An e-commerce system commodity search sorting method and system

Abstract

The present disclosure relates to the technical field of electronic commerce, in particular to an e-commerce system commodity search sorting method and system. The method comprises the following steps: collecting a search request, and obtaining query content based on the search request; matching a plurality of commodities in a preset commodity library based on the query content, and calculating text relevance scores, commodity quality scores and label matching scores of the commodities; obtaining text relevance weights, commodity quality weights and label matching weights of the commodities, wherein the sum of the text relevance weights, the commodity quality weights and the label matching weights of each commodity is less than or equal to 1, and the text relevance weights, the commodity quality weights and the label matching weights correspond to the text relevance scores, the commodity quality scores and the label matching scores of the commodity respectively; calculating final search sorting scores of the commodities based on the text relevance scores, the commodity quality scores, the label matching scores, the text relevance weights, the commodity quality weights and the label matching weights; sorting all the commodities based on the final search sorting scores, generating a commodity list and displaying the commodity list.

Description

Technical Field

[0001] This disclosure relates to the field of e-commerce technology, and more specifically, to a product search and sorting method for an e-commerce system. Background Technology

[0002] Currently, most e-commerce platforms primarily use text relevance ranking methods based on keyword matching for product search ranking, or combine ranking based on a single dimension such as product sales volume and positive review rate. Some e-commerce systems use a multi-indicator weighted scoring method with fixed weights, such as calculating a comprehensive score based on indicators like product quality, tag matching, and text relevance according to preset proportions. However, these methods generally have the following drawbacks: (1) Search ranking relies on static weight configuration, which cannot flexibly adapt to business changes, resulting in rigid ranking results.

[0003] (2) The impact of dynamic combination of multi-dimensional indicators on search ranking is not considered, making it difficult to meet the search needs of different user groups or different scenarios.

[0004] (3) The lack of a dynamic adjustment mechanism for weights such as product quality, labels, and text relevance leads to inaccurate ranking results.

[0005] (4) The scoring model is not transparent, making it difficult to debug and optimize the system, which affects the iterative efficiency of the ranking strategy.

[0006] The aforementioned defects are generally not isolated, but are an inevitable result of the core design logic of existing technologies. The reason is that existing technologies are generally limited to the sorting approach of "static configuration + single / fixed combination dimension", and have not realized the decisive influence of "dynamic weight adaptation + multi-dimensional collaborative optimization" on the sorting effect, nor have they proposed any technical ideas that can solve the above-mentioned defects at the same time.

[0007] While some existing technologies mention "multi-index weighting," none of them involve dynamic adjustment of the weights, nor do they incorporate user behavior feedback for real-time adaptation. Therefore, they cannot fundamentally overcome the technical bottlenecks of rigid sorting and insufficient accuracy. Summary of the Invention

[0008] To address the problems existing in the prior art, this disclosure proposes a product search and sorting method and system for e-commerce systems to solve at least one of the aforementioned technical problems. The technical solution adopted in this disclosure is as follows: In a first aspect, this disclosure provides a method for product search and sorting in an e-commerce system, the method comprising: Collect the input search request, and parse the search request to obtain the query content; Based on the query content, several products are matched from the preset product database, and the text relevance score, product quality score and tag matching score of each product are calculated respectively. Obtain the text relevance weight, product quality weight, and tag matching weight for each product. The sum of the text relevance weight, product quality weight, and tag matching weight for each product is ≤1, and each product is assigned a text relevance score, product quality score, and tag matching score. Based on the text relevance score, product quality score, and tag matching score, as well as the text relevance weight, product quality weight, and tag matching weight, the final search ranking score for each product is calculated. All products are sorted based on the final search ranking score, and a sorted product list is generated and displayed.

[0009] Preferably, the calculation of the text relevance score for each product specifically includes: Extract the query keywords from the query content; Based on the query keywords, multiple fields (such as product name, category, attribute, description, etc.) in the preset product database are matched to obtain the matching score for each field; The text relevance score is obtained by weighted summation based on the matching scores of each field and the preset weights.

[0010] Preferably, the calculation of the product quality score for each product specifically includes: Retrieve multiple quality indicators (such as sales volume, sales revenue, positive review rate, certification level, inventory turnover rate, etc.) for the corresponding product from the preset quality indicator library; The product quality score is obtained by weighted summation based on the values ​​of each quality indicator and their preset weights.

[0011] Preferably, the calculation of the tag matching score for each product specifically includes: Based on a preset product category tag library, multiple product category tags corresponding to the query content are matched; Calculate the matching score between each of the product category tags and the query content; The tag matching score is obtained by weighted summation based on the matching score corresponding to each product category tag and the preset tag weight.

[0012] Preferably, before calculating the final search ranking score of each product, the values ​​of the text relevance weight, product quality weight, and tag matching weight can be adjusted in real time based on user behavior feedback, and then the final search ranking score of each product can be calculated separately, thereby adapting to the needs of different user groups or business scenarios.

[0013] Preferably, the step of sorting all products based on the final search ranking score, generating a sorted product list, and displaying it specifically includes: Based on the final search ranking score, all products are sorted in descending order to generate a product list in descending order. The product list is returned to the user's front-end interface for display.

[0014] A second aspect of this disclosure provides a product search and sorting system for an e-commerce system, the system comprising: The search request module is used to collect the input search request and parse the search request to obtain the query content. The product matching module is used to match several products from a preset product database based on the query content, and calculate the text relevance score, product quality score and tag matching score of each product respectively. The weight acquisition module is used to acquire the text relevance weight, product quality weight and tag matching weight of each product. The sum of the text relevance weight, product quality weight and tag matching weight of each product is ≤1 and corresponds to the text relevance score, product quality score and tag matching score of the product respectively. The scoring calculation module is used to calculate the final search ranking score of each product based on the text relevance score, product quality score, and tag matching score, as well as the text relevance weight, product quality weight, and tag matching weight. The list generation module is used to sort all products based on the final search ranking score, generate a sorted list of products, and display it.

[0015] In a third aspect, this disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the product search and sorting method for an e-commerce system as described above.

[0016] In a fourth aspect, this disclosure provides an electronic device including a processor and a memory, the processor being configured to execute a computer program stored in the memory to implement the e-commerce system product search and sorting method described above.

[0017] The beneficial effects of this disclosure are as follows: This disclosure provides a product search ranking method and system for e-commerce systems. By introducing multi-dimensional dynamic weight configuration and real-time calculation mechanisms, a flexible, interpretable, and highly configurable product search ranking method and system is constructed. This disclosure fully considers text relevance, product quality, and product tag categories, and dynamically calculates and obtains text relevance weights, product quality weights, and tag matching weights respectively, thus avoiding static weight configuration. Furthermore, it dynamically calculates weight values ​​from multiple dimensions to influence the final search ranking score and product ranking, covering product quality, tags, text relevance, etc., avoiding a one-sided arrangement and display of the product list, creating a clearer and more explicit calculation model, and making it more conducive to iteratively matching and displaying products based on user product search requests, thereby improving the recommendation efficiency and accuracy of the e-commerce system.

[0018] This disclosure can be flexibly configured into various e-commerce systems, seamlessly integrated into existing e-commerce systems, or used as a functional module to upgrade e-commerce systems, exhibiting excellent scalability and configurability. In addition to dynamically calculating and obtaining text relevance weights, product quality weights, and tag matching weights, this disclosure can also dynamically adjust these weights based on user behavior feedback. This comprehensively considers text relevance, product quality, product tag categories, and user behavior feedback, further adjusting the weights around user behavior. This makes product recommendations more aligned with user preferences and behaviors, further improving recommendation accuracy and avoiding the technical shortcomings of only considering weights without considering user behavior.

[0019] Compared with the prior art, this disclosure has the following advantages: (1) Significantly improves the accuracy of search results and user satisfaction. This disclosure, by comprehensively considering three core dimensions—text relevance, product quality, and tag matching—and introducing a dynamic weighting mechanism, can more accurately capture the complex relationship between user search intent and products. This overcomes the technical problem that traditional search ranking methods often rely on a single dimension or static weight, making it difficult to fully reflect the true value of products and user intent. For example, during peak promotional periods, the system can automatically increase the weight of product quality to ensure that best-selling products are displayed first; while when users conduct precise keyword searches, the weight of text relevance is enhanced to improve search accuracy. This multi-dimensional, adaptive ranking strategy can effectively reduce the exposure of irrelevant products, significantly improve the efficiency of users finding target products, and thus greatly improve user experience and shopping satisfaction.

[0020] (2) Supports flexible configuration and rapid response of business strategies, adapting to diversified operational scenarios. This disclosure designs text relevance weight, product quality weight, and tag matching weight as parameters that can be dynamically loaded from the configuration center. This allows for rapid response to changes in business needs without modifying code or redeploying, demonstrating high configurability. Operations personnel can flexibly adjust the weight allocation of each dimension according to different marketing activities (such as Double Eleven promotions, brand days), user segments (such as new users, high-value users), or product categories (such as clothing, electronic products), making implementation more convenient and flexible. For example, for new product launches, tag matching weight can be temporarily increased to quickly drive traffic; for clearance products, the weight of quality indicators such as sales volume and price can be increased. This flexibility makes search ranking no longer a rigid technical function, but an important strategic tool supporting business operations.

[0021] (3) It provides a transparent and interpretable scoring mechanism, greatly reducing the cost of system debugging and optimization. This disclosure, through a clear three-level weighted calculation model (field level → dimension level → global level) and a clear calculation formula model, ensures that the final search ranking score of each product is fully traceable. Developers can clearly see whether the poor ranking of certain products is due to insufficient text relevance matching, low product quality scores, or unreasonable tag matching weight settings, thus enabling them to quickly locate problems and perform targeted optimizations, demonstrating good interpretability. This interpretability not only accelerates the daily troubleshooting and debugging process but also provides a reliable data foundation and logical basis for the long-term iterative optimization of the algorithm model, fundamentally reducing the system's maintenance costs and iteration risks. Attached Figure Description

[0022] The accompanying drawings, which form part of this application, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0023] Figure 1 This is a flowchart of a product search and sorting method for an e-commerce system according to Embodiment 1 of this disclosure.

[0024] Figure 2 This is an architecture diagram of an e-commerce system product search and sorting system according to Embodiment 2 of this disclosure. Detailed Implementation

[0025] The present disclosure will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the present application can be combined with each other.

[0026] The following detailed descriptions are exemplary and intended to provide further detailed explanation of this disclosure. Unless otherwise specified, all technical terms used in this disclosure have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this disclosure.

[0027] Example 1: like Figure 1 As shown, this disclosure provides a product search and sorting method for an e-commerce system, the method including steps S100 to S500.

[0028] Step S100: Collect the input search request and parse the query content based on the search request.

[0029] Step S200: Based on the query content, match several products from the preset product database, and calculate the text relevance score, product quality score and tag matching score for each product.

[0030] Furthermore, the calculation of the text relevance score for each product specifically includes: Step S211: Extract query keywords from the query content; Step S212: Match multiple fields (such as product name, category, attribute, description, etc.) in the preset product database based on the query keywords to obtain the matching score of each field; Step S213: Based on the matching scores and preset weights of each field, the text relevance score is obtained by weighted summation.

[0031] Furthermore, the formula for calculating the text relevance score through weighted summation can be as follows: TextRelevance = Σ(FieldScore i × Weight i (1) In equation (1), i represents the field number; FieldScore i This represents the matching score for the corresponding field i; Weight i This represents the preset weight of the corresponding field i; TextRelevance represents the text relevance score.

[0032] In this disclosure, the calculation method for text relevance score is not the simple addition of keyword matching scores as in existing technologies. Instead, it achieves a refined evaluation of text relevance by separately matching and weighting multiple core fields of the product. This allows for a more accurate capture of the correlation between the query content and the product, which has significant technical advantages compared to existing technologies such as "single field matching" or "undifferentiated field weighting". Furthermore, by combining it with a subsequent dynamic weighting mechanism, the accuracy of recommendations is greatly improved.

[0033] Furthermore, the calculation of the product quality score for each product specifically includes: Step S221: Obtain multiple quality indicators (such as sales volume, sales revenue, positive review rate, certification level, inventory turnover rate, etc.) of the corresponding product from the preset quality indicator library; Step S222: Based on the values ​​of each quality indicator and the preset weights, the product quality score is obtained by weighted summation.

[0034] Furthermore, the formula for calculating the product quality score through weighted summation can be as follows: ProductQuality = Σ(Metric j × λ j (2) In equation (2), j represents the number of the quality index; Metric j This represents the value of quality indicator j; λ j The preset weights represent the quality index j; ProductQuality represents the quality score of the product.

[0035] It is important to note that the product quality indicators selected in this disclosure cover the core dimensions of the entire product operation process, rather than just the single indicators such as sales volume and positive review rate used in existing technologies. Furthermore, the preset weights of each product quality indicator can be flexibly configured according to business scenarios, laying the foundation for subsequent dynamic weight adjustments. Through this multi-dimensional selection and combination of quality indicators, the true quality of products can be reflected more comprehensively and objectively, avoiding the ranking bias caused by the single quality indicator in existing technologies.

[0036] Furthermore, the calculation of the tag matching score for each product specifically includes: Step S231: Match multiple product category tags corresponding to the query content based on the preset product category tag library; Step S232: Calculate the matching score between each of the product category tags and the query content; Step S233: Based on the matching score corresponding to each product category label and the preset label weight, the label matching score is calculated by weighted summation.

[0037] Furthermore, the formula for calculating the tag matching score through weighted summation is as follows: TagScore = Σ(Tag k × ω k (3) In equation (3), k represents the product category label number; Tag k The matching score represents the product category tag k; ω k The preset label weight representing the product category label k; TagScore represents the tag matching score.

[0038] In practice, this disclosure can match multiple product category tags to the query content, and calculate the tag matching score of each product by weighted summation of the matching score corresponding to the product category tag and the preset tag weight. This can overcome the limitation of "single tag matching" in the prior art and accurately capture the potential user needs behind the query content (for example, when a user searches for "summer outfits", multiple product category tags such as "short sleeve", "shorts", and "casual style" can be matched to comprehensively evaluate the product suitability). The product category tag matching logic forms a synergistic mechanism with text relevance and product quality score to jointly improve the ranking accuracy.

[0039] Step S300: Obtain the text relevance weight, product quality weight, and tag matching weight of each product. The sum of the text relevance weight, product quality weight, and tag matching weight of each product is ≤1, and each product is assigned a text relevance score, product quality score, and tag matching score.

[0040] It is understood that the text relevance weight, product quality weight, and tag matching weight can reflect the relevant weight values ​​in e-commerce product recommendations from multiple dimensions, and the sum of the three is <= 1. This allows for flexible configuration in the implementation of this disclosure, such as introducing more other weights to dynamically adjust product recommendations from more dimensions, or using only the three weights: text relevance weight, product quality weight, and tag matching weight. The text relevance weight, product quality weight, and tag matching weight of each product can be preset, read from a configuration file, dynamically calculated by referring to the text relevance score, product quality score, and tag matching score, or obtained or determined by referring to existing technologies.

[0041] Step S400: Based on the text relevance score, product quality score, and tag matching score, as well as the text relevance weight, product quality weight, and tag matching weight, calculate the final search ranking score for each product.

[0042] Furthermore, the formula for calculating the final search ranking score of each product can be as follows: FinalScore = (TextRelevance × Wt) + (ProductQuality × Wp) +(TagScore × Ws); (4) In equation (4), TextRelevance represents the text relevance score, and Wt represents the text relevance weight; ProductQuality represents the product quality score, and Wp represents the product quality weight. TagScore represents the tag matching score, and Ws represents the tag matching weight. FinalScore represents the final search ranking score.

[0043] It should be noted that the final search ranking score of each product is calculated based on equation (4). For example, when calculating the final search ranking score of a certain product, the text relevance score, text relevance weight, product quality score, tag matching weight, tag matching score, and tag matching weight are the parameters corresponding to that product.

[0044] Furthermore, before calculating the final search ranking score, the text relevance score, product quality score, and tag matching score of each product can be normalized to ensure that the scores of each dimension are on the same order of magnitude and avoid ranking deviations caused by inconsistent units.

[0045] Furthermore, before calculating the final search ranking score for each product, the values ​​of the text relevance weight, product quality weight, and tag matching weight can be adjusted in real time based on user behavior feedback, and then the final search ranking score for each product can be calculated separately, thereby adapting to the needs of different user groups or business scenarios.

[0046] Furthermore, the user behavior feedback includes at least one of click-through rate, conversion rate, browsing dwell time, product reviews, likes, and shares. In specific implementation, weight values ​​that should be adjusted can be set for each type of user behavior feedback, thereby dynamically adjusting the weights based on different user behavior feedback and calculating the final search ranking score, realizing the change of the product list arrangement based on user feedback behavior.

[0047] It is important to note that existing technologies lack a solution capable of dynamically adjusting multi-dimensional weights based on real-time user behavior feedback. The weights in existing technologies are all preset fixed values, failing to adapt to the personalized and real-time changes in user needs. This disclosure addresses this by deeply binding user behavior feedback (click-through rate, conversion rate, etc.) with weight adjustments, enabling dynamic weight adjustments based on user behavior feedback. This achieves self-iterative optimization of the ranking strategy, ensuring that the ranking results continuously align with user preferences and changes in business scenarios, thus overcoming the drawbacks of static design logic in existing technologies.

[0048] Step S500: Sort all products based on the final search ranking score, generate a sorted product list and display it.

[0049] Further, step S500, which involves sorting all products based on the final search ranking score, generating a sorted product list, and displaying it, specifically includes: Step S501: Sort all products in descending order based on the final search ranking score to generate a product list in descending order; Step S502: Return the product list to the user's front-end interface for display.

[0050] It is important to emphasize that the technical concept of the product search ranking method in the e-commerce system is not a simple improvement or parameter adjustment of existing technologies. Rather, it is based on a deep understanding of the core needs of existing e-commerce search ranking and proposes a brand-new "multi-dimensional dynamic collaborative ranking" technical framework. Its core innovation lies in breaking the limitations of static weights in existing technologies and constructing a closed-loop ranking mechanism of "accurate score calculation + dynamic weight adaptation + user feedback iteration". This technical concept can significantly improve the accuracy, flexibility and adaptability of search ranking, greatly improve user experience and business operation efficiency, and represents a significant improvement over existing technologies, fully meeting the requirements for technical contributions in creative responses.

[0051] Example 2: like Figure 2 As shown, Embodiment 2 of this disclosure provides a product search and sorting system for an e-commerce system, the system comprising: The search request module 100 is used to collect the input search request and parse the search request to obtain the query content. The product matching module 200 is used to match several products from a preset product database based on the query content, and calculate the text relevance score, product quality score and tag matching score of each product respectively. The weight acquisition module 300 is used to acquire the text relevance weight, product quality weight and tag matching weight of each product. The sum of the text relevance weight, product quality weight and tag matching weight of each product is ≤1 and corresponds to the text relevance score, product quality score and tag matching score of the product respectively. The scoring calculation module 400 is used to calculate the final search ranking score of each product based on the text relevance score, product quality score, and tag matching score, as well as the text relevance weight, product quality weight, and tag matching weight. The list generation module 500 is used to sort all products based on the final search ranking score, generate a sorted product list, and display it.

[0052] The search request module 100, product matching module 200, weight acquisition module 300, score calculation module 400, and list generation module 500 described in Example 2 correspond to steps S100, S200, S300, S400, and S500, respectively. It is worth noting that the system described in Example 2 is merely one implementation of the product search and ranking method for the e-commerce system, and does not imply that the product search and ranking method for the e-commerce system described in Example 1 must depend on the system described in Example 2.

[0053] In implementation, the modules of the system described in Example 2 can form a closely coordinated closed loop. In particular, the weight acquisition module 300 can be configured to adjust the weight values ​​in real time based on user behavior feedback, and the score calculation module 400 can accurately perform multi-dimensional weighted calculations. This constitutes a fundamental difference from the "single-function module splicing" system in the prior art. The design logic of the system described in Example 2 is highly consistent with the technical concept of the e-commerce system product search and ranking method described in Example 1. It also has outstanding substantive features and significant progress, which can effectively support the implementation of the above-mentioned creative method and solve the technical problems of rigid ranking and insufficient accuracy in existing systems. The innovation of its system architecture is beyond the reach of existing technologies.

[0054] Example 3: Embodiment 3 of this disclosure provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the product search and sorting method of the e-commerce system as described in Embodiment 1.

[0055] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program modules, or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory or other memory technologies, CD-ROM (Compact Disc Read-Only Memory), DVD or other optical disc storage, cartridges, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer.

[0056] Example 4: Embodiment 4 of this disclosure provides an electronic device, which includes a processor and a memory. The processor is used to execute a computer program stored in the memory to implement the e-commerce system product search and sorting method described in Embodiment 1.

[0057] Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, and read-only memory (ROM).

[0058] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied 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.

[0059] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. 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 apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine 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.

[0060] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function 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.

[0061] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable 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.

[0062] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0063] In summary, the product search ranking method and system provided in embodiments 1-4 of this disclosure, by introducing multi-dimensional dynamic weight configuration and real-time calculation mechanism, constructs a flexible, interpretable, and highly configurable product search ranking method and system. This disclosure fully considers text relevance, product quality, and product tag categories, and dynamically calculates and obtains text relevance weight, product quality weight, and tag matching weight respectively, thereby avoiding static weight configuration. Furthermore, it dynamically calculates weight values ​​from multiple dimensions to influence the final search ranking score and product ranking, covering product quality, tags, text relevance, etc., avoiding a one-sided arrangement and display of the product list, creating a clearer and more explicit calculation model, and is more conducive to iteratively matching and displaying products based on user product search requests, thus improving the recommendation efficiency and accuracy of the e-commerce system.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure and not to limit them. Although this disclosure has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of this disclosure. Any modifications or equivalent substitutions that do not depart from the spirit and scope of this disclosure should be covered within the protection scope of the claims of this disclosure.

Claims

1. A product search and sorting method for an e-commerce system, characterized in that, The method includes: Collect the input search request, and parse the search request to obtain the query content; Based on the query content, several products are matched from the preset product database, and the text relevance score, product quality score and tag matching score of each product are calculated respectively. Obtain the text relevance weight, product quality weight, and tag matching weight for each product. The sum of the text relevance weight, product quality weight, and tag matching weight for each product is ≤1, and each product is assigned a text relevance score, product quality score, and tag matching score. Based on the text relevance score, product quality score, and tag matching score, as well as the text relevance weight, product quality weight, and tag matching weight, the final search ranking score for each product is calculated. All products are sorted based on the final search ranking score, and a sorted product list is generated and displayed.

2. The method according to claim 1, characterized in that, The calculation yields a text relevance score for each product, specifically including: Extract the query keywords from the query content; Based on the query keywords, multiple fields in the preset product database are matched to obtain the matching score for each field; The text relevance score is obtained by weighted summation based on the matching scores of each field and the preset weights.

3. The method according to claim 1, characterized in that, The calculation yields a product quality score for each item, specifically including: Retrieve multiple quality indicators for the corresponding product from the preset quality indicator library; The product quality score is obtained by weighted summation based on the values ​​of each quality indicator and their preset weights.

4. The method according to claim 1, characterized in that, The calculation of the tag matching score for each product specifically includes: Based on a preset product category tag library, multiple product category tags corresponding to the query content are matched; Calculate the matching score between each of the product category tags and the query content; The tag matching score is obtained by weighted summation based on the matching score corresponding to each product category tag and the preset tag weight.

5. The method according to claim 1, characterized in that, The formula for calculating the final search ranking score of each product is as follows: FinalScore = (TextRelevance × Wt) + (ProductQuality × Wp) + (TagScore × Ws); (4) In equation (4), TextRelevance represents the text relevance score, and Wt represents the text relevance weight; ProductQuality represents the product quality score, and Wp represents the product quality weight. TagScore represents the tag matching score, and Ws represents the tag matching weight. FinalScore represents the final search ranking score.

6. The method according to claim 1, characterized in that, Before calculating the final search ranking score for each product, the values ​​of the text relevance weight, product quality weight, and tag matching weight are adjusted in real time based on user behavior feedback. Then, the final search ranking score for each product is calculated separately to adapt to the needs of different user groups or business scenarios. The user behavior feedback includes at least one of the following: click-through rate, conversion rate, browsing time, product reviews, likes, and shares.

7. The method according to claim 1, characterized in that, The process of sorting all products based on the final search ranking score, generating a sorted product list, and displaying it specifically includes: Based on the final search ranking score, all products are sorted in descending order to generate a product list in descending order. The product list is returned to the user's front-end interface for display.

8. A product search and sorting system for an e-commerce system, characterized in that, The system includes: The search request module is used to collect the input search request and parse the search request to obtain the query content. The product matching module is used to match several products from a preset product database based on the query content, and calculate the text relevance score, product quality score and tag matching score of each product respectively. The weight acquisition module is used to acquire the text relevance weight, product quality weight and tag matching weight of each product. The sum of the text relevance weight, product quality weight and tag matching weight of each product is ≤1 and corresponds to the text relevance score, product quality score and tag matching score of the product respectively. The scoring calculation module is used to calculate the final search ranking score of each product based on the text relevance score, product quality score, and tag matching score, as well as the text relevance weight, product quality weight, and tag matching weight. The list generation module is used to sort all products based on the final search ranking score, generate a sorted list of products, and display it.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the product search and sorting method of the e-commerce system according to any one of claims 1-7.

10. An electronic device comprising a processor and a memory, characterized in that, The processor is used to execute the computer program stored in the memory to implement the e-commerce system product search and sorting method according to any one of claims 1-7.

Images