Commodity data resource analysis method and system based on digital intelligence system

Abstract

The present invention relates to the technical field of commodity data analysis, and in particular to a commodity data resource analysis method and system based on a digital intelligence system, the method comprising: obtaining product keywords and price ranges; matching to obtain a set of product models; judging relevance; taking a set of historical user evaluation information of a product model whose relevance to feature information associated with any other product model reaches a preset value as an interactive subset of the product model; calculating the number of good reviews and bad reviews for each product model; calculating a recommended value based on the number of good reviews and bad reviews for each product model, and the number of user evaluation information in the associated user evaluation information set; sorting each product model according to the recommended value, obtaining a product selection recommendation sequence and pushing it to the user. This application has the effect of matching corresponding competing products from the perspective of similar real selling points, providing operators with reference, grasping user trends, and improving the quality of operator product selection.

Description

Technical Field

[0001] The present invention relates to the technical field of commodity data analysis, and in particular to a commodity data resource analysis system and method based on a digital intelligence system. Background Art

[0002] Marketing is also known as marketing, marketing or marketing. The market is a category of commodity economy and a form of economic connection with commodity exchange as its content. For enterprises, the market is the starting point and destination of marketing activities. Marketing is both a function and an activity, process and system for organizations to create, communicate, disseminate and deliver customer value for the benefit of themselves and their stakeholders, and to bring economic value to customers, clients, partners and the entire society. It mainly refers to the process by which marketers conduct business activities and sales behaviors in the market.

[0003] At present, the existing e-commerce intelligent product selection system can only perform normal product selection operations when in use. It cannot conduct effective analysis and comparison based on the market and the company itself, and cannot make the selected products quickly adjusted according to market demand and the company's own environment, affecting economic benefits. Especially for small-scale operators and companies in the early stages of operation, the evaluation has a greater impact on the development of the store. Therefore, it is often necessary to carefully choose between multiple different product models during the initial product selection. Summary of the invention

[0004] In order to match corresponding competing products from the perspective of similar real selling points, provide reference for operators, grasp user trends, and improve the quality of product selection for operators, this application provides a commodity data resource analysis system and method based on a digital intelligence system.

[0005] The above-mentioned invention objective of the present application is achieved through the following technical solutions:

[0006] A commodity data resource analysis method based on a digital intelligence system, comprising:

[0007] Get product keywords and price ranges;

[0008] A product model set is obtained by matching the product keywords and price range in a market database, wherein the market database stores product information of the shopping mall platform, and the product information includes product models, historical selling prices, and associated historical user evaluation information sets;

[0009] Get the feature information associated with the product model;

[0010] Determine the relevance of each piece of historical user evaluation information in the historical user evaluation information set of each product model with the feature information associated with other product models;

[0011] The set of historical user evaluation information in the historical user evaluation information set of the product model whose correlation with the feature information associated with any other product model reaches a preset value is taken as the interaction subset of the product model;

[0012] Based on the historical user evaluation information in the interaction subset of each product model and the feature information associated with the remaining product models, the number of good reviews and the number of bad reviews for each product model are obtained;

[0013] The recommendation value is calculated based on the number of good reviews and bad reviews for each product model and the number of user review information in the associated user review information set;

[0014] Each product model is sorted according to the recommendation value to obtain a product selection recommendation sequence and push it to the user.

[0015] In a preferred example, the present application can be further configured as follows: based on the historical user evaluation information in the interaction subset of each product model and the feature information associated with the remaining product models, the number of good reviews and the number of bad reviews for each product model are obtained, including:

[0016] Input historical user evaluation information and feature information into the comparison model, and infer and judge whether each historical user evaluation information is good or bad;

[0017] Counting the number of historical user evaluation information with good reviews in the interaction subset of each product model to obtain the number of good reviews for each product model, and counting the number of historical user evaluation information with bad reviews in the interaction subset of each product model to obtain the number of bad reviews for each product model;

[0018] The good review indicates that the evaluation of the product model associated with the historical user evaluation information is better than that of the product model associated with the feature information in the historical user evaluation information, and the bad review indicates that the evaluation of the product model associated with the historical user evaluation information is worse than that of the product model associated with the feature information in the historical user evaluation information.

[0019] In a preferred example, the present application can be further configured as follows: based on the number of good reviews and bad reviews of each product model, and the number of user review information in the associated user review information set, a recommendation value is calculated, including:

[0020] The difference between the number of good reviews and the number of bad reviews for each product model is taken as the net merit of each product model;

[0021] The ratio of the net merit of each product model to the number of user evaluation information in the associated user evaluation information set is taken as the recommended value.

[0022] In a preferred example, the present application can be further configured as follows: the feature information includes the brand name, brand alias, product name and product alias of the product.

[0023] The second object of the invention is achieved by the following technical solutions:

[0024] The commodity data resource analysis system based on the digital intelligence system includes:

[0025] Keyword acquisition module, used to obtain product keywords and price ranges;

[0026] A matching module, used for matching a product model set in a market database according to the product keywords and price range, wherein the market database stores product information of the mall platform, and the product information includes product model, historical selling price and associated historical user evaluation information set;

[0027] A feature acquisition module is used to acquire feature information associated with a product model;

[0028] A judgment module, used to judge the relevance between each historical user evaluation information in the historical user evaluation information set of each product model and the characteristic information associated with other product models;

[0029] A subset generation module, used to take the set of historical user evaluation information of the product model whose correlation with the feature information associated with any other product model reaches a preset value as the interaction subset of the product model;

[0030] A comparative statistics module, used to obtain the number of good reviews and the number of bad reviews for each product model based on the historical user evaluation information in the interaction subset of each product model and the characteristic information associated with the remaining product models;

[0031] A calculation module, used to calculate the recommendation value based on the number of good reviews and the number of bad reviews for each product model and the number of user evaluation information in the associated user evaluation information set;

[0032] The sorting and recommendation module is used to sort each product model according to the recommendation value, obtain a product selection recommendation sequence and push it to the user.

[0033] In a preferred example, the present application can be further configured as follows: the comparison statistics module includes:

[0034] An inference unit, used to input historical user evaluation information and feature information into a comparison model, and infer and judge whether each historical user evaluation information is a good review or a bad review;

[0035] The statistical unit is used to count the number of historical user evaluation information with good reviews in the interaction subset of each product model to obtain the number of good reviews for each product model, and to count the number of historical user evaluation information with bad reviews in the interaction subset of each product model to obtain the number of bad reviews for each product model.

[0036] In a preferred example, the present application can be further configured as follows: the computing module includes:

[0037] A net merit calculation unit, used to take the difference between the number of good reviews and the number of bad reviews of each product model as the net merit of each product model;

[0038] The recommendation value calculation unit is used to take the ratio of the net merit of each product model to the number of user evaluation information in the associated user evaluation information set as the recommendation value.

[0039] In summary, the present application includes at least one of the following beneficial technical effects:

[0040] 1. After the user enters the keywords and price of the product they want to select, the system matches the product model set that meets the requirements in the database. Then, by evaluating the comparison of the product model with other product models in all the reviews, it can reflect the pros and cons of using multiple different product models. From this data comparison, it can match the corresponding competing products from the perspective of similar real selling points. Finally, the pros and cons comparison can be given to the operator for reference to grasp the user trend;

[0041] 2. The proportion of net merit in all reviews can be calculated to reflect the user's comparative evaluation between different product models. If the product models are sorted from high to low, the higher-ranked product models will be more popular with users. This can more accurately grasp user trends and improve the quality of product selection for operators. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flowchart of the implementation of the commodity data resource analysis method based on the digital intelligence system in one embodiment of the present application;

[0043] Figure 2 This is a module diagram of a commodity data resource analysis system based on a digital intelligence system in one embodiment of the present application. DETAILED DESCRIPTION

[0044] The following is a description of exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0045] It should be noted that the terms "first", "second", etc. in the present invention are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure.

[0046] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0047] Figure 1 is a flowchart of a method for analyzing commodity data resources based on a digital intelligence system in an embodiment of the present application. Figure 1 As shown, the commodity data resource analysis method based on the digital intelligence system includes:

[0048] S1. Obtain product keywords and price range.

[0049] Specifically, the platform provides a user interface and a search box for users to enter product keyword information, and provides price options. For example, users can enter or select a lower limit and an upper limit to determine the price range of the selected product. For example, a sweeping robot has a price range of 3,000 to 5,000 yuan.

[0050] S2. Obtain a product model set by matching the product keywords and price range in the market database.

[0051] Among them, the market database stores the product information of the mall platform products, and the product information includes product models, historical selling prices and related historical user evaluation information sets; the market database can be a purchased product database, or a self-built crawler can be used to crawl product information from specific websites through web crawler technology, that is, the above-mentioned product models, historical selling prices and related historical user evaluation information sets to build the underlying database, and obtain the market database through data cleaning, denoising, etc.

[0052] For example, if "sweeping robot, price range 3000-5000 yuan" is input, multiple sweeping robot product models in the price range are matched in the market database to form a product model set.

[0053] S3. Obtain feature information associated with the product model.

[0054] Specifically, the characteristic information includes the product's brand name, brand alias, product name and product alias. For example, the characteristic information of a certain product model of a sweeping robot includes: yy brand, yy brand product name P, brand product name product alias Q, brand product name product alias M, brand product name product alias N, etc.

[0055] Input the product model and feature information questions into the ChatGPT model. Feature information questions are such as "What are the similar terms of J3 Elf?", etc. The ChatGPT model outputs the feature information answer. The ChatGPT model is trained by a large amount of user interaction data. The user interaction data includes the product model and the corresponding user evaluation information. Specifically, the text data (user interaction data) is preprocessed first and converted into a vector representation that the model can process. The processing method can be word segmentation and vectorization; then a classifier is added on the basis of the ChatGPT model to divide the text data into different categories or labels. Here, a fully connected layer or a convolutional neural network can be used as a classifier; after the model is built, the model is trained using the aforementioned preprocessed labeled data. The training process includes forward propagation, loss calculation, back propagation and parameter update. In this embodiment, an optimization algorithm (such as SGD, Adam, etc.) is used to optimize the model. Finally, the trained ChatGPT model is evaluated using test data. The evaluation indicators include accuracy, precision, and recall to determine the performance and optimization direction of the model. After training, the ChatGPT model can screen, classify and summarize the content in user reviews. For example, if a comment is "Little Whale is so easy to use", multiple entries such as "Little Whale" and "So easy to use" are intercepted, and then the similarity between each of the multiple entries and the product model is calculated. If the similarity between the entry "Little Whale" and the product model entry reaches the preset value, the entry "Little Whale" is identified and classified as belonging to feature information.

[0056] The term similarity algorithm may adopt calculation methods such as cosine or jaccard, or may adopt a similarity search method in feature information, such as Locality Sensitive Hashing (LSH), Vector Space Model (VSM), clustering algorithm, Approximate Nearest Neighbor Search (ANNS), etc. It is understandable that due to the particularity of the solution of this application, the accuracy of feature information does not need to be very high to achieve the subsequent comparison effect.

[0057] S4. Determine the relevance of each piece of historical user evaluation information in the historical user evaluation information set of each product model and the characteristic information associated with other product models.

[0058] Specifically, the characteristic information includes multiple characteristic sub-information such as the product's brand name, brand alias, product name and product alias. The aforementioned "little whale" is a characteristic sub-information, and the calculation of relevance can be the same as the calculation of the aforementioned term similarity.

[0059] It can be understood that the historical user evaluation information and the multiple sub-information in the feature information are all subjected to correlation calculation to obtain sub-correlations, and the maximum value of the multiple sub-correlations is used as the correlation with the feature information.

[0060] S5. The set of historical user evaluation information of the product model whose correlation with the feature information associated with any other product model reaches a preset value is taken as the interaction subset of the product model.

[0061] That is, if the correlation between a certain historical user evaluation information and the characteristic information associated with any other product model in the historical user evaluation information set of the product model reaches a preset value, the historical user evaluation information is used as an element in the interaction subset of the product model.

[0062] S6. Obtain the number of good reviews and the number of bad reviews for each product model based on the historical user evaluation information in the interaction subset of each product model and the feature information associated with the remaining product models.

[0063] Among them, a good review indicates that the evaluation of the product model associated with the historical user evaluation information is better than the product model associated with the feature information in the historical user evaluation information, and a bad review indicates that the evaluation of the product model associated with the historical user evaluation information is worse than the product model associated with the feature information in the historical user evaluation information.

[0064] Specifically, it includes:

[0065] S61, inputting historical user evaluation information and feature information into a comparison model, and inferring and judging whether each historical user evaluation information is a good review or a bad review.

[0066] The comparison model is trained in the following way:

[0067] Each comment text sample in the comment text sample training set is labeled to mark each comment text as good or bad, and the good or bad review is associated with all or part of the information in the comment text sample; and a neural network is trained using the labeled comment text sample training set to obtain a comparison model.

[0068] Each review text sample contains one review information and feature information. For example, a sample is “Faster than the response of the neighboring xx brand; xx brand, xx brand’s product name A, xx brand’s product name’s product alias B, xx brand’s product name’s product alias C, xx brand’s product name’s product alias D”, where “Faster than the response of xx” is one review information, and “xx brand, xx brand’s product name A, xx brand’s product name’s product alias B, xx brand’s product name’s product alias C, xx brand’s product name’s product alias D” is feature information.

[0069] S62. Count the number of historical user evaluation information with good reviews in the interaction subset of each product model to obtain the number of good reviews for each product model, and count the number of historical user evaluation information with bad reviews in the interaction subset of each product model to obtain the number of bad reviews for each product model.

[0070] S7. Calculate the recommendation value based on the number of good reviews and bad reviews for each product model and the number of user evaluation information in the associated user evaluation information set.

[0071] Specifically include:

[0072] S71. The difference between the number of good reviews and the number of bad reviews for each product model is taken as the net merit of each product model.

[0073] S72. Take the ratio of the net merit of each product model to the number of user evaluation information in the associated user evaluation information set as the recommended value.

[0074] S8. Sort each product model according to the recommended value, obtain the recommended product selection sequence and push it to the user.

[0075] In this way, the proportion of net merit in all reviews can be calculated to reflect the user's comparative evaluation between different product models. If the product models are sorted from high to low, the higher-ranked product models will be more popular with users. This can more accurately grasp user trends and improve the quality of product selection for operators.

[0076] This application also provides a commodity data resource analysis system based on a digital intelligence system, referring to Figure 2, including: a keyword acquisition module, a matching module, a feature acquisition module, a judgment module, a subset generation module, a comparison statistics module, a calculation module and a ranking recommendation module, wherein the keyword acquisition module is used to obtain product keywords and price ranges; the matching module is used to match the product model set in the market database according to the product keywords and price ranges, and the market database stores the product information of the mall platform products, and the product information includes the product model, the historical selling price and the associated historical user evaluation information set; the feature acquisition module is used to obtain the feature information associated with the product model; The judgment module is used to judge the correlation between each historical user evaluation information in the historical user evaluation information set of each product model and the characteristic information associated with the remaining product models; the subset generation module is used to take the set of historical user evaluation information in the historical user evaluation information set of the product model whose correlation with the characteristic information associated with any other product model reaches a preset value as the interactive subset of the product model; the comparative statistics module is used to obtain the number of good reviews and the number of bad reviews of each product model based on the historical user evaluation information in the interactive subset of each product model and the characteristic information associated with the remaining product models; the calculation module is used to calculate the recommended value based on the number of good reviews and the number of bad reviews of each product model, and the number of user evaluation information in the associated user evaluation information set; the sorting recommendation module is used to sort each product model according to the recommended value, obtain the product selection recommendation sequence and push it to the user.

[0077] In one embodiment, the comparison statistics module includes an inference unit and a statistics unit. The inference unit is used to input historical user evaluation information and feature information into the comparison model, and infer and judge whether each historical user evaluation information is good or bad; the statistics unit is used to count the number of historical user evaluation information with good reviews in the interaction subset of each product model to obtain the number of good reviews for each product model, and count the number of historical user evaluation information with bad reviews in the interaction subset of each product model to obtain the number of bad reviews for each product model. The calculation module includes a net merit calculation unit and a recommendation value calculation unit. The net merit calculation unit is used to use the difference between the number of good reviews and the number of bad reviews of each product model as the net merit of each product model; the recommendation value calculation unit is used to use the ratio of the net merit of each product model to the number of user evaluation information in the associated user evaluation information set as the recommendation value.

[0078] For the specific definition of the commodity data resource analysis system based on the digital intelligence system, please refer to the definition of the commodity data resource analysis method based on the digital intelligence system above, which will not be repeated here. Each step of the above-mentioned commodity data resource analysis method based on the digital intelligence system can be implemented in whole or in part by software, hardware and their combination.

[0079] Various implementations of the systems and techniques described herein can be realized in digital electronic circuit systems, integrated circuit systems, dedicated ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0080] These computer programs (also referred to as programs, software, software applications, or code) include machine instructions for programmable processors and can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, device, and / or means (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0081] To provide interaction with a user, the systems and techniques described herein may be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0082] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0083] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this application can be executed in parallel, sequentially or in different orders, as long as the expected results of the technical solution disclosed in this application can be achieved, and this document is not limited here.

[0084] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A commodity data resource analysis method based on a digital intelligence system, characterized in that: include: Get product keywords and price ranges; A product model set is obtained by matching the product keywords and price range in a market database, wherein the market database stores product information of the shopping mall platform, and the product information includes product models, historical selling prices, and associated historical user evaluation information sets; Get the feature information associated with the product model; Determine the correlation between each historical user evaluation information in the historical user evaluation information set of each product model and the feature information associated with other product models, the correlation being obtained in the following manner: calculating the correlation between the historical user evaluation information and multiple feature sub-information in the feature information to obtain sub-correlations, and taking the maximum value of the multiple sub-correlations as the correlation with the feature information; The set of historical user evaluation information in the historical user evaluation information set of the product model whose correlation with the feature information associated with any other product model reaches a preset value is taken as the interaction subset of the product model; Based on the historical user evaluation information in the interaction subset of each product model and the feature information associated with the remaining product models, the number of good reviews and the number of bad reviews for each product model are obtained; The recommendation value is calculated based on the number of good reviews and bad reviews for each product model and the number of user review information in the associated user review information set; Each product model is sorted according to the recommendation value to obtain a product selection recommendation sequence and push it to the user.

2. The commodity data resource analysis method based on a digital intelligence system according to claim 1, characterized in that: Based on the historical user evaluation information in the interaction subset of each product model and the feature information associated with the remaining product models, the number of good reviews and the number of bad reviews for each product model are obtained, including: The historical user evaluation information and feature information are input into the comparison model to infer and judge whether each historical user evaluation information is good or bad.

3. The commodity data resource analysis method based on a digital intelligence system according to claim 2, characterized in that: Counting the number of historical user evaluation information with good reviews in the interaction subset of each product model to obtain the number of good reviews for each product model, and counting the number of historical user evaluation information with bad reviews in the interaction subset of each product model to obtain the number of bad reviews for each product model; The good review indicates that the evaluation of the product model associated with the historical user evaluation information is better than that of the product model associated with the feature information in the historical user evaluation information, and the bad review indicates that the evaluation of the product model associated with the historical user evaluation information is worse than that of the product model associated with the feature information in the historical user evaluation information.

4. The commodity data resource analysis method based on a digital intelligence system according to claim 1, characterized in that: The recommended value is calculated based on the number of good and bad reviews for each product model and the number of user review information in the associated user review information set, including: The difference between the number of good reviews and the number of bad reviews for each product model is taken as the net merit of each product model; The ratio of the net merit of each product model to the number of user evaluation information in the associated user evaluation information set is taken as the recommended value.

5. The commodity data resource analysis method based on a digital intelligence system according to claim 1, characterized in that: The characteristic information includes the brand name, brand alias, product name and product alias of the product.

6. The commodity data resource analysis system based on the digital intelligence system is characterized by: include: Keyword acquisition module, used to obtain product keywords and price ranges; A matching module, used for matching a product model set in a market database according to the product keywords and price range, wherein the market database stores product information of the mall platform, and the product information includes product model, historical selling price and associated historical user evaluation information set; A feature acquisition module is used to acquire feature information associated with a product model; A judgment module is used to judge the correlation between each historical user evaluation information in the historical user evaluation information set of each product model and the characteristic information associated with other product models, wherein the correlation is obtained by: calculating the correlation between the historical user evaluation information and multiple characteristic sub-information in the characteristic information to obtain sub-correlations, and taking the maximum value of the multiple sub-correlations as the correlation with the characteristic information; A subset generation module, used to take the set of historical user evaluation information of the product model whose correlation with the feature information associated with any other product model reaches a preset value as the interaction subset of the product model; A comparative statistics module, used to obtain the number of good reviews and the number of bad reviews for each product model based on the historical user evaluation information in the interaction subset of each product model and the characteristic information associated with the remaining product models; A calculation module, used to calculate the recommendation value based on the number of good reviews and bad reviews of each product model and the number of user evaluation information in the associated user evaluation information set; The sorting and recommendation module is used to sort each product model according to the recommendation value, obtain a product selection recommendation sequence and push it to the user.

7. The commodity data resource analysis system based on a digital intelligence system according to claim 6, characterized in that: The comparison statistics module includes: an inference unit, which is used to input historical user evaluation information and feature information into a comparison model, and infer and determine whether each historical user evaluation information is a good review or a bad review.

8. The commodity data resource analysis system based on digital intelligence system according to claim 7, characterized in that: The comparative statistics module includes: a statistical unit, which is used to count the number of historical user evaluation information with good reviews in the interaction subset of each product model to obtain the number of good reviews for each product model, and to count the number of historical user evaluation information with bad reviews in the interaction subset of each product model to obtain the number of bad reviews for each product model.

9. The commodity data resource analysis system based on a digital intelligence system according to claim 6, characterized in that: The calculation module includes: a net merit calculation unit, which is used to use the difference between the number of good reviews and the number of bad reviews of each product model as the net merit of each product model.

10. The commodity data resource analysis system based on digital intelligence system according to claim 9, characterized in that: The calculation module includes: a recommendation value calculation unit, which is used to take the ratio of the net merit of each product model to the number of user evaluation information in the associated user evaluation information set as the recommendation value.

Images