A method, apparatus, computer equipment, and storage medium for displaying product pages.

Abstract

This application provides a product page display solution. When a target product is updated, the latest test report is quickly retrieved to ensure timely information acquisition. Then, the differences between the latest and current test reports are assessed. If differences exist, the discrepancies are further identified, the product display page is updated, and the discrepancies are highlighted. These steps work together to ensure the accuracy and timeliness of the product display page information, improve the efficiency of users obtaining product information, and allow users to intuitively understand changes in the product test report, thus optimizing the user experience.

Description

Technical Field

[0001] This application relates to the field of product page display technology, and in particular to a product page display method, apparatus, computer equipment, and storage medium. Background Technology

[0002] Currently, product pages typically display basic product information and testing reports to help consumers understand product quality and compliance. However, when product updates lead to changes in testing reports, there can be a problem of asynchronous report updates, which hinders the improvement of the consumer experience and the transparency of product information. Summary of the Invention

[0003] The purpose of this application is to at least solve one of the aforementioned technical defects, and in particular, to provide a product page display method that can solve the problem of asynchronous updates of product testing reports.

[0004] Firstly, this application provides a method for displaying a product page, including:

[0005] When an update to the target product is detected, the latest test report is retrieved from the test report database based on the target product's unique identifier.

[0006] Determine if there are any discrepancies between the latest test report and the current test report;

[0007] If so, then identify the differences between the latest test report and the current test report;

[0008] Update the product display page based on the latest test report and mark the differences in the latest test report.

[0009] In one embodiment, determining whether there is a difference between the latest test report and the current test report includes:

[0010] A first field table is generated based on the current test report; the first field table includes all the first fields in the current test report.

[0011] A second-field table is generated based on the latest test report; the second-field table includes all the second fields from the latest test report.

[0012] Match each first field and each second field based on the field name;

[0013] If there is no unmatchable first or second field, then if the change in the field value of each matching field pair is less than the first threshold, it is determined that there is no difference; if the change in the field value of each matching field pair is greater than the first threshold, it is determined that there is a difference.

[0014] If there is a first field and / or second field that cannot be matched, then a difference is determined to exist.

[0015] In one embodiment, the difference items include numerical changes, increases, and decreases. Determining the difference items between the latest test report and the current test report includes:

[0016] The field values ​​of each matched field that change by more than the first threshold are identified as numerical variation items.

[0017] The first field that cannot be matched is identified as the item to be reduced.

[0018] The second field that cannot be matched is determined to be an added item.

[0019] In one embodiment, the product page display method further includes:

[0020] In response to a target user's request for explanation of a discrepancy, the discrepancy to be explained is identified as the target discrepancy; the target user is the user currently browsing the target product page.

[0021] Based on the type label of the target difference item, the field name of the target difference item, and the description of the target product name, request prompt words are provided.

[0022] Input the explanation request prompt into the explanation model to obtain the explanation result;

[0023] The results will be presented to the target users.

[0024] In one embodiment, in response to a target user's request for explanation of the differences, the following steps are included:

[0025] If the system detects that the target user hovers over any of the differences, an explanation request for that difference will be generated.

[0026] In one embodiment, the differences are noted in the latest test report, including:

[0027] Obtain a user profile of the target users; the target users are those who are currently browsing the target product page.

[0028] Predict the target user's interest in each difference item based on user profiles;

[0029] The visual emphasis level of the difference item is determined based on the level of attention; the higher the level of interest, the higher the corresponding visual intensity level.

[0030] Select the appropriate annotation method based on the level of visual emphasis to annotate the differences.

[0031] In one embodiment, the user profile includes one or more attribute labels, each attribute label corresponding to a set of interest fields. The user profile is used to predict the target user's interest in each difference item, including:

[0032] Semantically match the differences with the sets of fields of interest corresponding to each attribute label;

[0033] The degree of interest is determined based on the highest semantic similarity in the semantic matching results.

[0034] In one embodiment, the product page display method further includes:

[0035] Monitor the batch number of the target product;

[0036] When the product batch number is updated, it is determined that the target product has been updated.

[0037] Secondly, this application provides a product page display device, comprising:

[0038] The retrieval module is used to retrieve the latest test report from the test report database based on the unique identifier of the target product when an update is detected.

[0039] The judgment module is used to determine whether there are any differences between the latest test report and the current test report;

[0040] The difference determination module is used to identify the differences between the latest test report and the current test report when there are differences between them.

[0041] The update module is used to update the display page of the target product based on the latest test report and to mark the differences in the latest test report.

[0042] Thirdly, this application provides a computer device including one or more processors and a memory storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, they perform the steps of the product page display method in any of the above embodiments.

[0043] Fourthly, this application provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the product page display method in any of the above embodiments.

[0044] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0045] Based on the product page display method in this solution, when a target product is updated, the latest test report is quickly retrieved first to ensure timely information acquisition. Then, the differences between the latest and current test reports are assessed. If differences exist, the discrepancies are further identified, the product display page is updated, and the discrepancies are highlighted. These steps work together to ensure the accuracy and timeliness of the product display page information, improve the efficiency of users obtaining product information, and allow users to intuitively understand changes in the product test reports, thus optimizing the user experience. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art 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.

[0047] Figure 1 A flowchart illustrating a product page display method provided in one embodiment of this application;

[0048] Figure 2 This is a flowchart illustrating the process of determining whether there is a difference between the current test report and the latest test report in one embodiment of this application;

[0049] Figure 3 This is a flowchart illustrating the categories of differences in one embodiment of this application;

[0050] Figure 4 This is a flowchart illustrating the explanation of the differences in one embodiment of this application;

[0051] Figure 5 This is a flowchart illustrating the process of annotating differences in one embodiment of this application;

[0052] Figure 6 This is an internal structural diagram of a computer device provided in one embodiment of this application. Detailed Implementation

[0053] The technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. The embodiments described in the specification are only some embodiments of this application, and 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.

[0054] This application provides a method for displaying product pages; please refer to [link / reference]. Figure 1 This includes steps S102 to S108.

[0055] S102, when an update to the target product is detected, the latest test report is retrieved from the test report database based on the unique identifier of the target product.

[0056] As we understand it, a target product refers to a specific product on an e-commerce platform that requires synchronized report updates. It possesses a unique product code to distinguish it from other products. A unique identifier is a unique string or number combination assigned to a target product, typically generated by the platform using a specific algorithm. This uniqueness ensures accurate identification of the product across different systems and databases. The testing report database stores product testing reports, including various reports such as quality testing, safety certification, and performance testing. These reports are stored according to preset data formats and indexing rules for rapid retrieval. The latest testing report refers to the most recent version generated after the target product has been retested. When the target product is updated, the unique identifier establishes a connection between the target product and the testing report database, thereby retrieving the latest testing report. When information about the target product, such as specifications, ingredients, or manufacturing processes, is updated, the system needs to promptly obtain testing reports reflecting these updated product statuses to ensure accurate updates to the product display page in subsequent steps.

[0057] When an update to a target product is detected, the system extracts the product's unique identifier and performs a query in the inspection report database using Structured Query Language (SQL). The system then finds the inspection report with the matching unique identifier and the most recent entry time as the latest inspection report. For example, after receiving a product update signal, the system retrieves the product's unique identifier "PROD123" from the product information table, and then executes the SQL statement "SELECT * FROM inspection_reports WHERE product_id='PROD123' ORDER BY report_date DESCLIMIT 1" in the inspection report database to obtain the latest inspection report.

[0058] S104, determine whether there is a difference between the latest test report and the current test report.

[0059] As we can understand, the current test report refers to the test report displayed on the product page before the current update of the target product. It records the various test indicators and results before the product update. Differences refer to the differences in content between the latest test report and the current test report, including changes in the values ​​of test indicators, additions or deletions of test items, and changes in test conclusions. Judgment refers to the process by which the system compares and analyzes the latest test report and the current test report using specific algorithms and logic to determine whether there are any differences between them.

[0060] The main function of step S104 is to take the latest test report obtained in step S102 and compare it with the currently used test report to determine if there are any differences. This is a crucial step in deciding whether to execute subsequent steps. If there are no differences, there is no need to perform subsequent difference identification and page update operations, saving system resources; if there are differences, step S106 is triggered to identify the differences. This step, through a systematic comparison of the two reports, provides a decision-making basis for the entire method, ensuring that subsequent processing is only performed when the test report has indeed changed. This makes the updates to the product display page more targeted and effective, forming an organic connection with the preceding and following steps, and ensuring the efficiency and accuracy of the method.

[0061] S106, if so, then identify the differences between the latest test report and the current test report.

[0062] It is understandable that differences refer to specific content units that differ between the latest and current test reports, such as the value of a single test indicator, the description of the test item, or the statement of the test conclusion. Identifying differences refers to the process of accurately locating and extracting the specific content that differs between the two test reports using specific technical means and methods. Step S106 is executed after step S104 determines the existence of differences; its main purpose is to clarify the specific differences between the two reports, providing specific content for the page update and difference annotation in step S108. This step involves in-depth analysis of the two differing reports, refining and extracting the differences to facilitate subsequent annotation on the latest test report.

[0063] S108. Update the display page of the target product according to the latest test report and mark the differences in the latest test report.

[0064] As we can understand it, a product display page refers to a webpage or application interface on an e-commerce platform used to present information about a target product to users, including product images, prices, specifications, and testing reports. Updating the target product display page refers to modifying, supplementing, or replacing the parts of the product display page related to the testing report based on the latest testing report. Highlighting differences refers to using specific visual markers in the latest testing report, such as color highlighting, underlining, or borders, to highlight the differences so that users can intuitively identify the content that differs from the current testing report.

[0065] Step S108 is the final execution step of the entire product page display method. Its function is to reflect the differences identified in step S106 on the product display page, allowing users to promptly understand the latest status and changes in the product inspection report. This step follows the previous steps, replacing the original inspection report with the latest one and highlighting the differences between the two reports, enabling users to more accurately monitor the dynamic changes of the target product during the update process.

[0066] Based on the product page display method in this solution, when a target product is updated, the latest test report is quickly retrieved first to ensure timely information acquisition. Then, the differences between the latest and current test reports are assessed. If differences exist, the discrepancies are further identified, the product display page is updated, and the discrepancies are highlighted. These steps work together to ensure the accuracy and timeliness of the product display page information, improve the efficiency of users obtaining product information, and allow users to intuitively understand changes in the product test reports, thus optimizing the user experience.

[0067] In one embodiment, to determine whether there is a difference between the latest test report and the current test report, please refer to [link to relevant documentation]. Figure 2 This includes steps S202 to S210.

[0068] S202, Generate a first field table based on the current test report. The first field table includes all the first fields from the current test report.

[0069] It can be understood that the first field table is a structured data set generated based on the current test report. The first field refers to an information unit with specific meaning in the current test report, including the names of various test items, test indicators, etc. Each first field corresponds to a unique field name and a corresponding field value. Generating the first field table refers to the process of extracting information from the current test report using data parsing technology and organizing it according to a preset data structure to form a structured table that can be processed by a computer. Its core function is to transform the unstructured current test report into a structured first field table, providing a standardized data foundation for subsequent steps of field matching and difference judgment. By decomposing the current test report into multiple first fields, each containing a clear field name and field value, subsequent comparisons with the second field table can be performed under a unified data structure, avoiding comparison errors caused by differences in report formats.

[0070] Specifically, regular expressions can be used to parse the current detection report and generate the first field table. The text format of the current detection report can be analyzed to determine the typical expressions for field names and values. Then, corresponding regular expressions can be written to extract field names and corresponding field values ​​from the report text through regular expression matching. All extracted field names and values ​​are stored in the first field table as key-value pairs. The first field table adopts a relational database table structure, containing two columns: "Field Name" and "Field Value," with each row corresponding to one first field. Alternatively, a deep learning-based natural language processing model can be used for field extraction and first field table generation. The system uses a large number of detection report samples labeled with field names and values ​​to train a large language model. This model can automatically identify field boundaries and semantic information in the current detection report. When generating the first field table, the current detection report is input into the trained model. The model analyzes the text sequence and outputs the field name and field value for each field.

[0071] S204, Generate a second field table based on the latest test report. The second field table includes all the second fields from the latest test report.

[0072] It can be understood that the second field table is a structured data set generated based on the latest test report. The second field refers to an information unit with specific meaning in the latest test report, corresponding to the definition of the first field. Each second field also contains a unique field name and corresponding field value. Generating the second field table refers to transforming the information in the latest test report into a form conforming to a preset data structure through data extraction and structuring processing techniques, so as to facilitate subsequent comparison operations. Its function is to transform the latest test report into a structured second field table, forming a comparison data pair with the first field table, providing additional data support for field matching in step S206. By adopting the same data structure and extraction rules as the first field table, the consistency of the two field tables in format is ensured, allowing field matching to be performed based on the same standard, reducing matching errors caused by structural differences. The second field table generated in this step, together with the first field table, constitutes the basic data for difference judgment. The structuring processing of both allows subsequent field matching, field value comparison, and other operations to be performed efficiently. The extraction of the second field table can also be performed in a similar manner to the first field table.

[0073] S206, Match each first field and each second field according to the field name.

[0074] As we can understand, a field name is a string or symbol used to identify the meaning of a field. In the first and second field tables, field names are identifiers that distinguish different fields, such as "appearance" and "fragrance." Matching refers to the process of associating the first field in the first field table with the second field in the second field table according to their field names. Matching forms matching field pairs, that is, pairs of data consisting of the first and second fields with the same field name. Specifically, it involves traversing all fields in the first and second field tables, determining the correspondence between fields by comparing field names, finding fields with the same field name, and establishing a connection. Its main function is to determine the correspondence between fields in the two field tables by comparing field names, providing a basis for subsequent judgments on whether there are any unmatchable fields and for calculating the range of field value changes. Through accurate field matching, it is possible to identify which fields are common to both reports and which fields are unique to one report, which is a prerequisite for the difference judgment in steps S208 and S210. This step establishes a correspondence between fields, linking fields in the two field tables so that subsequent difference assessments can be performed on specific field pairs or isolated fields, ensuring the accuracy of the difference assessment. It works in conjunction with the preceding and following steps to form a complete difference assessment logic chain, avoiding errors in difference assessment caused by incorrect field correspondence.

[0075] Specifically, a hash table lookup algorithm can be used for field matching. The system first stores all field names from the first field table into a hash table. The key of the hash table is the field name, and the value is the index position of the corresponding first field in the table. Then, it iterates through each second field in the second field table, extracts its field name, and searches it in the hash table. If a matching key is found, a matching relationship is established between the second field and the corresponding first field in the hash table, forming a matching field pair. If no match is found, it is marked as an unmatchable second field. Simultaneously, unmatched fields in the first field table are also marked as unmatchable first fields. For example, if the first field table contains the field names "hardness" and "density," and the second field table contains the field names "hardness" and "elastic modulus," after hash table matching, the "hardness" field forms a matching field pair, "density" is marked as an unmatchable first field, and "elastic modulus" is marked as an unmatchable second field.

[0076] S208, if there is no unmatchable first or second field, then if the change in the field value of each matching field pair is less than the first threshold, it is determined that there is no difference; if the change in the field value of each matching field pair is greater than the first threshold, it is determined that there is a difference.

[0077] It is understood that an unmatchable first field refers to a field that exists in the first field table but cannot be found in the second field table with the same or similar semantic name. An unmatchable second field refers to a field that exists in the second field table but cannot be found in the first field table with the same or similar semantic name. A matched field pair refers to a pair of data consisting of the first and second fields successfully matched in step S206, containing the same field name and their respective field values. The magnitude of field value change refers to the degree of difference between the field value of the second field and the field value of the first field in the matched field pair, usually calculated using a specific formula, such as the ratio of the absolute difference to the original field value. The first threshold is a preset critical value used to determine whether the field value change is significant. This value is set according to the field type and actual business needs; different first thresholds can be set for different fields. Determining no difference means that after analyzing the magnitude of field value changes in the matched field pair, it is determined that the two test reports are not significantly different in content. Determining difference means that after analysis, it is determined that the two test reports are significantly different in content.

[0078] This step focuses on fields shared by both reports, assessing differences by analyzing the degree of change in field values. This overcomes the limitations of simply judging by the presence or absence of fields, enabling the identification of situations where field names are identical but field values ​​have significantly different values. Building upon the confirmation in step S206 that all fields match, this step delves deeper into the analysis at the field value level, complementing step S210 and together forming a complete difference assessment logic. This ensures that both missing fields and significant changes in field values ​​are accurately identified, improving the comprehensiveness and accuracy of the difference assessment.

[0079] S210, if there is a first field and / or a second field that cannot be matched, then a difference is determined to exist.

[0080] It is understandable that a "first field that cannot be matched" refers to a field that exists in the first field table but has no corresponding matching field in the second field table, indicating that some field information in the current test report is missing from the latest test report. Similarly, a "second field that cannot be matched" refers to a field that exists in the second field table but has no corresponding matching field in the first field table, indicating that field information not present in the current test report has been added to the latest test report. The main function of step S210 is to directly determine a difference between the two test reports when a field that cannot be matched is found in step S206. This step addresses the difference in the field composition of the two reports, i.e., the situation where one report contains fields that are not present in the other report. This situation usually means that the test items for the product have been added or removed, constituting a substantial change in content, and therefore needs to be determined as a difference.

[0081] In one embodiment, the difference items include numerical changes, increases, and decreases; see [link to relevant documentation]. Figure 3 The differences between the latest test report and the current test report are identified, including steps S302 to S306.

[0082] S302, the field values ​​of each matched field pair that change by more than the first threshold are identified as numerical variation items.

[0083] As can be understood, a matched field pair refers to a pair of data units consisting of a first field and a second field that match successfully by field name. The first field comes from the first field table generated in the current detection report, and the second field comes from the second field table generated in the latest detection report; both have the same or semantically equivalent field names. The first threshold is a pre-set critical value used to determine whether a field value change has reached a significant level. This threshold can be adjusted according to the field type, importance, and business needs; different fields can be configured with different first thresholds. A numerical change item refers to a difference item in the matched field pair whose field value change exceeds the first threshold, indicating that the content of that field has changed significantly in both detection reports. This step allows for the precise identification of fields shared by both detection reports where the content has changed significantly.

[0084] S304, the first field that cannot be matched is identified as the reduction item.

[0085] As can be understood, the "unmatched first field" refers to a field that exists in the first field table but has no corresponding matching field in the second field table. In other words, the current test report contains a field that the latest test report does not. The "reduction item" refers to the difference caused by the existence of the unmatched first field, indicating that the latest test report has reduced the test content corresponding to that field compared to the current test report.

[0086] S306 determines the second field that cannot be matched as an addendum.

[0087] It can be understood that an unmatched second field refers to a second field that exists in the second field table but for which no corresponding matching field is found in the first field table. In other words, the latest test report contains a field that the current test report does not. An added item refers to a difference item created by the existence of an unmatched second field, indicating that the latest test report has added the corresponding test content for that field compared to the current test report. Determining an added item is the process of explicitly classifying an unmatched second field as an added item. This process clearly identifies the increase in the number of fields between the two test reports.

[0088] In one embodiment, please refer to Figure 4 The product page display method also includes steps S402 to S408.

[0089] S402, in response to the target user's request for explanation of the difference item, the difference item to be explained is identified as the target difference item.

[0090] It can be understood that a request for explanation of a difference item refers to a request issued by a target user through a specific interactive operation, seeking detailed information about the difference item. This request can be triggered by clicking an explanation button on the page, entering question text, etc. The difference item to be explained refers to the specific difference item for which the target user requests explanation; it can be one or more of a numerical change, addition, or subtraction. The target difference item refers to the difference item identified from the difference items to be explained, for which the system needs to generate an explanation; it usually corresponds directly to the user's specific request. The response refers to the process by which the system, upon receiving a user's explanation request, executes the corresponding operation according to preset processing logic, ensuring that the user's request is processed promptly.

[0091] S404, based on the type label of the target difference item, the field name of the target difference item, and the target product name, a description request prompt is generated.

[0092] It can be understood that the type label of the target difference item refers to the label used to identify the type to which the target difference item belongs, such as "numerical change item," "increase item," "decrease item," etc. This label is generated when the difference item is determined and is used to distinguish different types of difference items. The field name of the target difference item refers to the field name in the test report corresponding to the target difference item, such as "tensile strength" and "corrosion resistance grade," which is a key identifier for identifying the specific content of the difference item. The target product name refers to the name of the product to which the target difference item belongs. It is unique and can be used to clearly define the scope of the object to be explained. The explanation request prompt is the text information used to issue an explanation request to the explanation model. It contains the key information needed to generate the explanation result and can guide the model to generate explanation content that meets the requirements. Composing the explanation request prompt refers to the process of combining the type label, field name, target product name, and other elements of the target difference item into a prompt according to a preset format and logic. By incorporating the type label, field name, target product name, and other elements into the prompt, sufficient contextual information can be provided to the explanation model, enabling it to understand the specific object and difference type that needs to be explained, thereby generating targeted and accurate explanation results. Specifically, a fixed template filling method can be used to compose the explanation request prompt. The system pre-sets prompt templates, which include placeholders for type labels, field names, and target product names. For example, "Please explain the reasons for the differences in [field names] ([type labels]) of product [target product name] and the resulting impact."

[0093] S406, input the description request prompt into the description model to obtain the description result.

[0094] In essence, the "large-scale explanation model" refers to an AI model trained to generate explanations for discrepancies based on input prompts. This model is trained on extensive data from product testing, industry standards, and technical documents, possessing the ability to understand discrepancy information and generate professional explanations. The explanation result refers to the output of the large-scale explanation model based on the explanation request prompts, containing explanations about the target discrepancy, including the reasons, impacts, and basis for the discrepancy, presented in natural language text. "Inputting into the large-scale explanation model" refers to the process of converting the explanation request prompts according to the model's required format and then passing them to the model for processing. The model analyzes and understands the prompts to generate the corresponding explanation result. This step, through the large-scale model's processing, transforms structured prompt information into natural language explanations, ensuring users can quickly obtain explanatory information and enhancing their understanding of product discrepancies.

[0095] S408 will display the results to the target user.

[0096] As we can understand it, "display" refers to the process of presenting the explanatory results to the target user in a way that is perceptible to the user. This is typically achieved through pop-up windows or embedded text areas on the product display page, ensuring that users can intuitively and conveniently access the explanatory content. Presenting the explanatory results to the target user means that the system presents the generated explanatory results to the target user through a preset display method, allowing the user to view and understand the relevant explanations of the differences. Through this step, users can gain a deeper understanding of the reasons and impacts behind the differences, enhancing their trust and understanding of the product, thereby promoting their purchase decisions or increasing their satisfaction with the platform.

[0097] In one embodiment, responding to a target user's explanation request for a difference item includes: if it is detected that the target user hovers over any difference item, then generating an explanation request for the difference item.

[0098] In one embodiment, the differences are highlighted in the latest test report; please refer to [link / reference]. Figure 5 This includes steps S502 to S508.

[0099] S502, Obtain the user profile of the target user.

[0100] User profiling, as we understand it, refers to a user characteristic model constructed using data mining and analysis techniques based on multi-dimensional data such as the target user's basic information, behavioral data, and preference settings. Obtaining a target user's user profile involves the system calling a user database interface to extract various data associated with the target user's identifier, integrating and analyzing this data according to a pre-defined profile model to form structured user profile data. Information within the user profile, such as the user's historical attention to product detection indicators and consumption habits, is crucial for predicting interest levels. Through this step, the system can identify the unique characteristics of the current user from a massive user base, laying the foundation for differentiated labeling methods and ensuring that subsequent labeling operations are more targeted and effective.

[0101] Specifically, user profiles are obtained through user behavior log analysis. The system periodically extracts data from the user behavior log database, including the target user's browsing history, click behavior, types of purchased goods, and frequency of viewing test reports. For example, it counts the number of times a user has viewed test indicators such as "tensile strength" and "corrosion resistance" in the past 30 days, and the degree of attention paid to safety certification items in test reports when purchasing goods. This data is quantified according to preset weights, such as multiplying the number of views by the corresponding weight to convert it into an attention index. After integration, a user profile is formed, containing tag information such as the types of test indicators the user cares about and the intensity of their attention. Alternatively, a neural network classification model can be used to categorize the target user's behavior logs. That is, using the user's real-time behavior data, the user is tagged to form training data, and then the classification model is trained using the training data. For example, if a user quickly browses a product page but spends a long time on the "heavy metal content" test item, they can be given a tag indicating a concern for health.

[0102] S504, predict the target user's interest in each difference item based on the user profile.

[0103] It can be understood that interest level refers to the degree of attention a target user pays to each difference item. It is a quantitative indicator calculated based on user profiles and difference item features; the higher the value, the more attention the user pays to that difference item. The main function of this step is to connect user profiles with difference item annotations. By analyzing user preferences reflected in the user profile and combining features such as detection indicators involved in each difference item, it predicts the user's interest level for each difference item, providing a basis for subsequently determining the visual emphasis level. This step follows the user profile obtained in step S502, transforming user preferences into quantitative values ​​of attention to specific difference items, so that the annotation of difference items can revolve around the user's points of interest.

[0104] S506 determines the visual emphasis level of the difference item based on the level of attention. The higher the level of interest, the higher the corresponding visual intensity level.

[0105] It can be understood that attention level, or the target user's interest in each difference item, is a quantitative indicator predicted through step S504, reflecting the degree of attention users pay to different differences. Visual emphasis level refers to the classification used to distinguish the visual prominence of differences when displayed, typically divided into multiple levels, such as high, medium, and low. The higher the level, the more visually striking the difference item is on the page. Determining the visual emphasis level of differences based on attention level involves comparing the predicted quantitative value of attention with the preset level classification standard, assigning a corresponding visual emphasis level to each difference item. Through this step, the system can classify differences according to the user's attention level, tilting visual resources towards differences that users are more interested in, improving the efficiency of information delivery, and preventing users from missing important content among a large number of differences. A threshold range method can be used to determine the visual emphasis level. The system presets multiple attention threshold ranges, each corresponding to a visual emphasis level. For example, an attention level of 0.8-1.0 corresponds to high, 0.4-0.8 to medium, and 0-0.4 to low. For example, a difference item with an attention score of 0.85, falling within the 0.8-1.0 range, is identified as a high visual emphasis level; another difference item with an attention score of 0.5, falling within the 0.4-0.8 range, is identified as a medium level. A dynamic adaptive grading method can also be used to determine the visual emphasis level. The system dynamically adjusts the threshold range based on the number of difference items for the current target product and the distribution of attention scores for each difference item. For example, when the number of difference items is small (e.g., less than 5), the level is divided into two levels: the top 50% of attention scores are high-level, and the bottom 50% are low-level; when the number of difference items is large (e.g., more than 10), it is divided into four levels, and the threshold for each level is determined based on the normal distribution of attention scores. Simultaneously, the system adjusts the granularity of the grading based on the user's information processing ability characteristics in the user profile. For users with strong information processing abilities, more levels are assigned to provide more detailed visual differentiation; for users with weaker information processing abilities, the number of levels is reduced to avoid visual confusion. For example, if a user profile shows that a user is accustomed to browsing quickly, the system will simplify the levels to two levels, high and low, to make the key points more prominent.

[0106] S508: Select the corresponding annotation method to annotate the differences based on the visual emphasis level.

[0107] Visual emphasis level refers to the degree of visual prominence of a difference item during display, typically divided into multiple levels. A higher level indicates that the difference item needs to be presented in a more prominent manner. The annotation method refers to the specific form in which the difference item is marked in the latest test report, including various visual techniques such as color highlighting, bolding, adding borders, and dynamic flashing. This step involves selecting an appropriate method from a pre-set annotation method library based on the determined visual emphasis level for each difference item, visually highlighting the difference item to attract user attention. Its main function is to translate the visual emphasis level into a specific visual representation. Through differentiated annotation methods, target users can quickly identify the differences they are interested in, improving the efficiency of information acquisition. This step enables the system to personalize the annotation of difference items, highlighting the key points of user attention while avoiding interference from irrelevant information. This makes the display of the latest test report more in line with users' reading habits and information needs, thereby improving users' understanding and trust in product information. A pre-set level-method mapping table can be used to select the annotation method. The system pre-establishes a correspondence between visual emphasis levels and annotation methods. For example, high-level corresponds to red highlighting and bold text, medium-level corresponds to yellow highlighting, and low-level corresponds to gray borders. It employs dynamic combination annotation methods and optimizes the annotation effect based on user interaction feedback. Multiple annotation method combinations can be configured for each visual emphasis level. For example, high-level can choose combinations such as "red highlighting + dynamic underline + subtle animation" or "orange background + bold + icon." The system selects appropriate combinations based on user profiles and their sensitivity to different visual elements. For instance, users sensitive to color are given priority for combinations with significant color differences, while users sensitive to dynamic elements receive additional animation effects.

[0108] In one embodiment, the user profile includes one or more attribute tags, each attribute tag corresponding to a set of interest fields. Predicting the target user's interest in each difference item based on the user profile includes: semantically matching each difference item with the set of interest fields corresponding to each attribute tag; and determining the interest level based on the highest semantic similarity among the semantic matching results.

[0109] Attribute tags are identifiers used to describe user characteristics, such as "a follower of maternal and infant products" or "strong environmental awareness." Each attribute tag reflects a user's characteristics or preferences in a specific area. The set of fields of interest refers to the set of test report fields that the user is interested in, corresponding to each attribute tag. For example, the set of fields of interest for "a follower of maternal and infant products" includes fields such as "formaldehyde content," "heavy metal content," and "material safety." Semantic matching refers to the process of calculating the semantic similarity between the difference item and the fields in the set of fields of interest using natural language processing techniques. This is used to determine the degree of semantic association between the two. Semantic similarity is the quantitative result of semantic matching, usually represented by a value between 0 and 1. A higher value indicates a closer semantic similarity. The highest semantic similarity is the maximum similarity value obtained after semantically matching the difference item with all fields in the set of fields of interest corresponding to a certain attribute tag.

[0110] The core function of this step is to semantically match the differences with the sets of interest corresponding to each attribute tag in the user profile, find the highest semantic similarity, and determine the degree of interest accordingly, thus achieving accurate prediction of user interests. This step follows the construction of the user profile, using the attribute tags and their corresponding sets of interest in the user profile as references. By comparing the differences with these sets semantically, it can overcome the limitations of superficial differences in field names and more accurately capture the user's potential interests. By selecting the highest semantic similarity as the basis for determining the degree of interest, it ensures that the highest correlation between the differences and a certain attribute feature of the user is fully considered, avoiding misjudgments of interest due to insufficient matching of a single attribute tag. This step works closely with the subsequent step of determining the visual emphasis level. Accurate interest provides a reliable basis for the division of visual emphasis levels, enabling the labeling of differences to truly match the user's interests and preferences. This solves the problem of low efficiency in user information acquisition caused by the lack of targeting in traditional labeling methods. Through the synergistic effect of each step, personalized and precise labeling of differences is achieved.

[0111] Semantic similarity is calculated using a word vector model. The system pre-trains word vectors on all fields in the detection report, using the Word2Vec model to convert each field into a fixed-dimensional vector representation, forming a field vector library. Similarly, for each attribute tag's set of fields of interest in the user profile, each field is converted into a word vector. When processing discrepancies, the field names of the discrepancies are extracted and converted into word vectors. Then, cosine similarity is calculated between these vectors and the word vectors of all fields in the set of fields of interest corresponding to each attribute tag, resulting in multiple similarity values. The maximum value is selected as the semantic similarity for that attribute tag. Finally, the highest value among all semantic similarities for all attribute tags is selected as the highest semantic similarity, which is directly used as the interest score. For example, if the difference item field is named "Benzene Series Content," and the user profile includes the attribute tag "Decoration Material Buyer," whose interest field set includes fields such as "Formaldehyde Content" and "VOC Content," then using word vector calculations, the cosine similarity between "Benzene Series Content" and "VOC Content" is 0.8, which is the highest value in this field set. The highest semantic similarity for other attribute tags is 0.6. Therefore, 0.8 is determined as the highest semantic similarity, meaning the interest level for this difference item is 0.8. The system iteratively calculates the semantic similarity between each difference item and each interest field set to determine the interest level for all difference items.

[0112] In one embodiment, the product page display method further includes: monitoring the batch number of the target product. When the batch number is updated, it is determined that the target product has been updated. It can be understood that a batch number is a unique code used to identify a product's production batch, typically composed of numbers, letters, or a combination thereof, containing information such as production time, production line, and work group, and can be used to trace the product's production source and quality status. Monitoring the batch number of the target product refers to the process by which the system, through preset technical means, obtains the current batch number information of the target product in real time or periodically and compares it with stored historical batch numbers. A batch number update refers to a state where the current batch number of the target product is inconsistent with the historical batch numbers stored in the system, indicating that the product has undergone a change in production batch. The core function of this embodiment is to determine whether the target product has been updated by monitoring batch number updates, providing triggering conditions for subsequent steps such as test report retrieval and difference judgment. The batch number is directly associated with the product's production batch; an update of the batch number usually means that the product may have undergone changes in raw materials, process adjustments, or quality inspection standards during the production process, which may lead to changes in the content of the product's test report. This embodiment uses the product batch number as the basis for determining product updates, enabling timely detection of production batch changes and ensuring the system initiates subsequent processing when a substantial change in the product may occur. The system monitors product batch numbers through a timed polling mechanism. At fixed time intervals (e.g., hourly), the system retrieves the current batch number of the target product from the product information database via an application programming interface (API) and compares it with the historical batch numbers of the same product stored locally. If they do not match, the system determines that the product batch number has been updated, and thus determines that the target product has been updated. For example, if the system stores the historical batch number of the target product as "20231005A01" and the timed current batch number is "20231006A02," a discrepancy is found, the system determines that the product batch number has been updated, and the target product has been updated. Simultaneously, the system updates the current batch number to the historical batch number for the next monitoring and comparison.

[0113] This application provides a product page display device, including: a retrieval module, used to retrieve the latest test report from a test report database based on the unique identifier of the target product when an update is detected; a judgment module, used to determine whether there is a difference between the latest test report and the current test report; a difference determination module, used to determine the difference items between the latest test report and the current test report when a difference exists; and an update module, used to update the display page of the target product according to the latest test report and mark the difference items in the latest test report.

[0114] Specific limitations regarding the product page display device can be found in the above-described limitations on the product page display method, and will not be repeated here. Each module in the aforementioned product page display device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module. It should be noted that the module division in this embodiment is illustrative and only represents a logical functional division; other division methods may be used in actual implementation.

[0115] This application provides a computer device including one or more processors and a memory storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, they perform the steps of the product page display method in any of the above embodiments.

[0116] Indicatively, such as Figure 6 As shown, Figure 6 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. (Refer to...) Figure 6 The computer device 600 includes a processing component 602, which further includes one or more processors, and memory resources represented by memory 601 for storing instructions, such as application programs, that can be executed by the processing component 602. The application programs stored in memory 601 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 602 is configured to execute instructions to perform the steps of the product page display method of any of the above embodiments.

[0117] This application provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the product page display method in any of the above embodiments.

[0118] 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 apparatus 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 apparatus. 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 apparatus that includes the element.

[0119] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0120] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for displaying product pages, characterized in that, The method comprises the following steps: When it is detected that the target commodity is updated, the latest detection report is retrieved from the detection report database according to the unique identifier of the target commodity; It is judged whether there is a difference between the latest detection report and the current detection report; If yes, the difference between the latest detection report and the current detection report is determined; The display page of the target commodity is updated according to the latest detection report; In the latest detection report, the difference items are marked, including: obtaining the user portrait of the target user; the target user is a user who is browsing the target commodity page; according to the user preference characteristics reflected in the user portrait, the interest degree of the target user to each difference item is predicted, and according to the user information processing ability characteristics reflected in the user portrait, the interest degree level division granularity is determined, and according to the user's sensitivity to different visual elements reflected in the user portrait, the difference item marking scheme is determined; the interest degree of each difference item is divided into levels according to the level division granularity to determine the visual emphasis level of each difference item; the higher the interest degree, the higher the visual emphasis level; the visual emphasis level of each difference item is marked by using the difference item marking scheme.

2. The merchandise page presentation method of claim 1, wherein, The judgment of whether there is a difference between the latest detection report and the current detection report comprises: A first field table is generated according to the current detection report; all first fields in the current detection report are included in the first field table; A second field table is generated according to the latest detection report; all second fields in the latest detection report are included in the second field table; Each of the first fields and the second fields is matched according to the field name; If there is no first field or second field that cannot be matched, it is determined that there is no difference when the field value change amplitude of each matched field pair is less than a first threshold, and it is determined that there is a difference when the field value change amplitude of each matched field pair is greater than the first threshold; If there is a first field and / or a second field that cannot be matched, it is determined that there is a difference.

3. The merchandise page presentation method of claim 2, wherein, The difference items include numerical variation items, added items and reduced items, and the determination of the difference items between the latest detection report and the current detection report comprises: The field value change amplitude of each matched field pair greater than the first threshold is determined as the numerical variation item; The first field that cannot be matched is determined as the reduced item; The second field that cannot be matched is determined as the added item.

4. The merchandise page presentation method of claim 3, wherein, Further comprising: In response to the target user's explanation request for the difference item, the difference item to be explained is determined as the target difference item; The target user is a user who is browsing the target commodity page; The explanation request prompt word is composed of the type label of the target difference item, the field name of the target difference item and the target commodity name; The explanation request prompt word is input into the explanation large model to obtain an explanation result; The explanation result is displayed to the target user.

5. The merchandise page presentation method of claim 4, wherein, The response to the target user's explanation request for the difference item comprises: If it is monitored that the target user performs a hovering operation on any one of the difference items, the explanation request for the difference item is generated.

6. The merchandise page presentation method of claim 1, wherein, The user portrait includes one or more attribute labels, each attribute label is configured with a set of interested fields, and the interest degree of the target user for each difference item is predicted according to the user portrait, including: The difference items are respectively matched with the set of interested fields corresponding to each attribute label in terms of semantics; and The interest degree is determined according to the highest semantic similarity in the semantic matching result.

7. The merchandise page presentation method of claim 1, wherein, Further comprising: Monitoring the batch number of the target commodity; When the batch number of the commodity is updated, it is determined that the target commodity is updated.

8. A merchandise page presentation device, characterized by, Comprising: A retrieval module configured to retrieve the latest detection report from a detection report database according to the unique identifier of the target commodity when it is detected that the target commodity is updated; A judgment module configured to determine whether there is a difference between the latest detection report and a current detection report; A difference determination module configured to determine the difference between the latest detection report and the current detection report when there is a difference between the latest detection report and the current detection report; An update module configured to update the display page of the target commodity according to the latest detection report and mark the difference items in the latest detection report, including: obtaining a user portrait of a target user; the target user is a user who is browsing the target commodity page; predicting the interest degree of the target user for each difference item according to the user preference features reflected in the user portrait, determining the interest degree grade division granularity according to the user information processing ability features reflected in the user portrait, and determining the difference item marking scheme according to the sensitivity of the user to different visual elements reflected in the user portrait; dividing the interest degree of each difference item into grades according to the grade division granularity to determine the visual emphasis level of each difference item; the higher the interest degree, the higher the visual emphasis level; and marking the visual emphasis level of each difference item using the difference item marking scheme.

9. A computer device, comprising: One or more processors and a memory are included, and the memory stores computer readable instructions, which are executed by the one or more processors to perform the steps of the commodity page display method according to any one of claims 1-7.

10. A storage medium, characterized by The storage medium stores computer readable instructions, which are executed by one or more processors to cause the one or more processors to perform the steps of the commodity page display method according to any one of claims 1-7.

Images