, including:

[0028] Output the resources and / or information of the target object of the same category.

[0029] Optionally, the method further includes:

[0030] The comment information is analyzed to determine the current user's intent;

[0031] If the current user's intention is to increase resource allocation, the resources of the currently commented object, the e-commerce platform where the currently commented object is located, or the delivery platform associated with the currently commented object are sent to the current user's account.

[0032] Output the resources and / or information of the object being commented on, including:

[0033] Output a resource notification message indicating that resources have been allocated to the current user's account.

[0034] Optionally, the method further includes:

[0035] The comment information is analyzed to determine the current user's intent;

[0036] Based on the current user's intent, determine the problem to be solved;

[0037] Match the problem to be solved with a preset solution database and output the solution corresponding to the problem to be solved; and / or

[0038] The problem to be solved is input into a pre-trained solution prediction model to obtain and output the solution corresponding to the problem to be solved.

[0039] Optionally, the method further includes:

[0040] The operation of obtaining the current user's posted comments on other users;

[0041] Analyze the current user's actions to determine whether the current user's sentiment is the same as the sentiment of the comments posted by other users;

[0042] If the current user's sentiment is the same as the sentiment of comments posted by other users, the intent of the other users will be determined as the intent of the current user.

[0043] Output response information associated with the commented object, including:

[0044] In response to the current user's action, output response information associated with the object being commented on.

[0045] Optionally, semantic analysis is performed on the comment information to determine the object being commented on, including:

[0046] Perform semantic analysis on the comment information to obtain the paid orders associated with the comment information;

[0047] The object providing the service in the paid order is identified as the object to be commented on.

[0048] According to a second aspect of the present disclosure, an information processing apparatus is provided, the apparatus comprising:

[0049] The acquisition module is used to retrieve comment information displayed to the current user.

[0050] The first determining module is used to perform semantic analysis on the comment information to determine the object being commented on;

[0051] The output module is used to output response information associated with the commented object.

[0052] According to a third aspect of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the information processing method described in the first aspect above.

[0053] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the information processing method described in the first aspect.

[0054] The information processing method disclosed in this application involves acquiring comment information displayed to the current user, performing semantic analysis on the comment information to determine the commented object, and outputting response information associated with the commented object. This addresses the problems of low information processing efficiency and low accuracy in related technologies. Compared to the prior art, the information processing method provided in this application also has at least the following advantages:

[0055] In the information processing method provided in this application embodiment, since the comment information is the comment information displayed to the current user, the comment object determined based on the comment information can ensure that the output response information associated with the comment object has high accuracy, thereby effectively solving the current user's problem.

[0056] In the information processing method provided in this application embodiment, response information associated with the identified commented object is automatically searched and output. Since neither the determination nor the output process involves human operation, it eliminates the need for manual guessing of the user's current purpose and manual suggestions for response solutions. It also avoids the drawbacks of low accuracy, excessive time consumption, and inability to promptly resolve user problems caused by manual intervention. In other words, the information processing method provided in this application embodiment can improve the efficiency of resolving user problems while achieving high accuracy, thus optimizing the user experience.

[0057] Furthermore, for merchants or platforms, the information processing method provided in this application embodiment can solve user problems in a timely and highly accurate manner. Therefore, it can not only enable merchants or platforms to obtain user feedback information in a timely manner and improve the speed of problem handling, but also improve users' evaluation of merchants or platforms. It can also solve the drawbacks of high economic investment caused by manual processing for merchants or platforms. Attached Figure Description

[0058] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart illustrating an information processing method according to an embodiment of this application;

[0060] Figure 2 This application illustrates a service diagram based on an embodiment of intent-based output response information.

[0061] Figure 3 This is a schematic diagram of the structure of an information processing device according to an embodiment of this application. Detailed Implementation

[0062] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0063] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0064] As online transactions increase, users leave comments about their experiences with some transactions. Merchants and platforms generally use two methods to handle these comments: one is to provide a standardized response to comments that can be replied to with fixed content; the other is to manually reply to each comment individually, as fixed content is not sufficient.

[0065] However, the first technical method mentioned above, because it uses fixed response content for a uniform reply, fails to consider the specific dissatisfaction users have with the comments, resulting in low accuracy and an inability to substantively resolve user issues. The second technical method, which relies on manual, one-by-one responses, suffers from high latency and excessive manpower investment due to the inherent limitations of human interaction. Consequently, users experience a negative experience due to the inability to address their issues promptly, while businesses or platforms face the disadvantage of low economic efficiency due to the significant investment of human resources.

[0066] To address the aforementioned technical problems, the main technical concept proposed in this application is: to analyze the user's intent based on comment information, and then provide the user with a corresponding processing solution based on that intent. Based on this concept, this application provides an information processing method, apparatus, electronic device, and storage medium. The specific details are as follows:

[0067] In a first aspect of this embodiment, to address the problem of low accuracy and efficiency in information processing mentioned above, this application provides an information processing method, such as... Figure 1As shown, the information processing method of this application includes the following steps:

[0068] S1: Retrieve the comment information to be displayed to the current user.

[0069] In practice, after merchants and users conduct online transactions on a platform, users typically provide feedback on their experiences during the transaction. This feedback might include comments about the quality of goods provided by the merchant or the delivery service offered by the platform. These user comments are crucial for every platform and merchant, serving as an important means to identify problems and improve the user experience, and also providing significant guidance for resolving user issues. For example, these comments can help identify the subject of the comment and then, based on this identification, targeted response information can be delivered to the current user that is relevant to that subject.

[0070] Therefore, in specific implementation, this embodiment first obtains the comment information displayed to the current user (the comment information can be the comment information that the current user is browsing or editing), so as to accurately locate the comment object that the current user intends to refer to, such as whether the current user intends to refer to the platform, the merchant, or the product.

[0071] S2: Perform semantic analysis on the comment information to determine the object being commented on.

[0072] In practice, after obtaining the comment information in step S1, keywords are extracted from the comment information, and then semantic analysis is performed based on the extracted keywords to determine the object being commented on. For example, for a comment such as... Figure 2 The keywords extracted from the comments shown include: milk tea A, counterfeit, unpalatable, and throw away. When "milk tea A" and "counterfeit" are extracted, these two keywords can be directly integrated to obtain the semantic information "milk tea A is counterfeit," and the object of the comments corresponding to this semantic information is the merchant "milk tea brand A."

[0073] S3: Output the response information associated with the commented object.

[0074] In practice, after identifying the object being commented on in the previous step, the system automatically performs a related search based on that object. The search results are then output and displayed to the current user as response information. Continuing with the example from step S2 above, for instance, if the object being commented on by the current user is determined to be the merchant "Milk Tea Brand A," then "Milk Tea Brand A" is used as the search sample. The search results can be coupon information from that merchant, or information and corresponding links from other "Milk Tea Brand A" merchants. This information is then output and displayed to the current user (e.g., ...). Figure 2 (As shown).

[0075] The information processing method disclosed in this application involves acquiring comment information displayed to the current user, performing semantic analysis on the comment information to determine the commented object, and outputting response information associated with the commented object. This addresses the problems of low information processing efficiency and low accuracy in related technologies. Compared to the prior art, the information processing method provided in this application also has at least the following advantages:

[0076] In the information processing method provided in this application embodiment, since the comment information is the comment information displayed to the current user, the comment object determined based on the comment information can ensure that the output response information associated with the comment object has high accuracy, thereby effectively solving the current user's problem.

[0077] In the information processing method provided in this application embodiment, response information associated with the identified commented object is automatically searched and output. Since neither the determination nor the output process involves human operation, it eliminates the need for manual guessing of the user's current purpose and manual suggestions for response solutions. It also avoids the drawbacks of low accuracy, excessive time consumption, and inability to promptly resolve user problems caused by manual intervention. In other words, the information processing method provided in this application embodiment can improve the efficiency of resolving user problems while achieving high accuracy, thus optimizing the user experience.

[0078] Furthermore, for merchants or platforms, the information processing method provided in this application embodiment can solve user problems in a timely and highly accurate manner. Therefore, it can not only enable merchants or platforms to obtain user feedback information in a timely manner and improve the speed of problem handling, but also improve users' evaluation of merchants or platforms. It can also solve the drawbacks of high economic investment caused by manual processing for merchants or platforms.

[0079] Optionally, in another embodiment, the method provided by the present invention further includes:

[0080] Semantic analysis is performed on the comment information to determine its sentiment tendency.

[0081] In practice, semantic analysis is performed on the content of the comments information acquired in real time, and then the sentiment tendency of the comments information is determined based on the analyzed semantics.

[0082] In semantic analysis, keywords in the comment information can be used for analysis. Specifically, keywords are extracted from the real-time acquired comment information, and then semantic analysis is performed based on the extracted keywords. The sentiment tendency of the comment information is then determined based on the analyzed semantics. For example, when a comment is being written stating "This milk tea brand A is a counterfeit," the keywords "milk tea brand A" and "counterfeit" can be extracted in real time. These two keywords are then integrated in real time to obtain "milk tea brand A is a counterfeit." Analysis based on the real-time integration result yields the semantic meaning: this shop is a pirated store, selling counterfeit goods, and the corresponding sentiment tendency is negative.

[0083] Keywords can include, but are not limited to, attribute words and negative words.

[0084] Among these, attribute words can be: words representing the name of the product (such as shredded pork with garlic sauce, milk tea brand A, personal care brand B, etc.), words indicating whether the product is genuine (such as counterfeit, fake, pirated, etc.), words indicating whether the price is favorable (such as discount, cheap, coupon, package, etc.), and words indicating services (such as service attitude, home delivery, in-store pickup, and other services not within the scope of the merchant's or platform's services, etc.). Negative words can be words that contain negative emotions or negative sentiments of the user, such as: no, poor, bad, no service provided, too expensive, etc.

[0085] Correspondingly, in step 3 above, response information associated with the commented object is output, including:

[0086] If the emotional tendency is positive, output the resources and / or information of the object being commented on;

[0087] If the emotional tendency is negative, output resources and / or information of other objects similar to the object being commented on.

[0088] In practice, if the sentiment is positive, it indicates that the user is relatively satisfied with the object being commented on. In this case, outputting resources and / or information about the object being commented on will allow the user to better understand it and improve their experience. If the sentiment is negative, it indicates that the user may be dissatisfied with the object being commented on. In this case, instead of outputting related information about the object being commented on, resources and / or information about other objects similar to the object being commented on will be output. This will help the user quickly and accurately receive relevant resources and / or information about other objects to address their current needs for similar objects.

[0089] In this embodiment, the sentiment tendency of the comments is first determined, and then targeted response information is output based on the specific sentiment tendency type. This allows for the recommendation of the most likely-to-be-adopted information to the current user, thereby achieving a highly accurate and effective solution to the user's problem.

[0090] Optionally, in another embodiment, when the comment information is "no comment posted", the method further includes:

[0091] Semantic analysis is performed on the uncommented information to determine its sentiment tendency.

[0092] In practice, the specific analysis methods used when performing semantic analysis on uncommented information can be the same as those used in the above embodiments, and will not be elaborated upon in this embodiment.

[0093] Accordingly, response information associated with the commented object is output, including:

[0094] Detect requests to post comments on information for which no comments have been posted;

[0095] In response to the posting request, output response information associated with the commented object to guide the sentiment of the uncommented information to switch to a positive sentiment.

[0096] In practice, when the sentiment of uncommented information is analyzed, the system automatically detects the current user's posting request to respond in a timely manner before the current user posts an article. By outputting response information associated with the object being commented on, the system guides the sentiment of uncommented information to switch to a positive sentiment, thereby reversing the current user's sentiment.

[0097] By detecting posting requests, merchants or platforms can intervene before reviews are posted and respond promptly to the review information. This avoids negative reviews caused by misunderstandings, reduces subsequent communication costs between merchants and users, and improves the rating of merchants or platforms, thereby mitigating the impact on orders and revenue to some extent.

[0098] The response information in this embodiment is output when the current user clicks "post" and triggers the posting operation. Since the comment information has been written at this time, the response information output based on this comment information has a higher accuracy rate and can solve the user's problem more effectively and comprehensively.

[0099] To help those skilled in the art better understand these implementation steps, the applicant provides further illustrative examples based on the above instance. For instance, when the analyzed semantics are "This store is a pirated store, selling counterfeit goods," this semantics represents a negative sentiment. Then, to avoid a series of drawbacks caused by the current user posting this comment, upon detecting the user's posting action, the system promptly responds to the posting action by outputting response information to guide the sentiment of the comment information to switch to a positive sentiment (e.g., "Link to the authentic 'Milk Tea Brand A' merchant"). Figure 2 (As shown).

[0100] This embodiment intervenes promptly before users post comments, providing them with positive guidance and response information, such as information that can help them resolve issues encountered during the transaction. Thus, for users, this method can promptly address problems they face during the transaction; for merchants and platforms, this method allows them to promptly receive user feedback and respond accordingly, minimizing any negative impact on them.

[0101] In the information processing method provided in this application embodiment, since the comment information can reflect the current user's emotional tendency corresponding to the comment information, such as a negative emotional tendency towards the current comment object, in this application embodiment, by analyzing the comment information to determine whether the current user's emotional tendency is negative, it can ensure that the output response information corresponding to the comment information is highly accurate, can effectively solve the current user's problem, and switch the current user's emotional tendency to a positive emotional tendency.

[0102] Furthermore, in this embodiment, uncommented information can be either negative or positive. When the uncommented information is negative, the output response information can promptly resolve the user's problem and reduce the adverse effects caused by the negative uncommented information; when the uncommented information is positive, the output response information can further recommend better products or services so that the user can continue to have a better experience and prevent the user from having a bad experience in the future.

[0103] Optionally, in another embodiment, resources and / or information of other objects similar to the commented object are output, including:

[0104] The comment information is analyzed to determine the current user's intent;

[0105] If the current user's intent is to obfuscate the currently commented object, the system queries the historical order database for similar attribute objects within a preset geographical range of the currently commented object that have an attribute similarity greater than a preset similarity.

[0106] If multiple similar attribute objects exist within a preset geographical range of the currently commented object, then based on the comment information of each of the multiple similar attribute objects, determine the similar attribute target object whose positive review score is greater than the preset positive review score from among the multiple similar attribute objects;

[0107] Output the resources and / or information of the target object with the similar attributes.

[0108] In specific implementation, this embodiment involves: analyzing the comment information to determine the current user's intent, for example, if the current user's intent is that the commented object is misleading, i.e., the commented object is a counterfeit product; if the current user's intent is to mislead the commented object, the system queries the historical order database for similar attribute objects within a preset geographical range of the commented object, whose attribute similarity is greater than a preset similarity; if multiple similar attribute objects exist within the preset geographical range of the commented object, the system determines similar attribute target objects with a positive review score greater than a preset positive review score based on the comment information of each of the multiple similar attribute objects; and outputs the resources and / or information of the similar attribute target objects. For example, if the current user's intent is brand misleading, it can be determined that the current user is interested in the brand of the product, and based on the historical orders corresponding to the comment information, the system searches for positive-reviewed merchants of similar misleading brands, and recommends nearby positive-reviewed merchants of the same brand to the current user (e.g., Figure 2 (As shown).

[0109] The method provided in this embodiment can solve the problem of users being misled by similar merchants and placing the wrong order, recommend merchants that users like, and guide users to change merchants, thus solving the problem of users being dissatisfied with the brand.

[0110] Furthermore, in existing technologies, when recommending products or merchants to users, recommendations are often based on the user's historical purchase history, without considering user reviews. For example, if a user buys a product from merchant A on a platform and leaves a negative review stating that the product is bad or unsuitable for them, it indicates that the user is unlikely to buy from merchant A again, and may even avoid buying the same brand from other merchants. However, when the user returns to the platform, existing technologies still prioritize merchant A and recommend similar merchants or products. Therefore, this existing technology, which recommends products or guides purchases based on historical purchase records, is unreasonable and does not align with the user's current needs. The method provided in this embodiment, however, makes real-time recommendations based on the user's current review information, effectively preventing the recommendation of merchants or products that the user dislikes or finds unsuitable, and achieving effective recommendations based on user feedback.

[0111] Among these, attributes can be understood as the brand of the product or the services offered by the merchant.

[0112] Optionally, in another embodiment, the method provided by the present invention further includes:

[0113] If no similar attribute objects exist within the preset geographical range of the currently commented object, the object database is queried to find objects of the same category as the currently commented object within the preset geographical range of the currently commented object;

[0114] If there are multiple objects of the same category within the preset geographical range of the object being commented on, then, based on the comment information of each of the multiple objects of the same category, determine the target object of the same category whose positive review score is greater than the preset positive review score from among the multiple objects of the same category;

[0115] Output resources and / or information about other objects similar to the object being commented on, including:

[0116] Output the resources and / or information of the target object of the same category.

[0117] In specific implementation of this embodiment, the steps are as follows: If no similar attribute objects exist within the preset geographical range of the currently commented object, for example, if no brand similar to "Milk Tea A" (e.g., "Milk Tea Brand A") exists, the system queries the object database for objects of the same category as the currently commented object within the preset geographical range, such as "Milk Tea Brand B" or "Milk Tea Brand C". If multiple objects of the same category exist within the preset geographical range of the currently commented object, based on the review information of each of the multiple objects of the same category, the system determines the target object of the same category whose positive review score is greater than the preset positive review score. The system outputs resources and / or information of other objects similar to the commented object, including: outputting resources and / or information of the target object of the same category, such as outputting relevant information about the store corresponding to "Milk Tea Brand B".

[0118] In this embodiment, when brand confusion occurs and there are no similar brands nearby, other brands in the same category can be recommended. For example, if "milk tea brand A" is confused, and there are no similar brands nearby, then "milk tea brand B" or "milk tea brand C" can be recommended. This can promptly resolve the issue of users being misled by similar merchants and placing incorrect orders, providing users with effective response information.

[0119] Optionally, in another embodiment, the method provided by the present invention further includes:

[0120] The comment information is analyzed to determine the current user's intent;

[0121] If the current user's intention is to increase resource allocation, the resources of the currently commented object, the e-commerce platform where the currently commented object is located, or the delivery platform associated with the currently commented object are sent to the current user's account.

[0122] Output the resources and / or information of the object being commented on, including:

[0123] Output a resource notification message indicating that resources have been allocated to the current user's account.

[0124] In this embodiment, the object being commented on can be a merchant, a platform, or a product offered by a merchant. For example, when the comment is about whether a merchant offers discounts, the object being commented on can be identified as the merchant; when the comment is about evaluating rider service on a platform, the object being commented on can be identified as the platform; and when the comment suspects a brand of counterfeit goods, the object being commented on can be identified as the product.

[0125] In this embodiment, outputting the resources and / or information of the commented object includes: outputting resource notification information that resources have been sent to the current user's account. For example, if the current user's intention is for discount information, it can be determined that the current user is interested in merchant discounts. Therefore, by combining store discount information and platform discount information, coupon notification information corresponding to these discounts can be output to the current user, indicating that the store offers discounts or other discount information, and guiding the current user to use them.

[0126] The method provided in this embodiment can prevent users from giving negative reviews to merchants due to insufficient discount information, while also prompting users to indicate that the store has available discount information, guiding them to continue placing orders on the platform / in-store. Furthermore, in practical applications, the method provided in this embodiment, through the analysis of intent, can categorize intent based on the attributes and / or discounts of the reviewed object, facilitating corresponding processing at the business layer and achieving automated, efficient, and highly accurate handling of the current user's issues.

[0127] Optionally, in another embodiment, the method provided by the present invention further includes:

[0128] The comment information is analyzed to determine the current user's intent;

[0129] Based on the current user's intent, determine the problem to be solved;

[0130] Match the problem to be solved with a preset solution database and output the solution corresponding to the problem to be solved; and / or

[0131] The problem to be solved is input into a pre-trained solution prediction model to obtain and output the solution corresponding to the problem to be solved.

[0132] In practical implementation, if the current user's problem is not related to attributes and / or discounts as mentioned in the previous embodiment, the problem to be solved can be determined based on the current user's intent. For example, if the current user's intent is dissatisfaction with a product (e.g., a dish is not tasty), the problem to be solved can be determined as how to resolve the dissatisfaction with the product; similarly, if the current user's intent is dissatisfaction with a service (e.g., a service is not provided), the problem to be solved can be determined as how to resolve the dissatisfaction with the service. Since such dissatisfaction problems are common and cannot be resolved by directly outputting product or merchant links, in existing technologies, they are often addressed manually one by one. However, due to the inherent limitations of manual work, this results in low efficiency and high investment costs. In this implementation step, to solve these problems, the proposed technical means can be a combination of the following two technical means, or any one of the following two technical means.

[0133] Technical approach A: For these problems, pre-set solutions and store them in a unified solution database. When a specific problem to be solved is determined based on the current user's intent, the problem to be solved can be matched with the pre-set solution database, and the corresponding solution can be quickly output based on the matching result.

[0134] Technical approach B: To address these issues, a solution prediction model is pre-trained. When the specific problem to be solved is determined based on the current user's intent, the problem can be input into the pre-trained solution prediction model, which can then quickly obtain and output the corresponding solution.

[0135] The method provided in this embodiment, because the solution may be pre-set or based on a solution prediction model, can not only solve user problems in a timely and accurate manner, improving user satisfaction, but also eliminates the manpower cost for merchants to individually respond to users, and avoids negative reviews caused by misunderstandings such as differences in understanding between users and merchants. Here, differences in understanding can be understood as: for a certain service, there is a difference in perception where users believe that the service is available, but the merchant or rider cannot actually perform the service.

[0136] Optionally, in another embodiment, the method provided by the present invention further includes:

[0137] The operation of obtaining the current user's posted comments on other users;

[0138] Analyze the current user's actions to determine whether the current user's sentiment is the same as the sentiment of the comments posted by other users;

[0139] If the current user's sentiment is the same as the sentiment of comments posted by other users, the intent of the other users will be determined as the intent of the current user.

[0140] Output response information associated with the commented object, including:

[0141] In response to the current user's action, output response information associated with the object being commented on.

[0142] In this context, "current user" refers to a user who has not posted any comments. The action can be the time spent browsing comments posted by other users. For example, a dwell time exceeding 5 seconds indicates that the current user is carefully reading the content of other users' comments. Therefore, the length of the dwell time can be used to infer whether the current user shares the same emotional tendency as other users' comments. Actions can also be gestures such as "likes." Since liking represents agreement with the content of a comment, if the current user likes a comment, it can be determined that the current user shares the same emotional tendency as other users' comments.

[0143] In practice, when the current user views comments posted by other users through edge intelligence, they can also interact with those comments. The system then acquires and analyzes this interaction to determine if the current user's sentiment is the same as that of the comments posted by other users. If the current user's sentiment matches that of the comments posted by other users, the intent of the other users is determined to be the current user's intent. At this point, response information associated with the commented object is output, including: response information associated with the commented object in response to the current user's interaction.

[0144] In this embodiment, since the edge intelligence makes judgments based on the current user's behavioral data, its accuracy may be low. Therefore, to improve accuracy, after making a judgment, the edge intelligence can output a prompt message to the current user to indicate whether the user agrees with the judgment result. If the user agrees, the next step can be performed.

[0145] After determining the current user's action, the above method can be used to analyze the comment information related to that action, revealing the intentions of other users who wrote the comment. Since the current user's action pertains to that comment, the intentions of the other users who wrote it can be understood as also being the current user's intention. Therefore, after analyzing the intentions of the other users regarding the comment information related to the current user's action, these analyzed intentions can be taken as the current user's intention.

[0146] For users who haven't posted comments, existing technologies prevent merchants or platforms from ascertaining their true intentions and thus hinder their ability to take appropriate action. To address this issue, this embodiment analyzes the intentions of users who haven't posted comments by collecting their browsing behavior data, thereby resolving the problem for these users. Therefore, this embodiment also covers users who haven't posted comments, broadening the scope of the problem and creating more beneficial technical effects for users, merchants, and platforms.

[0147] Optionally, in another embodiment, semantic analysis is performed on the comment information to determine the object being commented on, including:

[0148] Perform semantic analysis on the comment information to obtain the paid orders associated with the comment information;

[0149] The object providing the service in the paid order is identified as the object to be commented on.

[0150] In practice, when performing semantic analysis on comment information, the associated paid orders can be directly obtained, and then the service provider within those paid orders can be identified as the commented object. Here, the service provider refers to the merchant or product consumed corresponding to the paid order.

[0151] In this embodiment, review information refers to the information provided by a user after placing an order, commenting on the service provider involved in the current transaction. Thus, the user's evaluation after the transaction has reliable reference value for the service provider (such as the merchant, product, platform, or other users). Therefore, the reviewed object in this embodiment refers to the merchant or product corresponding to the paid order.

[0152] Optionally, in another embodiment, analyzing the comment information to determine the current user's intent includes:

[0153] The comment information is input into a pre-trained intent analysis model to obtain the current user's intent; or

[0154] The comment information is matched with a preset thesaurus to obtain the current user's intent. The preset thesaurus includes multiple pre-set keywords associated with the intent.

[0155] In practice, intent analysis can be performed using an intent analysis model or by analyzing comment information based on a pre-set thesaurus. When using an intent analysis model, the comment information is input into the model, which then directly outputs the current user's intent. When analyzing comment information based on a pre-set thesaurus, the comment information is matched with keywords in the thesaurus, and the current user's intent is determined based on the matching results. The pre-set thesaurus includes several pre-defined keywords associated with intent.

[0156] The training of the intent analysis model includes:

[0157] Multiple historical comment samples are obtained, each carrying a pre-labeled intent tag; the multiple historical comment samples are used as training samples to train a preset model to obtain the intent analysis model.

[0158] In this embodiment, the intent analysis model can be trained by the terminal device itself, and then intent analysis can be performed directly on the terminal device. When using the intent analysis model for intent analysis, after the intent is analyzed, it can be classified to facilitate corresponding processing by the business layer.

[0159] This embodiment uses an intent analysis model or a preset thesaurus to automatically analyze the intent of comment information, which can avoid the problems of low efficiency, long time consumption and high investment costs that exist in manual analysis and providing corresponding solutions.

[0160] Optionally, in another embodiment, the information processing method provided in this application further includes:

[0161] To obtain the true intent of the current user in response to the response information;

[0162] In practice, the intent analyzed by the intent analysis model may differ from the user's actual intent. This means the intent analysis model may not accurately analyze intent, resulting in subsequent response information that doesn't precisely address the user's problem. Therefore, to improve the accuracy of subsequent intent analysis, the intent analysis model needs to be updated in real time. When updating the intent analysis model, since the current user directly interacts with the response information, to ensure targeted and effective updates, it's necessary to collect the user's actual intent in response to the response information.

[0163] The model parameters of the intent analysis model are updated based on the stated true intent and the comment information.

[0164] In practice, comment information is used as input samples, and the obtained true intent is used as intent labels to update the model parameters of the intent analysis model.

[0165] In this embodiment, by continuously updating the intent analysis model, the accuracy of the analyzed intent can be ensured, thereby effectively solving the current user's problem.

[0166] Optionally, in another embodiment, when the method is applied to a terminal device, the method further includes:

[0167] Obtain the intent analysis model generated by the server, and store the model parameters of the intent analysis model locally.

[0168] In a specific implementation, if the information processing method provided in this embodiment is applied to a terminal device, and the intent analysis model is trained and generated on a server, the information processing method provided in this embodiment further includes: obtaining the intent analysis model generated by the server and storing the model parameters of the intent analysis model locally.

[0169] In summary, the information processing method provided in this application, based on the intent analysis model, edge intelligence analysis (which refers to analyzing the current user's behavioral data through terminal devices), and user comment intent, matches marketing and problem-solving solutions from merchants and platforms to users based on different intents, guiding and recommending relevant merchants or products. This can greatly improve user experience, reduce negative merchant reviews, improve operational efficiency, and increase user trust in the platform. Compared to related technical solutions, its advantages include at least the following:

[0170] 1. Comments are the most important channel for obtaining user feedback. This application guides and recommends based on comments, which is efficient and accurate.

[0171] 2. This application not only targets users who comment, but also covers users who do not comment, thus addressing a wider range of issues and yielding greater benefits;

[0172] 3. For users who are about to post negative reviews, the method provided in this application can address their issues more proactively, ensuring that these issues are resolved before the negative review is posted, thereby reducing the negative review rate for merchants.

[0173] 4. Based on user intent, the method provided in this application offers corresponding solutions to users, which can increase user satisfaction and enhance their trust in the platform;

[0174] 5. The method provided in this application automatically analyzes user comments based on the intent analysis model and the terminal device, and the terminal device outputs corresponding response information, thereby saving merchants' human resource investment.

[0175] Based on the same inventive concept as the above embodiments, a second aspect of the present disclosure provides an information processing apparatus, such as... Figure 3 As shown, the information processing device includes:

[0176] Module 31 is used to obtain comment information displayed to the current user;

[0177] The first determining module 32 is used to perform semantic analysis on the comment information to determine the object being commented on;

[0178] Output module 33 is used to output response information associated with the commented object.

[0179] Optionally, the device further includes:

[0180] The second determining module is used to perform semantic analysis on the comment information to determine the sentiment tendency of the comment information;

[0181] Output module 33 includes:

[0182] The first output submodule is used to output the resources and / or information of the commented object when the sentiment tendency is a positive sentiment tendency;

[0183] The second output submodule is used to output resources and / or information of other objects similar to the object being commented on when the sentiment tendency is negative.

[0184] Optionally, if the comment information is "no comment information", the device further includes:

[0185] The third determining module is used to perform semantic analysis on the uncommented information to determine the sentiment tendency of the uncommented information.

[0186] Output module 33 includes:

[0187] The detection submodule is used to detect posting requests for the uncommented information;

[0188] The third output submodule is used to respond to the posting request by outputting response information associated with the commented object, which guides the sentiment tendency of the uncommented information to switch to a positive sentiment tendency.

[0189] Optionally, the second output submodule includes:

[0190] The analysis submodule is used to analyze the comment information to determine the current user's intent;

[0191] The query submodule is used to query the historical order database for similar attribute objects within a preset geographical range of the currently commented object that have an attribute similarity greater than a preset similarity with the currently commented object, when the current user's intent is to confuse the currently commented object.

[0192] The filtering submodule is used to determine, based on the comment information of each of the multiple similar attribute objects, a target object with a positive review score greater than a preset positive review score when multiple similar attribute objects exist within a preset geographical range of the currently commented object.

[0193] The third output submodule is used to output information about the target object with the similar attributes.

[0194] Optionally, the device further includes:

[0195] The first query module is used to query the object database for objects of the same category as the current object being commented on, within the preset geographical range of the current object being commented on, if there are no similar attribute objects within the preset geographical range of the current object being commented on.

[0196] The first filtering module is used to determine, when there are multiple objects of the same category within a preset geographical range of the currently commented object, a target object of the same category whose positive review score is greater than a preset positive review score from the multiple objects of the same category based on the review information of each of the multiple objects of the same category.

[0197] The second output submodule includes:

[0198] The fourth output submodule is used to output information about the target objects of the same category.

[0199] Optionally, the device further includes:

[0200] The first analysis module is used to analyze the comment information and determine the intent of the current user;

[0201] The sending module is used to send coupons for the currently commented object, coupons for the e-commerce platform where the currently commented object is located, or coupons for the delivery platform associated with the currently commented object to the user account of the current user when the current user's intention is to increase the discount.

[0202] Output module 33 includes:

[0203] The fifth output submodule is used to output coupon notification information that coupons have been issued to the current user's account.

[0204] Optionally, the device further includes:

[0205] The second analysis module is used to analyze the comment information to determine the current user's intent;

[0206] The problem identification module is used to identify the problem to be solved based on the current user's intent.

[0207] The sixth output submodule is used to match the problem to be solved with a preset solution database and output the solution corresponding to the problem to be solved; and / or

[0208] The problem to be solved is input into a pre-trained solution prediction model to obtain and output the solution corresponding to the problem to be solved.

[0209] Optionally, the device further includes:

[0210] The operation acquisition module is used to acquire the operations of the current user on other users' posted comments.

[0211] The operation analysis module is used to analyze the current user's operation and determine whether the current user's sentiment tendency is the same as the sentiment tendency of the comments posted by other users.

[0212] The intent determination module is used to determine the intent of the other users as the intent of the current user when the current user's sentiment tendency is the same as the sentiment tendency of the comments posted by other users.

[0213] Output module 33 includes:

[0214] The response output submodule is used to respond to the current user's operation and output response information associated with the commented object.

[0215] Optionally, the first determining module 32 includes:

[0216] The order information acquisition submodule is used to perform semantic analysis on the comment information and obtain the paid orders associated with the comment information;

[0217] The object determination submodule is used to determine the object providing services in the paid order as the object to be commented on.

[0218] According to a third aspect of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the information processing method described in the first aspect above.

[0219] A fourth aspect of this disclosure provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the information processing method described in the first aspect.

[0220] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0221] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0222] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0223] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0224] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0225] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0226] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0227] The above provides a detailed description of the information processing method, apparatus, electronic device, and storage medium provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An information processing method, characterized in that, The method includes: Retrieve comment information displayed to the current user; Perform semantic analysis on the comment information to determine the object being commented on; Output the response information associated with the commented object; If the comment information is that no comment has been posted, the method further includes: Perform semantic analysis on the uncommented information to determine its sentiment tendency; Detect requests to post comments on information for which no comments have been posted; In response to the posting request, output response information associated with the commented object to guide the sentiment of the uncommented information to switch to a positive sentiment.

2. The method according to claim 1, characterized in that, Also includes: Perform semantic analysis on the comment information to determine its sentiment tendency; Output response information associated with the commented object, including: If the emotional tendency is positive, output the resources and / or information of the object being commented on; If the emotional tendency is negative, output resources and / or information of other objects similar to the object being commented on.

3. The method according to claim 2, characterized in that, Output resources and / or information about other objects similar to the object being commented on, including: The comment information is analyzed to determine the current user's intent; If the current user's intent is to obfuscate the currently commented object, the system queries the historical order database for similar attribute objects within a preset geographical range of the currently commented object that have an attribute similarity greater than a preset similarity. If multiple similar attribute objects exist within a preset geographical range of the currently commented object, then based on the comment information of each of the multiple similar attribute objects, determine the similar attribute target object whose positive review score is greater than the preset positive review score from among the multiple similar attribute objects; Output the resources and / or information of the target object with the similar attributes.

4. The method according to claim 3, characterized in that, The method further includes: If no similar attribute objects exist within the preset geographical range of the currently commented object, the object database is queried to find objects of the same category as the currently commented object within the preset geographical range of the currently commented object; If there are multiple objects of the same category within the preset geographical range of the object being commented on, then, based on the comment information of each of the multiple objects of the same category, determine the target object of the same category whose positive review score is greater than the preset positive review score from among the multiple objects of the same category; Output resources and / or information about other objects similar to the object being commented on, including: Output the resources and / or information of the target object of the same category.

5. The method according to claim 2, characterized in that, The method further includes: The comment information is analyzed to determine the current user's intent; If the current user's intention is to increase resource allocation, the resources of the currently commented object, the e-commerce platform where the currently commented object is located, or the delivery platform associated with the currently commented object are sent to the current user's account. Output the resources and / or information of the object being commented on, including: Output a resource notification message indicating that resources have been allocated to the current user's account.

6. The method according to claim 1, characterized in that, The method further includes: The comment information is analyzed to determine the current user's intent; Based on the current user's intent, determine the problem to be solved; Match the problem to be solved with a preset solution database and output the solution corresponding to the problem to be solved; and / or The problem to be solved is input into a pre-trained solution prediction model to obtain and output the solution corresponding to the problem to be solved.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: The operation of retrieving the current user's posted comments on other users; Analyze the current user's actions to determine whether the current user's sentiment is the same as the sentiment of the comments posted by other users; If the current user's sentiment is the same as the sentiment of comments posted by other users, the intent of the other users will be determined as the intent of the current user. Output response information associated with the commented object, including: In response to the current user's action, output response information associated with the object being commented on.

8. The method according to any one of claims 1-6, characterized in that, Perform semantic analysis on the comment information to determine the object being commented on, including: Perform semantic analysis on the comment information to obtain the paid orders associated with the comment information; The object providing the service in the paid order is identified as the object to be commented on.

9. An information processing device, characterized in that, The device includes: The acquisition module is used to retrieve comment information displayed to the current user. The first determining module is used to perform semantic analysis on the comment information to determine the object being commented on; The output module is used to output response information associated with the commented object; The third determining module is used to perform semantic analysis on the uncommented information when the comment information is uncommented information, and to determine the sentiment tendency of the uncommented information; The detection submodule is used to detect posting requests for the uncommented information; The third output submodule is used to respond to the posting request by outputting response information associated with the commented object, which guides the sentiment tendency of the uncommented information to switch to a positive sentiment tendency.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the information processing method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the information processing method according to any one of claims 1 to 8.

Images