Commodity management method and device, computer equipment and storage medium

Abstract

The invention relates to a commodity management method and device, computer equipment and a storage medium. The method comprises the steps of obtaining historical commodity browsing records of a user and a target retrieval condition, and querying the historical commodity browsing records of the user according to the target retrieval condition to obtain a commodity display list; wherein the commodity display list comprises a plurality of target commodities, processing the displayed commodities according to a preset processing rule, and generating a commodity display list; the preset processing rule comprises one or a combination of more of a season processing rule, a visibility processing rule, a user preference processing rule, an associated commodity processing rule, a commodity evaluation processing rule and the like; and selecting any one or more commodities in the historical commodity browsing records of the user, judging the display state of the selected commodity, and displaying a preset processing rule corresponding to the selected commodity. By adopting the method, commodity management can be transparent, and the user experience can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of commodity management, and in particular to a commodity management method, apparatus, computer equipment and storage medium. Background Art

[0002] As the Internet e-commerce industry continues to develop, the scenarios for product search and recommendation are becoming more and more diverse. Online shopping means that users search and browse products on the Internet e-commerce mall, then place an order and pay for the purchase, and the manufacturer then delivers the goods to their door through a courier company.

[0003] In the prior art, the product display list displayed after the user's search is formed by filtering and re-sorting the original browsed products after applying multiple processing rules. Therefore, the product display list presented to the user will be different from the original product browsing record. Finally, the product display list will be presented to the user. In the process of forming the product display list, the application of multiple processing rules is usually performed by logical calculation in the background, and the application of multiple processing rules will not be presented to the user. In this way, the user cannot clearly know which application processing rule is applied, nor can he judge whether the applied processing rule is reasonable. Summary of the invention

[0004] Based on this, it is necessary to provide a product management method, device, computer equipment and storage medium to improve user experience in response to the above technical problems.

[0005] In order to solve the above technical problems, in a first aspect, a commodity management method is provided, the method comprising:

[0006] Obtaining the user's historical product browsing records and target search conditions, querying the user's historical product browsing records according to the target search conditions, and obtaining a product display list; wherein the product display list includes multiple target products, and the displayed products are processed according to preset processing rules to generate a product display list; the preset processing rules include a combination of one or more of the following rules: season processing rules, visibility processing rules, user preference processing rules, related product processing rules, product evaluation processing rules, etc.;

[0007] Select any one or several products in the user's historical product browsing records, determine the display status of the selected products, and display the preset processing rules corresponding to the selected products.

[0008] In one embodiment, the preset processing rule includes a season processing rule, and the method further includes: obtaining user's regional information, obtaining weather information within a preset time range corresponding to the user's regional information; performing feature extraction on the weather information within the preset time range corresponding to the user's regional information to obtain a first feature vector; the weather information includes temperature, humidity, wind speed, visibility, etc.;

[0009] Acquire seasonal attribute information of commodities in the commodity pool, perform feature extraction on the seasonal attribute information of commodities in the commodity pool, and obtain a second feature vector;

[0010] Calculate the similarity between the first eigenvector and the second eigenvector;

[0011] When the similarity is higher than a preset threshold, it is determined that the seasonal attribute information of the product in the product pool matches the weather information within a preset time range corresponding to the user's regional information;

[0012] The products in the product pool that match the weather information within a preset time range corresponding to the user's regional information are displayed to the user.

[0013] In one embodiment, the preset processing rule includes a visibility processing rule, and the method further includes:

[0014] Get the user's regional information;

[0015] Obtain the product inventory information of the product corresponding to the target search condition;

[0016] Perform intersection operation on the region information and product inventory information to determine whether the product corresponding to the target search condition is in stock at the user's location;

[0017] Filter out-of-stock products in the user's location from historical product browsing records.

[0018] In one embodiment, the preset processing rule includes a user preference processing rule, and the method further includes:

[0019] Obtain the user's historical product browsing records and the user's historical product purchase records;

[0020] Determine the contribution of each user's historical product browsing record for each category of products based on the similarity between the user's historical product browsing record and the user's historical product purchase record;

[0021] Determine the user's purchasing habits based on the number of historical browsing between adjacent purchase records of each category of goods;

[0022] The contribution of the user's historical product browsing records on each e-commerce platform is used as the weight, and the sum of the contribution of the user's historical product browsing records corresponding to each category of products in all historical browsing records between the most recent historical purchase record and the previous historical purchase record on each e-commerce platform is weighted to obtain the user's purchase preference.

[0023] In one embodiment, the preset processing rule includes an associated commodity processing rule, and the method further includes:

[0024] Obtain the user's historical commodity purchase records, which include the user's historical behavior data;

[0025] According to the access history data and purchase history data in the user's historical behavior data, association rules or collaborative filtering algorithms are used to calculate the associated products of the products involved in the historical behavior data, calculate the correlation between the products involved in the historical behavior data and their associated products, select a preset number of products with the highest correlation as the associated product set of the products involved in the historical behavior data, and display them in the product display list.

[0026] In one embodiment, the method further comprises:

[0027] Obtain the user's historical behavior data, and determine the corresponding incremental score of the user's visit to each product based on the number of visits to each product in the user's historical behavior data;

[0028] Obtain a product browsing list in the user's historical product browsing record, and calculate the intermediate change value required for similarity calculation between the product corresponding to the user's historical behavior data and each product in the product browsing list in the user's historical product browsing record based on the incremental score;

[0029] Update the similarity value in the corresponding element in the product similarity list according to the intermediate change value, where the product similarity list is used to store the similarity values ​​between two product objects in the product browsing list in the user's historical product browsing record;

[0030] In response to receiving the target search condition, the corresponding product in the product browsing list in the user's historical product browsing record corresponding to the target search condition is obtained, the product similarity list is queried, and the products whose similarity values ​​meet the preset conditions are displayed in the product display list.

[0031] In one embodiment, the preset processing rules include product evaluation processing rules, and the method further includes:

[0032] Get the historical product reviews of the products in the user's historical product browsing history;

[0033] Analyze historical product reviews through semantic analysis text, extract emotional sentences and emotional images that carry emotional information;

[0034] Score the emotional sentences and emotional images respectively, and sort the total scores of the emotional sentences and emotional images by numerical value through the evaluation weight model; form product content labels according to the numerical order of the total scores of the emotional sentences and emotional images, and display the products according to the product content labels;

[0035] Among them, scoring emotional sentences and emotional images includes:

[0036] By identifying the emotional confidence of emotional sentences, emotional sentences are classified into positive, negative, and neutral emotional categories, and emotional sentences of different emotional categories are scored;

[0037] Extracting the first evaluation feature vector and the second evaluation feature vector corresponding to the emotional image and the store recommendation image respectively;

[0038] Calculating the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector;

[0039] The emotional image is scored according to the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector.

[0040] In order to solve the above technical problems, in a second aspect, a commodity management device is provided, the device comprising:

[0041] The acquisition module is used to obtain the user's historical product browsing records and target search conditions, query the user's historical product browsing records according to the target search conditions, and obtain a product display list; wherein the product display list includes multiple target products, and the display products are processed according to the preset processing rules to generate a product display list;

[0042] The display module is used to select any one or several products in the user's historical product browsing records, determine the display status of the selected products, and display the preset processing rules corresponding to the selected products.

[0043] In order to solve the above technical problem, a third aspect provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: the processor implements the steps of the above first aspect method when executing the computer program.

[0044] In order to solve the above technical problems, in a fourth aspect, the present application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method of the above first aspect are implemented.

[0045] Different from the prior art, the present application provides a product management method, device, computer equipment and storage medium, the method comprising: obtaining the user's historical product browsing records and target retrieval conditions, querying the user's historical product browsing records according to the target retrieval conditions, and obtaining a product display list; wherein the product display list includes multiple target products, and the displayed products are processed according to the preset processing rules to generate a product display list; selecting any one or several products in the user's historical product browsing records, determining the display status of the selected products, and displaying the preset processing rules corresponding to the selected products. In this way, the preset processing rules corresponding to the products can be clearly viewed, making product management transparent, which is conducive to improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of a commodity management method in one embodiment;

[0047] Figure 2 is a structural block diagram of a commodity management device in one embodiment;

[0048] Figure 3 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0050] In one embodiment, Figure 1 As shown, a commodity management method is provided, which should specifically include the following steps:

[0051] Step 101, obtain the user's historical product browsing records and target search conditions, query the user's historical product browsing records according to the target search conditions, and obtain a product display list; wherein the product display list includes multiple target products, and the display products are processed according to preset processing rules to generate a product display list.

[0052] For any e-commerce platform, when users operate products on it, different record information will be generated through the operation, including each user's historical product purchase record, user's historical product browsing record, etc. The e-commerce platform will store these record information in a secure database and back up the record information. Users can use any one or a combination of time, product category, user ID, device number, home page function display entrance, home page schedule flow, schedule landing page, etc. as the target search condition to query the user's historical product browsing record and other record information to obtain a product display list. Among them, the product display list includes multiple target products, and the displayed products can be processed according to the preset processing rules to generate a product display list.

[0053] The preset processing rules may include a combination of one or more rules such as season processing rules, visibility processing rules, user preference processing rules, associated product processing rules, and product evaluation processing rules. The user's regional information is obtained, and the weather information within the preset time range corresponding to the user's regional information is obtained; the weather information within the preset time range corresponding to the user's regional information is feature extracted to obtain a first feature vector; the weather information includes temperature, humidity, wind force, visibility, etc.; the seasonal attribute information of the products in the product pool is obtained, and the seasonal attribute information of the products in the product pool is feature extracted to obtain a second feature vector; the similarity between the first feature vector and the second feature vector is calculated; when the similarity is higher than the preset threshold, it is determined that the seasonal attribute information of the products in the product pool matches the weather information within the preset time range corresponding to the user's regional information; and the products in the product pool that match the weather information within the preset time range corresponding to the user's regional information are displayed to the user. Optionally, the occluded image can be feature extracted by a feature extraction model to obtain a first feature vector and a second feature vector. The feature extraction model can be RNN, CNN, or resnet, etc., and its model type is not limited. By performing feature extraction on the weather information within the preset time range corresponding to the user's geographical information and the seasonal attribute information of the products in the product pool, after obtaining the corresponding first feature vector and the second feature vector, the similarity between the first feature vector and the second feature vector can be determined, that is, the features extracted from the weather information within the preset time range corresponding to the user's geographical information and the seasonal attribute information of the products in the product pool are compared.

[0054] The spatial distance between the first eigenvector and the second eigenvector can be calculated, and the calculated spatial distance can represent the similarity between the first eigenvector and the second eigenvector. In this embodiment, if the spatial distance between the first eigenvector and the second eigenvector is larger, it means that the similarity between the target facial image and the occluded image is lower. Thus, in this embodiment, if the spatial distance between the first eigenvector and the second eigenvector is less than a preset threshold, it can represent that the similarity between the first eigenvector and the second eigenvector is higher than a preset level; conversely, if the spatial distance between the first eigenvector and the second eigenvector is smaller. Optionally, the spatial distance can be Euclidean distance, Chebyshev distance, Minkowski distance, Manhattan distance or cosine distance, etc., which is not limited here.

[0055] Exemplarily, the user's delivery address can be obtained, and the weather information corresponding to the user's regional information can be obtained based on the user's delivery address, and the seasonal attribute information of the commodities in the commodity pool can be obtained. Then, the temperature, humidity, wind speed and other weather information of the user's region can be matched with the seasonal attribute information of the commodities in the commodity pool to obtain the commodity categories in the commodity pool that are suitable for the user at present and in the future, and the obtained commodity categories that are suitable for the user at present and in the future can be displayed to the user.

[0056] In one embodiment, it can be determined whether the user has selected a seasonal processing rule. After the user selects a seasonal processing rule, the products will be filtered according to the seasonal processing rule. More specifically, the user can select a season category to apply the corresponding seasonal processing rule. For example, if the user selects winter, the system will generate a winter filter mark. If there is a winter filter mark, the user needs to filter products in other seasonal categories. In one case, the user may need to purchase off-season products, or the user forgets to check a specific season processing rule. If the product to be determined is a product under the third-level category covered by the product sequence clicked by the user in the past two days, no filtering is required.

[0057] Visibility processing rules may include obtaining the user's regional information; obtaining the product inventory information of the product corresponding to the target search condition; performing an intersection operation on the regional information and the product inventory information to determine whether the product corresponding to the target search condition is in stock at the user's location; and filtering out products that are out of stock at the user's location from historical product browsing records.

[0058] Optionally, the user's regional information can be obtained from the user's basic information. The user selects a regional range, such as "same city", "same district", "5 kilometers around" and other regional restrictions, to obtain inventory information of products corresponding to the target search in different regions, and then find products that meet the user's regional information, target search conditions, and product inventory corresponding to the target search conditions. Products that meet the above conditions will be displayed as target products, and products that do not meet the conditions will be filtered out and not displayed or recommended in the product display list.

[0059] In one embodiment, if the user chooses to apply the visibility processing rules, the visibility processing rules will carry regional information. If no regional information is carried, the visibility processing rules will not be applied by default. The preset processing rules of this application may also include: rules for breaking up the same product category. For example, in order to prevent products of the same category from appearing in clusters, they need to be broken up. Exemplarily, the similarity between the products in the user's historical product browsing records and the products corresponding to the target search conditions can be calculated, where the similarity can be the similarity of the product ID or the similarity of the product category. A threshold is set, and the products in the user's historical product browsing records with a calculated similarity value greater than the threshold are used as related recommended products, and the products in the user's historical product browsing records with a calculated similarity value less than the threshold are used as non-related recommended products. These products can then be displayed to the user in a preset order, and the preset order can be a randomly shuffled order. Although users have a low level of interest in non-related recommended products for the target search conditions, the characteristics of the target search terms may not be comprehensive. Users may still be interested in products, but the characteristics of the products of interest are not reflected in the target search conditions. In addition, these non-related products may inspire new purchasing directions for target users. By applying the rules for breaking up products of the same product category, products corresponding to users' historical product browsing records can be displayed in random order. While ensuring that users do not become tired, new purchasing directions can be stimulated for users, which can greatly improve the user experience.

[0060] The preset processing rules may also include, for example, if a user clicks on a product that he is not interested in, the product needs to be sorted to the bottom, the ranking of products that the user is not interested in needs to be lowered to avoid appearing at the top of the user's browsing list, and so on.

[0061] In actual applications, the application levels of the various rules included in the preset processing rules are different. The application levels here can include the first level, the second level, and the third level. The first level preset processing rules are visibility processing rules. When a product does not meet the visibility processing rules, the product can be directly deleted from the product display list without subsequent product re-arrangement. The second level preset processing rules are season processing rules, which are relatively special rules and are applied when the product attribute information of different seasons is quite different, for example, the clothes worn in winter and summer are quite different. The third level preset processing rules may include rules for breaking up the same product category, rules for sinking negative feedback products to the bottom, and so on.

[0062] In one embodiment, the preset processing rules also include similarity processing rules, and the method also includes: obtaining user historical behavior data, determining the incremental score of the corresponding user's visits to each product based on the number of visits to each product in the user's historical behavior data; obtaining the product browsing list in the user's historical product browsing record, and calculating the intermediate change value required for the similarity calculation between the product corresponding to the user's historical behavior data and each product in the product browsing list in the user's historical product browsing record based on the incremental score; updating the similarity value within the corresponding element in the product similarity list according to the intermediate change value, and the product similarity list is used to store the similarity value between each two product objects in the product browsing list in the user's historical product browsing record; in response to receiving the target retrieval condition, obtaining the product corresponding to the product browsing list in the user's historical product browsing record corresponding to the target retrieval condition, querying the product similarity list, and displaying the products whose similarity values ​​meet the preset conditions in the product display list.

[0063] Specifically, the user's historical behavior data also includes behavior data such as the user clicking on the collection, like, forward, add to shopping cart, etc. on the product interface. In this embodiment, the parameters of the user's historical behavior data can be used as a scoring method to set a corresponding scoring table based on the type of behavior data such as the user clicking on the collection, like, forward, add to shopping cart, etc. on the product interface. Thus, the behavior scoring table can be queried to obtain the score corresponding to the user's behavior type.

[0064] Different scores are given to different behavior types. Specifically, the scores are set according to the relative difficulty, frequency, complexity, time spent, consumption, etc. of the behavior type. Therefore, the behavior types with higher relative difficulty, more frequency, greater complexity, longer time spent, and higher consumption amount should have higher corresponding scores. The specific score values ​​can be flexibly set by technical personnel in this field with reference to the aforementioned logic.

[0065] Obtain the product browsing list in the user's historical product browsing record, and calculate the intermediate change value required for similarity calculation between the product corresponding to the user's historical behavior data and each product in the product browsing list in the user's historical product browsing record based on the incremental score; the intermediate variable is the variable required for calculating the similarity between the product corresponding to the user's historical behavior data and other products in the product browsing list in the user's historical product browsing record according to the preset similarity algorithm, and the products in the product browsing list in the user's historical product browsing record include the products corresponding to the user's historical behavior data, that is, the intermediate change value required for similarity calculation between the product corresponding to the user's historical behavior data and each product in the product browsing list in the user's historical product browsing record is obtained. The similarity calculation formula can be selected from common formulas such as the cosine similarity formula, the Euclidean distance formula, the Pearson coefficient formula, and the Jaccard similarity formula. The intermediate variables are set accordingly according to the components in the preset similarity formula, the incremental score of the corresponding user visiting each product is determined by the number of visits to each product in the current user's historical behavior data, and the change value of the intermediate variable corresponding to the dependent variable item that changes due to the incremental score is calculated, that is, the intermediate change value. The intermediate variables corresponding to the fixed items that do not respond to the change do not need to be calculated, and then the intermediate variables required for the similarity formula calculation are obtained. Furthermore, once the user behavior data is received, the aforementioned incremental similarity calculation is performed, and the incremental similarity calculation method can be called in real time, which requires fewer computing resources and can quickly calculate the results. Real-time recommendations can be achieved in business scenarios where high-frequency similarity calculations are performed to match similar products for product recommendations.

[0066] In one embodiment, after obtaining multiple target products, a query can be performed based on the target search conditions to obtain a display list of products that meet the target search conditions in any time period, and the target products are displayed to facilitate users to browse the target products and the user's historical browsing records of the target products. For example, when a user selects a target product, the user can double-click the target product to enter the detailed browsing page of the target product. The detailed browsing page includes the target product sorting corresponding to the user's historical product browsing records, the product source of the target product, product parameters, product evaluation, product services included in the target product, etc.

[0067] Step 102: select any one or several commodities in the historical commodity browsing records of the user, determine the display status of the selected commodity, and display the preset processing rules corresponding to the selected commodity.

[0068] Since the final product display list is obtained after the preset processing rules are implemented on the products in the user's historical product browsing records according to the preset processing rules, in actual applications, the target products displayed in the product display list will be different from the products in the actual user's historical product browsing records. For example, after implementing the preset processing rules, some products that do not meet the preset processing rules may be filtered out, or after implementing the preset processing rules, the order of the products will be different from the original user's historical product browsing records. You can select any one or several products in the user's product browsing records to determine the display status of the selected products, where the display status includes displayed and not displayed. First, according to the target retrieval condition, the products that meet the target retrieval condition can be found from the user's historical product browsing records as the first batch of products. After being processed by the preset processing rules, multiple target products will be obtained and displayed to the user in the form of a product display list. When any selected product appears in the first batch of products but does not appear in multiple target products, it means that the product has been filtered out by the preset processing rules. At this time, you can click on the product to view which preset processing rule the product is filtered out by, and which rule a certain product is affected by during the reordering process. Subsequent users can change the preset processing rules according to actual needs. It is understandable that the user can select a certain number of entries of preset rules in the system settings, or can choose to apply all preset rules by default.

[0069] Different from the prior art, the present application can determine the display status of the selected products by selecting any one or several products in the user's historical product browsing records, and display the preset processing rules corresponding to the selected products, so that users can test and change the preset processing rules implemented in the product display process, thereby improving the user experience.

[0070] When a user views a product display list, the user's purchase preferences can be generated based on the user's historical product browsing records and the user's historical product purchase records. The user's purchase preferences reflect the user's needs and preferences to a certain extent, and products are displayed to the user based on the user's purchase preferences.

[0071] Specifically, the user's historical product browsing records and the user's historical product purchase records are obtained; the contribution of each user's historical product browsing records for each category of product is determined based on the similarity between the user's historical product browsing records and the user's historical product purchase records; the user's purchasing habits are determined based on the number of historical browsings between adjacent purchase records for each category of product; the user's purchasing preferences are determined based on the contribution of each user's historical product browsing records for each category of product and the user's purchasing habits, and products are displayed to the user based on the user's purchasing preferences.

[0072] The user's historical product browsing records can be traversed. When the user's historical product browsing records are similar to the user's historical product purchase records, it is believed that the browsing records will greatly promote the transaction. That is, the contribution of each user's historical product browsing records for each category of products is determined.

[0073] The data of the last completed purchase transaction can be used as a node to conduct a specific analysis of the browsing records that facilitated the last transaction, wherein the browsing data between two adjacent transactions is used as the browsing data that facilitated the transaction corresponding to the later transaction of the two adjacent transactions. In the historical browsing data before the historical purchase data, the user's historical browsing data must contain browsing data of the transaction product or similar products, and the historical browsing data will facilitate the user to complete the final purchase transaction. However, in actual applications, there will definitely be some browsing records other than purchased goods in the browsing records, and these browsing records will have a certain degree of similarity with the purchased goods, but some browsing records have different similarities with the purchased goods. The browsing records before each transaction are analyzed to better explore the user's purchasing preferences.

[0074] The contribution of the user's historical product browsing records on each e-commerce platform can be used as a weight, and the sum of the contribution of the user's historical product browsing records corresponding to each category of products in all historical browsing records between the most recent historical purchase record and the previous historical purchase record on each e-commerce platform can be weighted to obtain the user's purchasing preference.

[0075] In one embodiment, in the process of facilitating transactions, users may browse some commodities that are not similar to the purchased products. From the perspective of people's psychology of purchasing commodities, although the browsing of such commodities is relatively small, it has a certain contribution to facilitating transactions. Optionally, the user's historical commodity purchase records can be obtained, and the user's historical commodity purchase records include the user's historical behavior data; according to the access history data and purchase history data in the user's historical behavior data, the associated commodities of the commodities involved in the historical behavior data are calculated by using association rules or collaborative filtering algorithms, and the correlation between the commodities involved in the historical behavior data and its associated commodities is calculated, and the first preset number of commodities with the highest correlation are selected as the associated commodity set of the commodities involved in the historical behavior data, and displayed in the commodity display list.

[0076] Optionally, association rules can be mined in user historical behavior data based on the April algorithm.

[0077] In the present invention, a frequent itemset mining algorithm is adopted. The frequent itemset mining algorithm (AprioriAlgorithm) is a classic data mining algorithm for mining frequent itemsets and association rules. April means "from before" in Latin. When defining a problem, prior knowledge or assumptions are usually used, which is called "a priori". The April algorithm uses the prior properties of frequent itemsets to compress the search space. The April algorithm is the most classic algorithm for mining frequent itemsets. The goal of various association rule algorithms is to start from different aspects to reduce the possible search space size and reduce the number of times data is scanned. The core idea of ​​the algorithm is to generate candidate items and their support by connection, and then generate frequent itemsets by pruning.

[0078] Among them, association rules are used to reflect the association relationship between various product categories. For example, after a user purchases product A, he or she will generally also purchase product B. In this case, there is a purchase association relationship between product A and product B. Therefore, as a product recommendation scheme, after a user purchases product A, product B or similar products of product B can be recommended to the user.

[0079] By calculating the correlation between the commodities involved in the historical behavior data and their associated commodities, the first preset number of commodities with the highest correlation are selected as the associated commodity set of the commodities involved in the historical behavior data, and displayed in the commodity display list, the user's behavior data can be accurately analyzed, and personalized commodity display recommendations can be provided to the user, making the commodity display recommendations more accurate. The preset number here can be set according to actual needs.

[0080] In one embodiment, historical product evaluations of products in a user's historical product browsing history can be obtained; the historical product evaluations are analyzed through semantic analysis text to extract emotional sentences and emotional images that carry emotional information; the emotional sentences and emotional images are scored respectively, and the total scores of the emotional sentences and emotional images are numerically sorted by an evaluation weight model; product content labels are formed according to the total scores of the emotional sentences and emotional images, and the products are displayed according to the product content labels; wherein, by identifying the emotional confidence of the emotional sentences, the emotional sentences are divided into positive, negative, and neutral emotional categories, and emotional sentences of different emotional categories are scored; the first evaluation feature vector and the second evaluation feature vector corresponding to the emotional image and the store recommendation image are respectively extracted; the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector is calculated; and the score of the emotional image is calculated according to the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector.

[0081] The historical product reviews of the products in the user's historical product browsing records can be obtained from the e-commerce platform, and then the semantic sentiment tendency and content richness of the evaluation text are analyzed respectively through semantic analysis text, and then the semantic sentiment tendency and the content richness are numerically calculated to obtain the final score of the evaluation text. Furthermore, the analysis of the semantic sentiment tendency distinguishes positive, negative, and neutral sentiment polarities by judging the confidence of the sentiment polarity category of the evaluation text, and counts the number of evaluations with negative sentiment polarity categories. Furthermore, the analysis of the content richness includes first extracting the sentiment opinions involved in the evaluation sentiment vocabulary in the evaluation text through the three dimensions of product, logistics, and service, and obtaining two polarity comment opinion labels of good and bad reviews.

[0082] Scoring the evaluation screen includes: respectively identifying the feature vectors of the emotional picture and the main picture of the store's goods, obtaining the Euclidean distance of the feature vectors as a similarity score, and calculating the final score of the evaluation screen through the similarity score. Exemplarily, the image recognition model can be used to calculate the image feature vectors of the user's emotional picture and multiple main pictures of goods, and then the Euclidean distance between the feature vector of each user's emotional picture and the feature vector of multiple main pictures of goods, that is, the similarity score, is calculated respectively, and finally the highest value of multiple similarity scores is taken as the similarity score of the user's emotional picture, and finally the emotional picture score is scored through the calculation model. When displaying goods, the score on the product content label can be considered. Assuming that the score range is in the numerical range of 0-100, the goods with a score below 50 can be filtered out, which helps users quickly filter out goods with poor evaluations.

[0083] In one embodiment, the user's historical product purchase records may be obtained and analyzed, which may specifically include extracting product combinations that appear in the same user's historical product purchase records. When the same product combination appears multiple times in the user's historical product purchase records, it is considered that the products in the product combination are related. When displaying products, the related products may be placed in adjacent positions.

[0084] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0085] In one embodiment, Figure 2 As shown, a commodity management device is provided, including: an acquisition module 20 and a display module 21, wherein:

[0086] The acquisition module 20 is used to acquire the user's historical product browsing records and target search conditions, and query the user's historical product browsing records according to the target search conditions to obtain a product display list; wherein the product display list includes multiple target products, and the displayed products are processed according to preset processing rules to generate a product display list; the preset processing rules include a combination of one or more rules of season processing rules, visibility processing rules, user preference processing rules, related product processing rules, product evaluation processing rules, etc.;

[0087] The display module 21 is used to select any one or several commodities in the historical commodity browsing records of the user, determine the display status of the selected commodity, and display the preset processing rules corresponding to the selected commodity.

[0088] In one embodiment, another implementation of the commodity management method that can be implemented by the above device includes the following specific steps: the preset processing rules include a season processing rule, and the method further includes:

[0089] Acquire the user's regional information, and acquire the weather information within a preset time range corresponding to the user's regional information; perform feature extraction on the weather information within the preset time range corresponding to the user's regional information to obtain a first feature vector; the weather information includes temperature, humidity, wind speed, visibility, etc.;

[0090] Acquire seasonal attribute information of commodities in the commodity pool, perform feature extraction on the seasonal attribute information of commodities in the commodity pool, and obtain a second feature vector;

[0091] Calculate the similarity between the first eigenvector and the second eigenvector;

[0092] When the similarity is higher than a preset degree, it is considered that the seasonal attribute information of the product in the product pool matches the weather information within a preset time range corresponding to the user's regional information;

[0093] The products in the product pool that match the weather information within a preset time range corresponding to the user's region are displayed to the user.

[0094] In one embodiment, another implementation of the commodity management method that can be implemented by the above device includes the following specific steps: the preset processing rule includes a visibility processing rule, and the method further includes:

[0095] Get the user's regional information;

[0096] Obtain the product inventory information of the product corresponding to the target search condition;

[0097] Perform intersection operation on the region information and product inventory information to determine whether the product corresponding to the target search condition is in stock at the user's location;

[0098] Filter out-of-stock products in the user's location from historical product browsing records.

[0099] In one embodiment, another implementation of the commodity management method that can be implemented by the above device includes the following specific steps: the preset processing rule includes a user preference processing rule, and the method further includes:

[0100] Obtain the user's historical product browsing records and the user's historical product purchase records;

[0101] Determine the contribution of each user's historical product browsing record for each category of products based on the similarity between the user's historical product browsing record and the user's historical product purchase record;

[0102] Determine the user's purchasing habits based on the number of historical browsing between adjacent purchase records of each category of goods;

[0103] The contribution of the user's historical product browsing records on each e-commerce platform is used as the weight, and the sum of the contribution of the user's historical product browsing records corresponding to each category of products in all historical browsing records between the most recent historical purchase record and the previous historical purchase record on each e-commerce platform is weighted to obtain the user's purchase preference.

[0104] In one embodiment, another implementation of the commodity management method that can be implemented by the above device includes the following specific steps: the preset processing rule includes an associated commodity processing rule, and the method further includes:

[0105] Obtain the user's historical commodity purchase records, which include the user's historical behavior data;

[0106] According to the access history data and purchase history data in the user's historical behavior data, association rules or collaborative filtering algorithms are used to calculate the associated products of the products involved in the historical behavior data, calculate the correlation between the products involved in the historical behavior data and their associated products, select a preset number of products with the highest correlation as the associated product set of the products involved in the historical behavior data, and display them in the product display list.

[0107] In one embodiment, another implementation of the commodity management method that can be implemented by the above device includes the following specific steps: The method further includes:

[0108] Obtain the user's historical behavior data, and determine the corresponding incremental score of the user's visit to each product based on the number of visits to each product in the user's historical behavior data;

[0109] Obtain a product browsing list in the user's historical product browsing record, and calculate the intermediate change value required for similarity calculation between the product corresponding to the user's historical behavior data and each product in the product browsing list in the user's historical product browsing record based on the incremental score;

[0110] Update the similarity value in the corresponding element in the product similarity list according to the intermediate change value, where the product similarity list is used to store the similarity values ​​between two product objects in the product browsing list in the user's historical product browsing record;

[0111] In response to receiving the target search condition, the corresponding product in the product browsing list in the user's historical product browsing record corresponding to the target search condition is obtained, the product similarity list is queried, and the products whose similarity values ​​meet the preset conditions are displayed in the product display list.

[0112] In one embodiment, another implementation of the commodity management method that can be implemented by the above device includes the following specific steps: the preset processing rules include commodity evaluation processing rules, and the method further includes:

[0113] Get the historical product reviews of the products in the user's historical product browsing history;

[0114] Analyze historical product reviews through semantic analysis text, extract emotional sentences and emotional images that carry emotional information;

[0115] Score the emotional sentences and emotional images respectively, and sort the total scores of the emotional sentences and emotional images by numerical value through the evaluation weight model; form product content labels according to the numerical order of the total scores of the emotional sentences and emotional images, and display the products according to the product content labels;

[0116] Among them, scoring emotional sentences and emotional images includes:

[0117] By identifying the emotional confidence of emotional sentences, emotional sentences are classified into positive, negative, and neutral emotional categories, and emotional sentences of different emotional categories are scored;

[0118] Extracting the first evaluation feature vector and the second evaluation feature vector corresponding to the emotional image and the store recommendation image respectively;

[0119] Calculating the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector;

[0120] The emotional image is scored according to the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector.

[0121] For the specific definition of the commodity management device, please refer to the definition of the commodity management method above, which will not be repeated here. Each module in the above commodity management device can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0122] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a commodity management method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0123] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0124] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0125] Step 101, obtain the user's historical product browsing records and target search conditions, query the user's historical product browsing records according to the target search conditions, and obtain a product display list; wherein the product display list includes multiple target products, and the displayed products are processed according to preset processing rules to generate a product display list; the preset processing rules include a combination of one or more rules such as season processing rules, visibility processing rules, user preference processing rules, related product processing rules, and product evaluation processing rules.

[0126] Step 102: select any one or several commodities in the historical commodity browsing records of the user, determine the display status of the selected commodity, and display the preset processing rules corresponding to the selected commodity.

[0127] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0128] The preset processing rules include a season processing rule, and the method further includes:

[0129] Acquire the user's regional information, and acquire the weather information within a preset time range corresponding to the user's regional information; perform feature extraction on the weather information within the preset time range corresponding to the user's regional information to obtain a first feature vector; the weather information includes temperature, humidity, wind speed, visibility, etc.;

[0130] Acquire seasonal attribute information of commodities in the commodity pool, perform feature extraction on the seasonal attribute information of commodities in the commodity pool, and obtain a second feature vector;

[0131] Calculate the similarity between the first eigenvector and the second eigenvector;

[0132] When the similarity is higher than a preset threshold, it is determined that the seasonal attribute information of the product in the product pool matches the weather information within a preset time range corresponding to the user's regional information;

[0133] The products in the product pool that match the weather information within a preset time range corresponding to the user's regional information are displayed to the user.

[0134] In one embodiment, when the processor executes the computer program, the following steps are further implemented: the preset processing rule includes a visibility processing rule, and the method further includes:

[0135] Get the user's regional information;

[0136] Obtain the product inventory information of the product corresponding to the target search condition;

[0137] Perform intersection operation on the region information and product inventory information to determine whether the product corresponding to the target search condition is in stock at the user's location;

[0138] Filter out-of-stock products in the user's location from historical product browsing records.

[0139] In one embodiment, when the processor executes the computer program, the following steps are further implemented: the preset processing rule includes a user preference processing rule, and the method further includes:

[0140] Obtain the user's historical product browsing records and the user's historical product purchase records;

[0141] Determine the contribution of each user's historical product browsing record for each category of products based on the similarity between the user's historical product browsing record and the user's historical product purchase record;

[0142] Determine the user's purchasing habits based on the number of historical browsing between adjacent purchase records of each category of goods;

[0143] The contribution of the user's historical product browsing records on each e-commerce platform is used as the weight, and the sum of the contribution of the user's historical product browsing records corresponding to each category of products in all historical browsing records between the most recent historical purchase record and the previous historical purchase record on each e-commerce platform is weighted to obtain the user's purchase preference.

[0144] In one embodiment, when the processor executes the computer program, the following steps are further implemented: the preset processing rule includes an associated commodity processing rule, and the method further includes:

[0145] Obtain the user's historical commodity purchase records, which include the user's historical behavior data;

[0146] According to the access history data and purchase history data in the user's historical behavior data, association rules or collaborative filtering algorithms are used to calculate the associated products of the products involved in the historical behavior data, calculate the correlation between the products involved in the historical behavior data and their associated products, select a preset number of products with the highest correlation as the associated product set of the products involved in the historical behavior data, and display them in the product display list.

[0147] In one embodiment, when the processor executes the computer program, the method further implements the following steps:

[0148] Obtain the user's historical behavior data, and determine the corresponding incremental score of the user's visit to each product based on the number of visits to each product in the user's historical behavior data;

[0149] Obtain a product browsing list in the user's historical product browsing record, and calculate the intermediate change value required for similarity calculation between the product corresponding to the user's historical behavior data and each product in the product browsing list in the user's historical product browsing record based on the incremental score;

[0150] Update the similarity value in the corresponding element in the product similarity list according to the intermediate change value, where the product similarity list is used to store the similarity values ​​between two product objects in the product browsing list in the user's historical product browsing record;

[0151] In response to receiving the target search condition, the corresponding product in the product browsing list in the user's historical product browsing record corresponding to the target search condition is obtained, the product similarity list is queried, and the products whose similarity values ​​meet the preset conditions are displayed in the product display list.

[0152] In one embodiment, when the processor executes the computer program, the method further implements the following steps:

[0153] The preset processing rules include product evaluation processing rules, and the method further includes:

[0154] Get the historical product reviews of the products in the user's historical product browsing history;

[0155] Analyze historical product reviews through semantic analysis text, extract emotional sentences and emotional images that carry emotional information;

[0156] Score the emotional sentences and emotional images respectively, and sort the total scores of the emotional sentences and emotional images by numerical value through the evaluation weight model; form product content labels according to the numerical order of the total scores of the emotional sentences and emotional images, and display the products according to the product content labels;

[0157] Among them, scoring emotional sentences and emotional images includes:

[0158] By identifying the emotional confidence of emotional sentences, emotional sentences are classified into positive, negative, and neutral emotional categories, and emotional sentences of different emotional categories are scored;

[0159] Extracting the first evaluation feature vector and the second evaluation feature vector corresponding to the emotional image and the store recommendation image respectively;

[0160] Calculating the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector;

[0161] The emotional image is scored according to the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector.

[0162] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0163] Step 101, obtain the user's historical product browsing records and target search conditions, query the user's historical product browsing records according to the target search conditions, and obtain a product display list; wherein the product display list includes multiple target products, and the displayed products are processed according to preset processing rules to generate a product display list; the preset processing rules include a combination of one or more rules such as season processing rules, visibility processing rules, user preference processing rules, related product processing rules, and product evaluation processing rules.

[0164] Step 102: select any one or several commodities in the historical commodity browsing records of the user, determine the display status of the selected commodity, and display the preset processing rules corresponding to the selected commodity.

[0165] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0166] Acquire the user's regional information, and acquire the weather information within a preset time range corresponding to the user's regional information; perform feature extraction on the weather information within the preset time range corresponding to the user's regional information to obtain a first feature vector; the weather information includes temperature, humidity, wind speed, visibility, etc.;

[0167] Acquire seasonal attribute information of commodities in the commodity pool, perform feature extraction on the seasonal attribute information of commodities in the commodity pool, and obtain a second feature vector;

[0168] Calculate the similarity between the first eigenvector and the second eigenvector;

[0169] When the similarity is higher than a preset threshold, it is determined that the seasonal attribute information of the product in the product pool matches the weather information within a preset time range corresponding to the user's regional information;

[0170] The products in the product pool that match the weather information within a preset time range corresponding to the user's regional information are displayed to the user.

[0171] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0172] The preset processing rule includes a visibility processing rule, and the method further includes:

[0173] Get the user's regional information;

[0174] Obtain the product inventory information of the product corresponding to the target search condition;

[0175] Perform intersection operation on the region information and product inventory information to determine whether the product corresponding to the target search condition is in stock at the user's location;

[0176] Filter out-of-stock products in the user's location from historical product browsing records.

[0177] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0178] The preset processing rules include user preference processing rules, and the method further includes:

[0179] Obtain the user's historical product browsing records and the user's historical product purchase records;

[0180] Determine the contribution of each user's historical product browsing record for each category of products based on the similarity between the user's historical product browsing record and the user's historical product purchase record;

[0181] Determine the user's purchasing habits based on the number of historical browsing between adjacent purchase records of each category of goods;

[0182] The contribution of the user's historical product browsing records on each e-commerce platform is used as the weight, and the sum of the contribution of the user's historical product browsing records corresponding to each category of products in all historical browsing records between the most recent historical purchase record and the previous historical purchase record on each e-commerce platform is weighted to obtain the user's purchase preference.

[0183] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0184] The preset processing rules include associated product processing rules, and the method further includes:

[0185] Obtain the user's historical commodity purchase records, which include the user's historical behavior data;

[0186] According to the access history data and purchase history data in the user's historical behavior data, association rules or collaborative filtering algorithms are used to calculate the associated products of the products involved in the historical behavior data, calculate the correlation between the products involved in the historical behavior data and their associated products, select a preset number of products with the highest correlation as the associated product set of the products involved in the historical behavior data, and display them in the product display list.

[0187] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0188] The method also includes:

[0189] Obtain the user's historical behavior data, and determine the corresponding incremental score of the user's visit to each product based on the number of visits to each product in the user's historical behavior data;

[0190] Obtain a product browsing list in the user's historical product browsing record, and calculate the intermediate change value required for similarity calculation between the product corresponding to the user's historical behavior data and each product in the product browsing list in the user's historical product browsing record based on the incremental score;

[0191] Update the similarity value in the corresponding element in the product similarity list according to the intermediate change value, where the product similarity list is used to store the similarity values ​​between two product objects in the product browsing list in the user's historical product browsing record;

[0192] In response to receiving the target search condition, the corresponding product in the product browsing list in the user's historical product browsing record corresponding to the target search condition is obtained, the product similarity list is queried, and the products whose similarity values ​​meet the preset conditions are displayed in the product display list.

[0193] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0194] The preset processing rules include product evaluation processing rules, and the method further includes:

[0195] Get the historical product reviews of the products in the user's historical product browsing history;

[0196] Analyze historical product reviews through semantic analysis text, extract emotional sentences and emotional images that carry emotional information;

[0197] Score the emotional sentences and emotional images respectively, and sort the total scores of the emotional sentences and emotional images by numerical value through the evaluation weight model; form product content labels according to the numerical order of the total scores of the emotional sentences and emotional images, and display the products according to the product content labels;

[0198] Among them, scoring emotional sentences and emotional images includes:

[0199] By identifying the emotional confidence of emotional sentences, emotional sentences are classified into positive, negative, and neutral emotional categories, and emotional sentences of different emotional categories are scored;

[0200] Extracting the first evaluation feature vector and the second evaluation feature vector corresponding to the emotional image and the store recommendation image respectively;

[0201] Calculating the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector;

[0202] The emotional image is scored according to the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector.

[0203] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0204] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0205] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A commodity management method, characterized in that: include: Obtaining a user's historical product browsing records and a target search condition, querying the user's historical product browsing records according to the target search condition, and obtaining a product display list; wherein the product display list includes a plurality of target products, and processing the display products according to a preset processing rule to generate a product display list; the preset processing rule includes a combination of one or more of a season processing rule, a visibility processing rule, a user preference processing rule, an associated product processing rule, a product evaluation processing rule, and the like; Select any one or several products in the user's historical product browsing records, determine the display status of the selected products, and display the preset processing rules corresponding to the selected products.

2. The method according to claim 1, characterized in that The preset processing rules include season processing rules, and the method further includes: Acquire the user's regional information, and acquire the weather information within a preset time range corresponding to the user's regional information; perform feature extraction on the weather information within the preset time range corresponding to the user's regional information to obtain a first feature vector; the weather information includes temperature, humidity, wind speed, visibility, etc.; Acquire seasonal attribute information of commodities in the commodity pool, perform feature extraction on the seasonal attribute information of commodities in the commodity pool, and obtain a second feature vector; Calculating the similarity between the first feature vector and the second feature vector; When the similarity is higher than a preset threshold, it is determined that the seasonal attribute information of the product in the product pool matches the weather information within a preset time range corresponding to the user's regional information; The products in the product pool that match the weather information within a preset time range corresponding to the user's regional information are displayed to the user.

3. The method according to claim 1, characterized in that The preset processing rule includes a visibility processing rule, and the method further includes: Get the user's regional information; Obtain the product inventory information of the product corresponding to the target search condition; Performing an intersection operation on the region information and the product inventory information to determine whether the product corresponding to the target search condition is in stock at the user's location; Filter out-of-stock products in the user's location from historical product browsing records.

4. The method according to claim 1, characterized in that: The preset processing rule includes a user preference processing rule, and the method further includes: Obtain the user's historical product browsing records and the user's historical product purchase records; Determining the contribution of each user's historical product browsing record for each category of products according to the similarity between the user's historical product browsing record and the user's historical product purchase record; Determine the user's purchasing habits based on the number of historical browsing between adjacent purchase records of each category of goods; The contribution of the user's historical product browsing records on each e-commerce platform is used as the weight, and the sum of the contribution of the user's historical product browsing records corresponding to each category of products in all historical browsing records between the most recent historical purchase record and the previous historical purchase record on each e-commerce platform is weighted to obtain the user's purchase preference.

5. The method according to claim 1, characterized in that: The preset processing rules include associated commodity processing rules, and the method further includes: Obtaining a user's historical commodity purchase records, wherein the user's historical commodity purchase records include user's historical behavior data; According to the access history data and purchase history data in the user's historical behavior data, an association rule or a collaborative filtering algorithm is used to calculate the associated products of the products involved in the historical behavior data, calculate the correlation between the products involved in the historical behavior data and their associated products, select a preset number of products with the highest correlation as the associated product set of the products involved in the historical behavior data, and display them in the product display list.

6. The method according to claim 1, characterized in that The method further comprises: Obtain the user's historical behavior data, and determine the corresponding incremental score of the user's visit to each product based on the number of visits to each product in the user's historical behavior data; Obtain a product browsing list in the user's historical product browsing record, and calculate an intermediate change value required for similarity calculation between the product corresponding to the user's historical behavior data and each product in the product browsing list in the user's historical product browsing record based on the incremental score; Update the similarity value in the corresponding element in the product similarity list according to the intermediate change value, wherein the product similarity list is used to store the similarity values ​​between two product objects in the product browsing list in the user's historical product browsing record; In response to receiving the target search condition, the corresponding product in the product browsing list in the user's historical product browsing record corresponding to the target search condition is obtained, the product similarity list is queried, and the products whose similarity values ​​meet the preset conditions are displayed in the product display list.

7. The method according to claim 1, characterized in that The preset processing rules include commodity evaluation processing rules, and the method further includes: Get the historical product reviews of the products in the user's historical product browsing history; Analyze the historical product reviews through semantic analysis text to extract emotional sentences and emotional images carrying emotional information; Scoring the emotional sentences and emotional images respectively, and numerically sorting the total scores of the emotional sentences and emotional images by using an evaluation weight model; forming product content labels according to the numerical sorting of the total scores of the emotional sentences and emotional images, and displaying the products according to the product content labels; Among them, scoring emotional sentences and emotional images includes: By identifying the emotional confidence of the emotional sentences, the emotional sentences are classified into positive, negative and neutral emotional categories, and the emotional sentences of different emotional categories are scored; Extracting the first evaluation feature vector and the second evaluation feature vector corresponding to the emotional image and the store recommendation image respectively; Calculating the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector; The emotional image is scored according to the Euclidean distance between the first evaluation feature vector and the second evaluation feature vector.

8. A commodity management device, characterized in that: The device comprises: The acquisition module is used to acquire the user's historical product browsing records and target search conditions, and query the user's historical product browsing records according to the target search conditions to obtain a product display list; wherein the product display list includes multiple target products, and the display products are processed according to preset processing rules to generate a product display list; the preset processing rules include a combination of one or more rules of season processing rules, visibility processing rules, user preference processing rules, related product processing rules, product evaluation processing rules, etc.; The display module is used to select any one or several products in the user's historical product browsing records, determine the display status of the selected products, and display the preset processing rules corresponding to the selected products.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images