E-commerce product accurate recommendation method, device and equipment, medium and product

Abstract

The invention discloses an e-commerce product accurate recommendation method and device, equipment, a medium and a product, and relates to the technical field of big data analysis. The method comprises the following steps: extracting sample data of various e-commerce products based on e-commerce product browsing record data and e-commerce product purchase record data, and training based on the sample data to obtain a purchase event prediction model of various e-commerce products; and then, based on the prediction model and the e-commerce product recommendation scheme, estimating to obtain the confidence coefficient of an event that various e-commerce products are purchased by a target customer in the current next starting operation period, and finally, when it is judged that the sum of the confidence coefficients of the various e-commerce products exceeds a preset threshold value, confirming that the recommendation accuracy of the recommendation scheme meets a preset target. And in the current next starting operation period, the recommendation scheme is executed through the e-commerce platform application program, so that the purchase willingness of the user can be effectively improved when the method is applied to multiple types of e-commerce product recommendation scenes, and the purpose of further improving the product recommendation accuracy is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of big data analysis, and specifically relates to a method, device, equipment, medium and product for accurately recommending e-commerce products. Background Art

[0002] With the booming development of e-commerce, users' demand for personalized recommendations is increasing. Traditional e-commerce product recommendation solutions are mainly based on users' historical purchase records and evaluation data made after purchase. For example, the existing patent "CN114429384A, Product Intelligent Recommendation Method and System Based on E-commerce Platform" provides a product intelligent recommendation solution based on e-commerce platform, which specifically includes: querying the data of a target product in the historical shopping data of a target user and its corresponding evaluation data; obtaining a similar user group with similar shopping behavior as the target user; obtaining a list of candidate products; selecting a preset number of candidate products from the list of candidate products; obtaining the sentiment scores corresponding to the preset key evaluation word categories based on the evaluation data of the selected candidate products and the target product and forming a score input matrix with these sentiment scores; inputting the score input matrix into a recommendation model so that the recommendation model outputs the recommended product. However, the aforementioned solution obviously has the problem of insufficient continuity of product recommendation quality and insufficient comprehensiveness of portrait type generation due to the lack of precise adjustment of the product recommendation interval when users have less feedback on the products on the browsing page, which leads to a decrease in the accuracy of product recommendations.

[0003] In response to the above problems, the existing patent "CN117829965A, a product recommendation system and method for a cross-border e-commerce platform" provides a product recommendation solution for a cross-border e-commerce platform, which specifically includes: a data collection module; a data processing module, which is used to calculate and process the first-level operation characteristic data of the browsing page to output the second-level operation characteristic data of the browsing page; a product update module, including a portrait generation unit for generating a user portrait based on the user characteristic data and the user's page browsing data and a product recommendation unit connected to the portrait generation unit for scheduling the corresponding recommended product information of the user portrait to the browsing page; a control module, which is used to determine the corresponding update method of the product recommendation unit according to the proportion of the number of product orders, or, according to the proportion of the number of product orders and the number of overlaps between the searched and recommended product types, determine the portrait classification method of the portrait generation unit. The above-mentioned scheme improves the accuracy of product recommendations on cross-border e-commerce platforms.

[0004] However, the above two solutions only consider the impact of recommending a single e-commerce product on customer purchasing decisions, but do not consider the situation where recommending different types of e-commerce products will have a competitive or linkage impact on customer purchasing decisions. For example, recommending tablets will have a competitive impact on customers' decisions to purchase other electronic devices such as smartphones or laptops. For example, recommending smartphones will have a linkage impact on customers' decisions to purchase related supporting products such as Bluetooth headsets or mobile phone cases. Therefore, how to provide a new e-commerce product recommendation solution that is suitable for multiple e-commerce product recommendation scenarios and can effectively improve users' willingness to buy, so as to further improve the accuracy of product recommendations, is a topic that technical personnel in this field urgently need to study. Summary of the invention

[0005] The purpose of the present invention is to provide a method, device, computer equipment, computer-readable storage medium and computer program product for accurately recommending e-commerce products, so as to solve the problem that the accuracy of product recommendations needs to be improved because the existing e-commerce product recommendation schemes only consider the impact of recommending a single e-commerce product on customer purchasing decisions, but do not consider the situation that recommending different types of e-commerce products will have competitive or linkage effects on customer purchasing decisions.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In the first aspect, a method for accurately recommending e-commerce products is provided, comprising:

[0008] Collect e-commerce product browsing record data and e-commerce product purchase record data generated by target customers during each historical startup and operation of the e-commerce platform application;

[0009] According to the e-commerce product browsing record data, the browsing event indication value or browsing time proportion value of each type of e-commerce product during each historical startup operation is obtained, and according to the e-commerce product purchase record data, the purchase event indication value of each type of e-commerce product during each historical startup operation is obtained, wherein the browsing event indication value is used to indicate whether an event of browsing the corresponding e-commerce product by the target customer occurs during the startup operation of the e-commerce platform application, the browsing time proportion value refers to the ratio of the time duration of browsing the corresponding e-commerce product by the target customer to the total time duration of browsing all e-commerce products during the startup operation of the e-commerce platform application, and the purchase event indication value is used to indicate whether an event of purchasing the corresponding e-commerce product by the target customer occurs during the startup operation of the e-commerce platform application;

[0010] For each type of e-commerce product, the corresponding sample data is extracted in the following manner: the browsing event indication values ​​or browsing time proportion values ​​of all e-commerce products during the m consecutive historical startup runs are taken as model input items, and the purchase event indication value of the corresponding e-commerce product during the last historical startup run during the m consecutive historical startup runs is taken as the model output item, and then the model input item and the model output item are taken as a sample data, wherein m represents a positive integer greater than or equal to 2;

[0011] For each type of e-commerce product, all the corresponding sample data are used to calibrate and validate the machine learning model based on the artificial intelligence algorithm to obtain the corresponding purchase event prediction model;

[0012] For each type of e-commerce product, the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme of the e-commerce platform application during the current next startup operation are imported into the corresponding purchase event prediction model, and the confidence of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup operation is output, wherein the e-commerce product recommendation scheme includes the recommendation event indication values ​​or recommendation frequency proportion values ​​of each type of e-commerce product during the current next startup operation, the recommendation event indication value is used to indicate whether the event of the corresponding e-commerce product being recommended to the target customer for browsing is triggered during the current next startup operation, and the recommendation frequency proportion value refers to the ratio of the frequency of the corresponding e-commerce product being recommended to the target customer for browsing during the current next startup operation to the total frequency of all e-commerce products being recommended to the target customer for browsing;

[0013] Determining whether the sum of the confidence levels of the various e-commerce products exceeds a first preset threshold;

[0014] If so, during the next startup operation, the e-commerce product recommendation solution is executed through the e-commerce platform application.

[0015] Based on the above invention content, a new solution for accurately recommending e-commerce products in multi-category e-commerce product recommendation scenarios is provided, that is, firstly, based on the e-commerce product browsing record data and the e-commerce product purchase record data, sample data of various e-commerce products are extracted, and purchase event prediction models of various e-commerce products are trained based on the sample data, and then, based on the prediction model and the e-commerce product recommendation scheme, the confidence of the event that various e-commerce products are purchased by target customers during the current next startup operation is estimated, and finally, when it is determined that the sum of the confidences of various e-commerce products exceeds a preset threshold, it is confirmed that the recommendation accuracy of the recommendation scheme meets the preset target, and during the current next startup operation, the recommendation scheme is executed through the e-commerce platform application, so that both the impact of recommending a single e-commerce product on the customer's purchase decision and the situation that recommending different types of e-commerce products will have a competitive or linkage impact on the customer's purchase decision are taken into account, so that when applied to multi-category e-commerce product recommendation scenarios, the user's willingness to buy can be effectively improved, and the purpose of further improving the accuracy of product recommendations is achieved, which is convenient for practical application and promotion.

[0016] In one possible design, the method further includes:

[0017] For each type of e-commerce product, based on the purchase event indication value of the corresponding e-commerce product during each of the historical startup and operation periods and the purchase event indication value of other types of e-commerce products during each of the historical startup and operation periods, respectively calculate the linkage purchase influence coefficient of the corresponding e-commerce product and other types of e-commerce products;

[0018] Determining whether the sum of the confidences of the various e-commerce products exceeds a first preset threshold includes:

[0019] For each type of e-commerce product, according to the corresponding confidence and the linkage purchase influence coefficient between the corresponding e-commerce product and other types of e-commerce products, the corresponding confidence correction value is calculated according to the following formula:

[0020]

[0021] In the formula, N represents the total number of categories of e-commerce products, i represents a positive integer less than or equal to N, and c i represents the confidence of the i-th e-commerce product, represents the confidence correction value of the i-th e-commerce product, i′ represents a positive integer less than or equal to N-1, c i,i′ represents the confidence of other i′th category e-commerce products relative to the i-th category e-commerce products, η i,i′ represents the linkage purchase influence coefficient between the i-th e-commerce product and the other i′-th e-commerce products;

[0022] Determine whether the sum of the confidence correction values ​​of the various types of e-commerce products exceeds a first preset threshold.

[0023] In a possible design, for each type of e-commerce product, the linkage purchase influence coefficients of the corresponding e-commerce product and other types of e-commerce products are calculated according to the purchase event indication value of the corresponding e-commerce product during each of the historical startup operations and the purchase event indication values ​​of other types of e-commerce products during each of the historical startup operations, including:

[0024] For a certain type of e-commerce product, extract the purchase event indicator value of the corresponding e-commerce product during the current M consecutive historical startup operations to form the first sample S 1 , where M represents a positive integer greater than m;

[0025] For other e-commerce products of a certain type relative to the e-commerce products of the certain type, the purchase event indication values ​​of the corresponding e-commerce products during the current most recent M consecutive historical startup operations are extracted to form a second sample S 2 ;

[0026] For the first sample S 1 Perform a normal distribution KS test to calculate the first test statistic p value p 1 , and also for the second sample S 2 Perform the normal distribution KS check to calculate the second check statistic p value p 2 ;

[0027] If the first test statistic p value p 1 and the second test statistic p-value p 2 are greater than the second preset threshold, the first sample S is calculated according to the following formula 1 With the sample S 2 The correlation coefficient r q :

[0028]

[0029] In the formula, i represents a positive integer, x i In the first sample S 1 The i-th sample value in y i In the second sample S 2 The i-th sample value in , The first sample S 1 The sample mean of The second sample S 2 The sample mean of ;

[0030] The correlation coefficient r q As the linkage purchase influence coefficient between the certain type of e-commerce products and the other certain type of e-commerce products.

[0031] In a possible design, after determining whether the sum of the confidences of the various e-commerce products exceeds a first preset threshold, if it is determined that the sum of the confidences of the various e-commerce products does not exceed the first preset threshold, the method further includes the following steps S701 to S704:

[0032] S701. Adjust the current e-commerce product recommendation plan to obtain a new e-commerce product recommendation plan, and then execute step S702;

[0033] S702. For each type of e-commerce product, the browsing event indicator values ​​or browsing time proportion values ​​of all e-commerce products during the most recent m-1 consecutive historical startups and runs and the new e-commerce product recommendation scheme are imported into the corresponding purchase event prediction model, and a new confidence level of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup and run is obtained as an output, and then step S703 is executed;

[0034] S703. Determine whether the sum of the new confidences of the various e-commerce products exceeds the first preset threshold. If so, execute step S704; otherwise, execute step S701;

[0035] S704. During the current next startup operation, the new e-commerce product recommendation plan is executed through the e-commerce platform application.

[0036] In a possible design, when there are multiple e-commerce product recommendation solutions, the method further includes:

[0037] For each of the various e-commerce products and each of the multiple e-commerce product recommendation schemes, the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the corresponding schemes are imported into the purchase event prediction model of the corresponding e-commerce product, and the event of the corresponding e-commerce product being purchased by the target customer during the current next startup operation and the corresponding confidence level are output;

[0038] For each of the e-commerce product recommendation solutions, the corresponding confidences of each type of e-commerce product are cumulatively calculated to obtain a corresponding total confidence value;

[0039] Determining whether the maximum confidence sum value exceeds the first preset threshold;

[0040] If so, during the current next startup operation, an e-commerce product recommendation scheme among the multiple e-commerce product recommendation schemes and corresponding to the maximum confidence sum value is executed through the e-commerce platform application.

[0041] In one possible design, the method further includes:

[0042] Applying the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the most recent m-1 consecutive historical startup runs and the purchase event prediction model of each type of e-commerce product, the e-commerce product recommendation scheme is optimized based on an optimization algorithm to obtain the e-commerce product recommendation scheme and an optimized search result for maximizing the sum of confidences of each type of e-commerce product;

[0043] During the current next startup operation, the e-commerce product recommendation solution based on the optimized search results is executed by the e-commerce platform application.

[0044] In a second aspect, an e-commerce product accurate recommendation device is provided, comprising a record data collection unit, a record data processing unit, a sample data extraction unit, a prediction model training unit, a prediction model application unit, a threshold comparison and judgment unit, and a recommendation scheme execution unit, which are sequentially connected in communication;

[0045] The record data collection unit is used to collect the e-commerce product browsing record data and e-commerce product purchase record data generated by the target customer during each historical startup and operation of the e-commerce platform application;

[0046] The record data processing unit is used to obtain browsing event indication values ​​or browsing time proportion values ​​of various types of e-commerce products during the various historical startups and operations according to the e-commerce product browsing record data, and to obtain purchase event indication values ​​of various types of e-commerce products during the various historical startups and operations according to the e-commerce product purchase record data, wherein the browsing event indication value is used to indicate whether an event of browsing the corresponding e-commerce product by the target customer occurs during the startup and operation of the e-commerce platform application, the browsing time proportion value refers to the ratio of the time duration of browsing the corresponding e-commerce product by the target customer to the total time duration of browsing all e-commerce products during the startup and operation of the e-commerce platform application, and the purchase event indication value is used to indicate whether an event of purchasing the corresponding e-commerce product by the target customer occurs during the startup and operation of the e-commerce platform application;

[0047] The sample data extraction unit is used to extract corresponding sample data for each type of e-commerce product in the following manner: taking the browsing event indication value or browsing time proportion value of all e-commerce products during the m consecutive historical startup operation periods as a model input item, and taking the purchase event indication value of the corresponding e-commerce product during the last historical startup operation period during the m consecutive historical startup operation periods as a model output item, and then taking the model input item and the model output item as a sample data, wherein m represents a positive integer greater than or equal to 2;

[0048] The prediction model training unit is used to apply all the corresponding sample data to calibrate and verify the machine learning model based on the artificial intelligence algorithm for each type of e-commerce product, so as to obtain a corresponding purchase event prediction model;

[0049] The prediction model application unit is used to import the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme of the e-commerce platform application during the current next startup operation into the corresponding purchase event prediction model for each type of e-commerce product, and output the confidence of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup operation, wherein the e-commerce product recommendation scheme includes the recommendation event indication values ​​or recommendation frequency proportion values ​​of each type of e-commerce product during the current next startup operation, the recommendation event indication value is used to indicate whether the event of the corresponding e-commerce product being recommended to the target customer for browsing is triggered during the current next startup operation, and the recommendation frequency proportion value refers to the ratio of the frequency of the corresponding e-commerce product being recommended to the target customer for browsing during the current next startup operation to the total frequency of all e-commerce products being recommended to the target customer for browsing;

[0050] The threshold comparison and judgment unit is used to judge whether the sum of the confidences of the various types of e-commerce products exceeds a first preset threshold;

[0051] The recommendation scheme execution unit is used to execute the e-commerce product recommendation scheme through the e-commerce platform application during the current next startup operation when it is determined that the total confidence of the various e-commerce products exceeds the first preset threshold.

[0052] In a third aspect, the present invention provides a computer device comprising a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the e-commerce product accurate recommendation method as described in the first aspect or any possible design of the first aspect.

[0053] In a fourth aspect, the present invention provides a computer-readable storage medium having instructions stored thereon. When the instructions are executed on a computer, the method for accurately recommending e-commerce products as described in the first aspect or any possible design of the first aspect is executed.

[0054] In a fifth aspect, the present invention provides a computer program product, including a computer program or instructions, which, when executed by a computer, implements the method for accurate recommendation of e-commerce products as described in the first aspect or any possible design of the first aspect.

[0055] Beneficial effects of the above scheme:

[0056] (1) The present invention creatively provides a new solution for accurately recommending e-commerce products in a multi-category e-commerce product recommendation scenario, namely, firstly extracting sample data of each type of e-commerce product based on e-commerce product browsing record data and e-commerce product purchase record data, and training a purchase event prediction model of each type of e-commerce product based on the sample data, and then estimating the confidence of the event that each type of e-commerce product is purchased by the target customer during the current next startup operation period based on the prediction model and the e-commerce product recommendation solution, and finally, when it is determined that the sum of the confidences of each type of e-commerce product exceeds a preset threshold, confirming that the recommendation accuracy of the recommendation solution meets the preset target, and executing the recommendation solution through the e-commerce platform application during the current next startup operation period, thereby taking into account both the impact of recommending a single e-commerce product on the customer's purchase decision and the situation that recommending different types of e-commerce products will have a competitive or linkage impact on the customer's purchase decision, so that when applied to a multi-category e-commerce product recommendation scenario, it can effectively improve the user's willingness to buy, achieve the purpose of further improving the accuracy of product recommendations, and facilitate practical application and promotion;

[0057] (2) By calculating the linkage purchase influence coefficient between e-commerce products based on the historical purchase situation of various e-commerce products and making confidence corrections, the estimation accuracy of user purchase intention can be improved by further considering the linkage purchase between different types of e-commerce products, thereby further improving the accuracy of product recommendations;

[0058] (3) When the recommendation accuracy of the recommendation scheme does not meet the preset target, the e-commerce product recommendation scheme can be continuously adjusted until the recommendation accuracy meets the preset target, thereby ensuring the e-commerce product recommendation effect;

[0059] (4) When there are multiple e-commerce product recommendation plans, the recommended plan can be selected to maximize the recommendation accuracy and meet the preset goals, thereby ensuring the best e-commerce product recommendation effect;

[0060] (5) It is also possible to find the best e-commerce product recommendation solution to maximize the accuracy of the recommendation, thereby ensuring the best e-commerce product recommendation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 A flowchart of a method for accurately recommending e-commerce products provided in an embodiment of the present application.

[0063] Figure 2 A supplementary flow chart of the method for accurately recommending e-commerce products when the sum of the confidence levels of various e-commerce products does not exceed a first preset threshold value provided in an embodiment of the present application.

[0064] Figure 3 A supplementary flow chart of the method for accurately recommending e-commerce products when there are multiple e-commerce product recommendation schemes provided in the embodiment of the present application.

[0065] Figure 4 A schematic diagram of the structure of an e-commerce product precision recommendation device provided in an embodiment of the present application.

[0066] Figure 5 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structures of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these embodiments without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0068] It should be understood that although the terms first and second, etc. may be used herein to describe various objects, these objects should not be limited by these terms. These terms are only used to distinguish one object from another object. For example, a first object can be referred to as a second object, and similarly, a second object can be referred to as a first object without departing from the scope of the exemplary embodiments of the present invention.

[0069] It should be understood that the term "and / or" that may appear in this document is merely a description of the association relationship between associated objects, indicating that there may be three relationships. For example, A and / or B can indicate three situations: A exists alone, B exists alone, or A and B exist at the same time. For another example, A, B and / or C can indicate the existence of any one of A, B and C or any combination of them. The term " / and" that may appear in this document describes another type of association object relationship, indicating that there may be two relationships. For example, A / and B can indicate two situations: A exists alone or A and B exist at the same time. In addition, the character " / " that may appear in this document generally indicates that the previous and next associated objects are in an "or" relationship.

[0070] Example

[0071] like Figure 1 As shown, the e-commerce product accurate recommendation method provided in the first aspect of this embodiment can be, but is not limited to, executed by a computer device with certain computing resources, such as a cloud server, a personal computer (PC, a multi-purpose computer with a size, price and performance suitable for personal use; desktops, laptops, small laptops, tablets and ultrabooks are all personal computers), smart phones, personal digital assistants (PDA) or wearable devices. Figure 1 As shown, the e-commerce product accurate recommendation method may include but is not limited to the following steps S1 to S75.

[0072] S1. Collect the e-commerce product browsing record data and e-commerce product purchasing record data generated by the target customers during each historical startup and operation of the e-commerce platform application.

[0073] In step S1, the e-commerce platform application refers to an APP (Application) program that can be started and run on a client terminal (such as a smartphone or tablet computer, etc.) to purchase e-commerce products online, and specific examples include but are not limited to Taobao APP, JD APP or Pinduoduo APP, etc. The e-commerce product browsing record data is used to reflect the e-commerce product browsing behavior made by the target customer during each historical startup and operation of the e-commerce platform application, and the e-commerce product purchase record data is used to reflect the e-commerce product purchase behavior made by the target customer during each historical startup and operation of the e-commerce platform application, and they can all be collected by conventionally recording and uploading the e-commerce platform application through existing technologies.

[0074] S2. Obtain browsing event indication values ​​or browsing time proportion values ​​of various e-commerce products during each historical startup and operation period according to the e-commerce product browsing record data processing, and obtain purchase event indication values ​​of various e-commerce products during each historical startup and operation period according to the e-commerce product purchase record data processing, wherein the browsing event indication value is used to indicate whether an event of browsing the corresponding e-commerce product by the target customer occurs during the startup and operation of the e-commerce platform application, the browsing time proportion value refers to the ratio of the time the target customer browses the corresponding e-commerce product to the total time the target customer browses all e-commerce products during the startup and operation of the e-commerce platform application, and the purchase event indication value is used to indicate whether an event of purchasing the corresponding e-commerce product by the target customer occurs during the startup and operation of the e-commerce platform application.

[0075] In the step S2, the specific processing method of the e-commerce product browsing record data and the e-commerce product purchase record data is the existing conventional data statistics method; for example, if there are Class A e-commerce products, Class B e-commerce products, Class C e-commerce products, Class D e-commerce products, Class E e-commerce products and Class F e-commerce products, and the e-commerce product browsing record data recorded during a certain historical startup operation of the e-commerce platform application reflects that the target user browsed Class B e-commerce products (the corresponding browsing time is 3 minutes), Class C e-commerce products (the corresponding browsing time is 5 minutes) and Class F e-commerce products (the corresponding browsing time is 2 minutes), and the e-commerce product purchase record data recorded during the certain historical startup operation reflects that the target user purchased Class C e-commerce products (which is generally one of the browsed products), then the browsing event indication values ​​of the Class A e-commerce products, the Class B e-commerce products, the Class C e-commerce products, the Class D e-commerce products, the Class E e-commerce products and the Class F e-commerce products during the certain historical startup operation can be processed to be 0, 1, 1, 0, 0 and 1 respectively (that is, for example but not limited to : 0 is used to indicate that no event of the corresponding e-commerce product being browsed by the target customer occurs during the certain historical startup operation, and 1 is used to indicate that an event of the corresponding e-commerce product being browsed by the target customer occurs during the certain historical startup operation), or it can be processed to obtain that the browsing time proportion values ​​of the Class A e-commerce products, the Class B e-commerce products, the Class C e-commerce products, the Class D e-commerce products, the Class E e-commerce products and the Class F e-commerce products during the certain historical startup operation are 0, 30%, 50%, 0, 0 and 20% respectively, and it can also be processed to obtain that the purchase event indication values ​​of the Class A e-commerce products, the Class B e-commerce products, the Class C e-commerce products, the Class D e-commerce products, the Class E e-commerce products and the Class F e-commerce products during the certain historical startup operation are 0, 0, 1, 0, 0 and 0 respectively (that is, for example but not limited to: 0 is used to indicate that no event of the corresponding e-commerce product being purchased by the target customer occurs during the certain historical startup operation, and 1 is used to indicate that an event of the corresponding e-commerce product being purchased by the target customer occurs during the certain historical startup operation). In addition, if each of the historical startup and operation periods is regarded as a time node, then for each type of e-commerce product, the browsing event indication values ​​or browsing time proportion values ​​that are arranged in chronological order and correspond to each of the historical startup and operation periods will constitute the corresponding time series data, and the purchase event indication values ​​that are arranged in chronological order and correspond to each of the historical startup and operation periods will constitute another corresponding time series data.

[0076] S3. For each type of e-commerce product, the corresponding sample data is extracted in the following manner: the browsing event indication values ​​or browsing time proportion values ​​of all e-commerce products during the m consecutive historical startup operations are taken as model input items, and the purchase event indication value of the corresponding e-commerce product during the last historical startup operation period among the m consecutive historical startup operations is taken as the model output item, and then the model input item and the model output item are taken as a sample data, wherein m represents a positive integer greater than or equal to 2.

[0077] In step S3, based on the example of step S2, for the B-type e-commerce products, the browsing event indication values ​​or browsing time proportion values ​​of the A-type e-commerce products, the B-type e-commerce products, the C-type e-commerce products, the D-type e-commerce products, the E-type e-commerce products, and the F-type e-commerce products during the m consecutive historical startups can be used as model input items (specifically a 6×m matrix), and the purchase event indication value of the B-type e-commerce products during the last historical startup during the m consecutive historical startups can be used as a model output item (specifically 0 or 1). In addition, the specific value of m can be pre-set according to actual needs or optimization algorithms, for example, it is designed to be 5.

[0078] S4. For each type of e-commerce product, apply all the corresponding sample data to calibrate and verify the machine learning model based on the artificial intelligence algorithm to obtain the corresponding purchase event prediction model.

[0079] In the step S4, the artificial intelligence algorithm is a core artificial intelligence algorithm that specifically studies how computers simulate or implement human learning behavior to acquire new knowledge or skills, reorganize existing knowledge structures to continuously improve their own performance, and is the fundamental way to make computers intelligent. Specifically, the machine learning model can be, but is not limited to, an LSTM model, a Bi-LSTM model, or an Attention-LSTM model. The aforementioned LSTM (Long Short-Term Memory, long short-term memory network) model, Bi-LSTM model, or Attention-LSTM model are all existing machine learning models, wherein the Attention-LSTM model is an improved model that adds a layer of attention mechanism to the existing LSTM model (which is commonly used for time series data prediction), and the added attention mechanism is used to allow the LSTM model to dynamically focus on part of the information during training, so as to be able to capture more important information and improve the performance of the model. Therefore, this embodiment preferably uses the Attention-LSTM model to train the purchase event prediction model. The specific calibration and verification modeling process of the machine learning model (wherein model calibration is to adjust parameters or structure to make the model fit the actual data; and model verification is to evaluate the accuracy and applicability of the model by comparing the model output with the actual data) is a prior art means; specifically, the parameters of the machine learning model can be searched by but not limited to a grid search algorithm or a Bayesian search algorithm.

[0080] S5. For each type of e-commerce product, the browsing event indication value or browsing time proportion value of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme of the e-commerce platform application during the current next startup operation are imported into the corresponding purchase event prediction model, and the confidence of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup operation is output, wherein the e-commerce product recommendation scheme includes but is not limited to the recommendation event indication value or recommendation frequency proportion value of each type of e-commerce product during the current next startup operation, etc. The recommendation event indication value is used to indicate whether the event of the corresponding e-commerce product being recommended to the target customer for browsing is triggered during the current next startup operation, and the recommendation frequency proportion value refers to the ratio of the frequency of the corresponding e-commerce product being recommended to the target customer for browsing during the current next startup operation to the total frequency of all e-commerce products being recommended to the target customer for browsing during the current next startup operation.

[0081] In step S5, the e-commerce product recommendation scheme is used as an e-commerce product recommendation plan, which can be randomly generated or generated by the e-commerce platform service provider, wherein the recommendation event indicator value is used to be equivalent to the browsing event indicator value (that is, because the e-commerce product must be browsed by the target customer after being recommended, and for example but not limited to: 0 is used to indicate that the event of the corresponding e-commerce product being recommended to the target customer for browsing is not triggered during the current next startup operation, and 1 is used to indicate that the event of the corresponding e-commerce product being recommended to the target customer for browsing is triggered during the current next startup operation), and the recommendation frequency percentage value is used to be equivalent to the browsing time percentage value (that is, because the recommendation frequency must be positively correlated with the browsing time). Based on the example of step S2, for the Class B e-commerce products, the browsing event indication values ​​or browsing time proportion values ​​of the Class A e-commerce products, the Class B e-commerce products, the Class C e-commerce products, the Class D e-commerce products, the Class E e-commerce products and the Class F e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme of the e-commerce platform application during the current next startup operation (the scheme is a 6×1 matrix, and the entire imported data can be made into a 6×m matrix) are imported into the corresponding purchase event prediction model, and the confidence level of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup operation is output (i.e., the confidence level of the output value 1). Specifically, when the purchase event prediction model is obtained by calibration and verification modeling based on browsing event indication values ​​and purchase event indication values, the data imported into the purchase event prediction model in step S5 are specifically the browsing event indication values ​​of all e-commerce products during the current most recent m-1 consecutive historical startup operations and the recommended event indication values ​​of each type of e-commerce product during the current next startup operation, and when the purchase event prediction model is obtained by calibration and verification modeling based on browsing time proportion values ​​and purchase event indication values, the data imported into the purchase event prediction model in step S5 are the browsing time proportion values ​​of all e-commerce products during the current most recent m-1 consecutive historical startup operations and the recommended frequency proportion values ​​of each type of e-commerce product during the current next startup operation. In addition, the aforementioned step S5 and the subsequent step S6 are specifically performed between the end time of the current most recent historical startup operation period and the start time of the current next startup operation period.

[0082] S6. Determine whether the sum of the confidence levels of the various types of e-commerce products exceeds a first preset threshold.

[0083] In step S6, the sum of the confidences of the various e-commerce products reflects the total purchase intention of the target customers in the multi-category e-commerce product recommendation scenario. The higher the value, the stronger the total purchase intention and the higher the product recommendation accuracy. The first preset threshold is used as a basis for judging whether the recommendation accuracy of the e-commerce product recommendation scheme meets the preset target, which can be pre-set by the platform service provider. In addition, the aforementioned step S6 can also be replaced by: judging whether the average confidence value of the various e-commerce products exceeds another preset threshold (which can also be pre-set by the platform service provider).

[0084] S7. If yes, then during the next startup operation, the e-commerce product recommendation solution is executed through the e-commerce platform application.

[0085] In the step S7, based on the example of step S2, if the specific content of the e-commerce product recommendation plan includes: the recommendation event indication values ​​of the Class A e-commerce products, the Class B e-commerce products, the Class C e-commerce products, the Class D e-commerce products, the Class E e-commerce products and the Class F e-commerce products during the current next startup operation are 1, 0, 0, 1, 0 and 0 respectively, then during the current next startup operation, the e-commerce product recommendation plan is executed through the e-commerce platform application, specifically including: during the current next startup operation, only recommending the Class A e-commerce products and the Class D e-commerce products through the e-commerce platform application. If the specific content of the e-commerce product recommendation scheme includes: the recommendation frequency proportions of the Class A e-commerce products, the Class B e-commerce products, the Class C e-commerce products, the Class D e-commerce products, the Class E e-commerce products and the Class F e-commerce products during the current next startup operation period are 10%, 30%, 30%, 30%, 0 and 0 respectively, then during the current next startup operation period, the e-commerce product recommendation scheme is executed through the e-commerce platform application, specifically including: among every 10 recommended e-commerce products, 1 e-commerce product is the Class A e-commerce product, 3 e-commerce products are the Class B e-commerce products, 3 e-commerce products are the Class C e-commerce products and 3 e-commerce products are the Class D e-commerce products, or each time an e-commerce product is recommended, the Class A e-commerce product is recommended with a probability of 10%, the Class B e-commerce product is recommended with a probability of 30%, the Class C e-commerce product is recommended with a probability of 30% and the Class D e-commerce product is recommended with a probability of 30%.

[0086] Therefore, based on the e-commerce product accurate recommendation method described in the above steps S1 to S7, a new solution for accurate recommendation of e-commerce products in multi-category e-commerce product recommendation scenarios is provided, that is, firstly, based on the e-commerce product browsing record data and the e-commerce product purchase record data, sample data of various e-commerce products are extracted, and purchase event prediction models of various e-commerce products are trained based on the sample data, and then, based on the prediction model and the e-commerce product recommendation scheme, the confidence of the event that various e-commerce products are purchased by the target customer during the current next startup operation is estimated, and finally, when it is determined that the sum of the confidences of various e-commerce products exceeds the preset threshold, it is confirmed that the recommendation accuracy of the recommendation scheme meets the preset target, and during the current next startup operation, the recommendation scheme is executed through the e-commerce platform application. In this way, the influence of recommending a single e-commerce product on the customer's purchase decision is taken into account, and the situation that recommending different types of e-commerce products will have a competitive or linkage influence on the customer's purchase decision is taken into account, so that when applied to multi-category e-commerce product recommendation scenarios, the user's willingness to buy can be effectively improved, and the purpose of further improving the accuracy of product recommendations is achieved, which is convenient for practical application and promotion.

[0087] Based on the technical solution of the first aspect mentioned above, this embodiment further provides a possible design for obtaining the linkage purchase influence coefficient between e-commerce products and performing confidence correction, that is, the method also includes but is not limited to the following step S8.

[0088] S8. For each type of e-commerce product, the linkage purchase influence coefficients of the corresponding e-commerce product and other types of e-commerce products are calculated based on the purchase event indication values ​​of the corresponding e-commerce product during each of the historical startup operations and the purchase event indication values ​​of other types of e-commerce products during each of the historical startup operations.

[0089] In step S8, based on the example of step S2, for the B-type e-commerce products, the A-type e-commerce products, the C-type e-commerce products, the D-type e-commerce products, the E-type e-commerce products and the F-type e-commerce products are the corresponding other types of e-commerce products. Specifically, for each type of e-commerce product, according to the purchase event indication value of the corresponding e-commerce product during each historical startup operation and the purchase event indication value of the other types of e-commerce products during each historical startup operation, the linkage purchase influence coefficient of the corresponding e-commerce product and the other types of e-commerce products is calculated respectively, including but not limited to the following steps S81 to S85.

[0090] S81. For a certain type of e-commerce product, extract the purchase event indication value of the corresponding e-commerce product during the current M consecutive historical startup operations to form a first sample S 1 , where M represents a positive integer greater than m.

[0091] In step S81, if the value of m is 5, then the value of M can be 50, for example.

[0092] S82. For another type of e-commerce product relative to the type of e-commerce product, extract the purchase event indication value of the corresponding e-commerce product during the current M consecutive historical startup operations to form a second sample S 2 .

[0093] S83. For the first sample S 1 Perform a normal distribution KS test to calculate the first test statistic p value p 1 , and also for the second sample S 2 Perform the normal distribution KS check to calculate the second check statistic p value p 2 .

[0094] S84. If the first check statistic p value is p 1 and the second test statistic p-value p 2 are greater than the second first preset threshold, then the first sample S is calculated according to the following formula 1 With the sample S 2 The correlation coefficient r q :

[0095]

[0096] In the formula, i represents a positive integer, x i In the first sample S 1 The i-th sample value in y i In the second sample S 2 The i-th sample value in , The first sample S 1 The sample mean of The second sample S 2 The sample mean of .

[0097] In the step S84, the second preset threshold may be 0.05, for example.

[0098] S85. The correlation coefficient r q As the linkage purchase influence coefficient between the certain type of e-commerce products and the other certain type of e-commerce products.

[0099] Determining whether the sum of the confidences of the various e-commerce products exceeds a first preset threshold includes but is not limited to the following steps S61 to S62.

[0100] S61. For each type of e-commerce product, according to the corresponding confidence and the linkage purchase influence coefficient between the corresponding e-commerce product and other types of e-commerce products, the corresponding confidence correction value is calculated according to the following formula:

[0101]

[0102] In the formula, N represents the total number of categories of e-commerce products, i represents a positive integer less than or equal to N, and c i represents the confidence of the i-th e-commerce product, represents the confidence correction value of the i-th e-commerce product, i′ represents a positive integer less than or equal to N-1, c i,i′ represents the confidence of other i′th category e-commerce products relative to the i-th category e-commerce products, η i,i′ It represents the linkage purchase influence coefficient between the i-th e-commerce product and the other i′-th e-commerce products.

[0103] S62. Determine whether the sum of the confidence correction values ​​of the various types of e-commerce products exceeds a first preset threshold.

[0104] Therefore, based on the aforementioned possible design one, by calculating the linkage purchase influence coefficient between e-commerce products based on the historical purchase situation of various e-commerce products and performing confidence correction, the estimation accuracy of user purchase intention can be improved by further considering the linkage purchase between different types of e-commerce products, thereby further improving the accuracy of product recommendation.

[0105] Based on the technical solution of the first aspect or possible design one, this embodiment further provides a possible design two for adjusting and re-evaluating the recommendation scheme when the recommendation accuracy of the recommendation scheme does not meet the preset target, that is, after determining whether the sum of the confidences of the various e-commerce products exceeds the first preset threshold, if it is determined that the sum of the confidences of the various e-commerce products does not exceed the first preset threshold, then Figure 2 As shown, the method also includes but is not limited to the following steps S701 to S704.

[0106] S701. Adjust the current e-commerce product recommendation plan to obtain a new e-commerce product recommendation plan, and then execute step S702.

[0107] In the step S701, the specific adjustment method of the e-commerce product recommendation scheme is the existing conventional method, such as randomly modifying the corresponding recommendation event indicator value or recommendation frequency ratio value of a certain type of e-commerce product.

[0108] S702. For each type of e-commerce product, the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the new recommended solutions for the e-commerce products are imported into the corresponding purchase event prediction model, and the new confidence level of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup operation is output, and then step S703 is executed.

[0109] In the step S702, the specific technical details can be found in the aforementioned step S5, which will not be repeated here.

[0110] S703. Determine whether the sum of the new confidences of the various e-commerce products exceeds the first preset threshold. If so, execute step S704; otherwise, execute step S701.

[0111] S704. During the current next startup operation, the new e-commerce product recommendation plan is executed through the e-commerce platform application.

[0112] Therefore, based on the aforementioned possible design 2, when the recommendation accuracy of the recommendation scheme does not meet the preset target, the e-commerce product recommendation scheme can be continuously adjusted until the recommendation accuracy meets the preset target, thereby ensuring the e-commerce product recommendation effect.

[0113] Based on the technical solution of the first aspect or possible design one, this embodiment further provides a possible design three for selecting the best recommendation scheme when there are multiple e-commerce product recommendation schemes, that is, when there are multiple e-commerce product recommendation schemes, such as Figure 3 As shown, the method also includes but is not limited to the following steps S91 to S94.

[0114] S91. For each type of e-commerce product and each e-commerce product recommendation scheme in the multiple e-commerce product recommendation schemes, the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the corresponding schemes are imported into the purchase event prediction model of the corresponding e-commerce product, and the event of the corresponding e-commerce product being purchased by the target customer during the current next startup operation and the corresponding confidence level are output.

[0115] In the step S91, the specific technical details can be found in the aforementioned step S5, which will not be repeated here.

[0116] S92. For each of the e-commerce product recommendation solutions, the corresponding confidences of each type of e-commerce product are cumulatively calculated to obtain a corresponding total confidence value.

[0117] S93. Determine whether the maximum confidence sum value exceeds the first preset threshold.

[0118] S94. If yes, during the next startup operation, an e-commerce product recommendation scheme among the multiple e-commerce product recommendation schemes and corresponding to the maximum confidence sum value is executed through the e-commerce platform application.

[0119] Therefore, based on the aforementioned possible design three, when there are multiple e-commerce product recommendation plans, the recommended plan can be selected to maximize the recommendation accuracy and meet the preset goals, thereby ensuring the best e-commerce product recommendation effect.

[0120] Based on the technical solution of the aforementioned first aspect or possible design one, this embodiment also provides a possible design four for how to obtain the optimal e-commerce product recommendation solution, that is, the method also includes but is not limited to the following steps S10 to S11.

[0121] S10. Apply the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the purchase event prediction model of each type of e-commerce product, and optimize the e-commerce product recommendation scheme based on the optimization algorithm to obtain the e-commerce product recommendation scheme and the optimized search result for maximizing the sum of the confidence scores of each type of e-commerce product.

[0122] In the step S10, specifically, the optimization algorithm may be, but is not limited to, a particle swarm optimization algorithm, a Newton optimization algorithm, a genetic optimization algorithm, a Grey Wolf Algorithm, a whale optimization algorithm, or a tuna school optimization algorithm. Considering that different optimization algorithms have different advantages and disadvantages, in order to integrate the performance of the aforementioned optimization algorithms to achieve the purpose of leveraging strengths and avoiding weaknesses, this embodiment preferably applies the browsing event indication value or browsing time proportion value of all e-commerce products during the current most recent m-1 consecutive historical startup operations and the purchase event prediction model of each type of e-commerce product, optimizes the e-commerce product recommendation scheme based on the optimization algorithm, obtains the e-commerce product recommendation scheme and is used to maximize the optimal search results for the sum of the confidences of the various types of e-commerce products, including but not limited to the following steps S101 to S116.

[0123] S101. Initialize the number of search individuals I, the maximum number of iterations T, and the first learning factor α of the search population 1 , the second learning factor α 2 , the third learning factor α 3 、The fourth learning factor α 4 、Fifth learning factor α 5 and the sixth learning factor α 6The optimization algorithm parameters are set, and the initial position of each search individual in the search population is randomly generated, and then step S102 is performed, wherein the initial position of each search individual is randomly generated based on the following formula:

[0124]

[0125] In the formula, i′ represents a positive integer less than or equal to I, d′ represents a positive integer less than or equal to D′, D′ represents the number of dimensions of the parameter vector to be optimized, and the parameter vector to be optimized includes the e-commerce product recommendation solution, represents the component of the initial position of the i′th search individual in the search population on the d′th dimension of the parameter vector to be optimized, u c,d′ represents the upper limit of the parameter search space in the d′th dimension, l c,d′ represents the lower limit of the parameter search space in the d′th dimension, and rand(0,1) represents a pure decimal random generation function.

[0126] S102. For each search individual, the corresponding initial position is used as the e-commerce product recommendation plan, and then the browsing event indication value or browsing time proportion value of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation plan are respectively imported into the purchase event prediction model of each type of e-commerce product, and the first confidence level of the event that the each type of e-commerce product is purchased by the target customer during the current next startup operation is output, and finally the cumulative sum of each of the first confidence levels corresponding to each type of e-commerce product is used as the corresponding fitness, and then step S103 is executed.

[0127] In step S102, the specific technical details can be found in the aforementioned step S5, which will not be described in detail here. In addition, considering that some values ​​(such as the recommended event indication value) in the e-commerce product recommendation scheme need to be integers, when the initial position is used as the e-commerce product recommendation scheme, it is necessary to round the corresponding values, such as rounding the recommended event indication value.

[0128] S103. Take the initial position of a search individual with the maximum fitness as the initial global optimal position x best , and also initialize and set the current iteration number t′=0, and then execute step S104.

[0129] S104. Calculate the current average position of the search population according to the current position of each search individual, and determine the corresponding first new position for each search individual, and then execute step S105, wherein the first new position of each search individual is determined according to the following formula:

[0130] x new1,i′,d′ =x best,d′ +α 1 ×rand(0,1)×(x mean,d′ -x i′,d′ )

[0131] In the formula, x new1,i′,d′ represents the component of the first new position of the i′th search individual in the d′th dimension, x best,d′ Denotes the global optimal position x best The component in the d′th dimension, x mean,d′ represents the component of the current average position of the search population in the d′th dimension, x i′,d′ Represents the component of the current position of the i′th search individual in the d′th dimension.

[0132] S105. For each search individual, the corresponding first new position is used as an e-commerce product recommendation scheme, and then the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme are respectively imported into the purchase event prediction model of each type of e-commerce product, and the second confidence level of the event that the each type of e-commerce product is purchased by the target customer during the current next startup operation is outputted, and finally the cumulative sum of each of the second confidence levels corresponding one by one to the each type of e-commerce product is used as the corresponding and new fitness level, and then step S106 is executed.

[0133] In step S105, the specific technical details can be found in the aforementioned step S5, which will not be described in detail here. In addition, considering that some values ​​(such as the recommended event indication value) in the e-commerce product recommendation scheme need to be integers, when the first new position is used as the e-commerce product recommendation scheme, it is necessary to round the corresponding values, such as rounding the recommended event indication value.

[0134] S106. Determine whether the fitness corresponding to the current position of the i′th search individual is less than the current fitness of the i′th search individual. If so, update the current position of the i′th search individual to the first new position of the i′th search individual, and then execute step S107; otherwise, directly execute step S107.

[0135] S107. Determine the global optimal position x best Is the corresponding fitness less than the minimum value among the current fitness of each search individual? If so, the global optimal position x bestUpdate to the current position of any search individual with the minimum value, and then execute step S108, otherwise directly execute step S108.

[0136] S108. The current average position of the search population is calculated based on the current position of each search individual, and the corresponding second new position is determined for each search individual, and then step S109 is executed, wherein the second new position of each search individual is determined according to the following formula:

[0137]

[0138] In the formula, x new2,i′,d′ represents the component of the second new position of the i′th search individual in the d′th dimension, β(i′) represents the first intermediate variable corresponding to the i′th search individual, δ(i′) represents the second intermediate variable corresponding to the i′th search individual, i″ represents a positive integer less than or equal to I, mod() represents a remainder function, x i″,d′ represents the component of the current position of the i′-th search individual in the search population on the d′-th dimension, βr(i′) represents the third intermediate variable corresponding to the i′-th search individual, δr(i′) represents the fourth intermediate variable corresponding to the i′-th search individual, max() represents the maximum value function, θ(i′) represents the fifth intermediate variable corresponding to the i′-th search individual, r(i′) represents the sixth intermediate variable corresponding to the i′-th search individual, and π represents 180 degrees.

[0139] S109. For each search individual, the corresponding second new position is used as an e-commerce product recommendation scheme, and then the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme are respectively imported into the purchase event prediction model of each type of e-commerce product, and the third confidence level of the event that the each type of e-commerce product is purchased by the target customer during the current next startup operation is output, and finally the cumulative sum of each third confidence level corresponding to each type of e-commerce product is used as the corresponding and new fitness level, and then step S110 is executed.

[0140] In step S109, the specific technical details can be found in the aforementioned step S5, which will not be described in detail here. In addition, considering that some values ​​in the e-commerce product recommendation scheme (such as the recommended event indication value) need to be integers, when the second new position is used as the e-commerce product recommendation scheme, it is necessary to round the corresponding values, such as rounding the recommended event indication value.

[0141] S110. Determine whether the fitness corresponding to the current position of the i′th search individual is less than the current fitness of the i′th search individual. If so, update the current position of the i′th search individual to the second new position of the i′th search individual, and then execute step S111; otherwise, directly execute step S111.

[0142] S111. Determine the global optimal position x best Is the corresponding fitness less than the minimum value among the current fitness of each search individual? If so, the global optimal position x best Update to the current position of any search individual with the minimum value, and then execute step S112, otherwise directly execute step S112.

[0143] S112. The current average position of the search population is calculated and updated according to the current position of each search individual, and the corresponding third new position is determined for each search individual, and then step S113 is executed, wherein the third new position of each search individual is determined according to the following formula:

[0144]

[0145] In the formula, x new3,i′,d′ represents the component of the third new position of the i′th search individual in the d′th dimension, represents the seventh intermediate variable corresponding to the i′th search individual, represents the eighth intermediate variable corresponding to the i′th search individual, represents the ninth intermediate variable corresponding to the i′th search individual, represents the tenth intermediate variable corresponding to the i′th search individual, Represents the eleventh intermediate variable corresponding to the i′th search individual.

[0146] S113. For each of the search individuals, the corresponding third new position is used as an e-commerce product recommendation scheme, and then the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme are respectively imported into the purchase event prediction model of each type of e-commerce product, and the fourth confidence level of the event that the each type of e-commerce product is purchased by the target customer during the current next startup operation is output, and finally the cumulative sum of each of the fourth confidence levels corresponding to each type of e-commerce product is used as the corresponding and new fitness level, and then step S114 is executed.

[0147] In step S113, the specific technical details can be found in the aforementioned step S5, which will not be described in detail here. In addition, considering that some values ​​(such as the recommended event indication value) in the e-commerce product recommendation scheme need to be integers, when the third new position is used as the e-commerce product recommendation scheme, it is necessary to round the corresponding values, such as rounding the recommended event indication value.

[0148] S114. Determine whether the fitness corresponding to the current position of the i′th search individual is less than the current fitness of the i′th search individual. If so, update the current position of the i′th search individual to the third new position of the i′th search individual, and then execute step S115; otherwise, directly execute step S115.

[0149] S115. Determine the global optimal position x best Is the corresponding fitness less than the minimum value among the current fitness of each search individual? If so, the global optimal position x best Update to the current position of any search individual with the minimum value, and then execute step S116, otherwise directly execute step S116.

[0150] S116. Increment the current number of iterations t′ by 1, and determine whether the current number of iterations t′ reaches the maximum number of iterations T. If so, set the global optimal position x best The optimized search result is used as the e-commerce product recommendation scheme and is used to maximize the sum of the confidences of the various e-commerce products; otherwise, the process returns to step S104.

[0151] Based on the aforementioned steps S101 to S116, since each search individual performs three fitness calculations with the help of spirals and group centers in each iteration process, it can be regarded as a fusion of the particle swarm optimization algorithm and the whale optimization algorithm. Therefore, under all the same conditions, its optimization performance will be better than other optimization algorithms, which is conducive to quickly and accurately obtaining the e-commerce product recommendation plan and the optimized search result for maximizing the total confidence of each type of e-commerce product.

[0152] S11. During the current next startup operation, the e-commerce platform application executes an e-commerce product recommendation solution based on the optimized search results.

[0153] In the step S11, it is considered that some values ​​in the e-commerce product recommendation scheme (such as the recommended event indication value) need to be integers. Therefore, in the process of obtaining the e-commerce product recommendation scheme based on the optimized search results, it is necessary to round the corresponding values, such as rounding the recommended event indication value.

[0154] Based on the above possible design four, the optimal e-commerce product recommendation scheme can also be optimized to maximize the recommendation accuracy, thereby ensuring the optimal e-commerce product recommendation effect. In addition, the above steps S10 to S11 can be specifically performed when it is determined that the total confidence value / maximum total confidence value of the various e-commerce products does not exceed the first preset threshold value, so as to optimize the optimal e-commerce product recommendation scheme.

[0155] like Figure 4 As shown, the second aspect of this embodiment provides a virtual device for implementing the e-commerce product accurate recommendation method described in the first aspect or any possible design in the first aspect, including a record data collection unit, a record data processing unit, a sample data extraction unit, a prediction model training unit, a prediction model application unit, a threshold comparison and judgment unit, and a recommendation scheme execution unit that are sequentially connected in communication;

[0156] The record data collection unit is used to collect the e-commerce product browsing record data and e-commerce product purchase record data generated by the target customer during each historical startup and operation of the e-commerce platform application;

[0157] The record data processing unit is used to obtain browsing event indication values ​​or browsing time proportion values ​​of various types of e-commerce products during the various historical startups and operations according to the e-commerce product browsing record data, and to obtain purchase event indication values ​​of various types of e-commerce products during the various historical startups and operations according to the e-commerce product purchase record data, wherein the browsing event indication value is used to indicate whether an event of browsing the corresponding e-commerce product by the target customer occurs during the startup and operation of the e-commerce platform application, the browsing time proportion value refers to the ratio of the time duration of browsing the corresponding e-commerce product by the target customer to the total time duration of browsing all e-commerce products during the startup and operation of the e-commerce platform application, and the purchase event indication value is used to indicate whether an event of purchasing the corresponding e-commerce product by the target customer occurs during the startup and operation of the e-commerce platform application;

[0158] The sample data extraction unit is used to extract corresponding sample data for each type of e-commerce product in the following manner: taking the browsing event indication value or browsing time proportion value of all e-commerce products during the m consecutive historical startup operation periods as a model input item, and taking the purchase event indication value of the corresponding e-commerce product during the last historical startup operation period during the m consecutive historical startup operation periods as a model output item, and then taking the model input item and the model output item as a sample data, wherein m represents a positive integer greater than or equal to 2;

[0159] The prediction model training unit is used to apply all the corresponding sample data to calibrate and verify the machine learning model based on the artificial intelligence algorithm for each type of e-commerce product, so as to obtain a corresponding purchase event prediction model;

[0160] The prediction model application unit is used to import the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme of the e-commerce platform application during the current next startup operation into the corresponding purchase event prediction model for each type of e-commerce product, and output the confidence of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup operation, wherein the e-commerce product recommendation scheme includes the recommendation event indication values ​​or recommendation frequency proportion values ​​of each type of e-commerce product during the current next startup operation, the recommendation event indication value is used to indicate whether the event of the corresponding e-commerce product being recommended to the target customer for browsing is triggered during the current next startup operation, and the recommendation frequency proportion value refers to the ratio of the frequency of the corresponding e-commerce product being recommended to the target customer for browsing during the current next startup operation to the total frequency of all e-commerce products being recommended to the target customer for browsing;

[0161] The threshold comparison and judgment unit is used to judge whether the sum of the confidences of the various types of e-commerce products exceeds a first preset threshold;

[0162] The recommendation scheme execution unit is used to execute the e-commerce product recommendation scheme through the e-commerce platform application during the current next startup operation when it is determined that the total confidence of the various e-commerce products exceeds the first preset threshold.

[0163] The working process, working details and technical effects of the aforementioned device provided in the second aspect of this embodiment can be referred to the first aspect or any possible design of the e-commerce product accurate recommendation method described in the first aspect, and will not be repeated here.

[0164] like Figure 5As shown, the third aspect of this embodiment provides a computer device for executing the e-commerce product accurate recommendation method as described in the first aspect or any possible design in the first aspect, including a memory, a processor and a transceiver connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program, and execute the e-commerce product accurate recommendation method as described in the first aspect or any possible design in the first aspect. For example, the memory may include, but is not limited to, a random access memory (Random-Access Memory, RAM), a read-only memory (Read-Only Memory, ROM), a flash memory (Flash Memory), a first-in-first-out memory (First Input First Output, FIFO) and / or a first-in-last-out memory (First Input Last Output, FILO), etc.; the processor may be, but is not limited to, a microprocessor of the STM32F105 series. In addition, the computer device may also include, but is not limited to, a power module, a display screen and other necessary components.

[0165] The working process, working details and technical effects of the aforementioned computer device provided in the third aspect of this embodiment can be referred to the first aspect or any possible design of the e-commerce product accurate recommendation method described in the first aspect, and will not be repeated here.

[0166] In a fourth aspect of this embodiment, there is provided a computer-readable storage medium storing instructions for the e-commerce product accurate recommendation method as described in the first aspect or any possible design in the first aspect, that is, the computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the e-commerce product accurate recommendation method as described in the first aspect or any possible design in the first aspect is executed. The computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to computer-readable storage media such as floppy disks, optical disks, hard disks, flash memories, USB flash drives, and / or memory sticks, and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0167] The working process, working details and technical effects of the aforementioned computer-readable storage medium provided in the fourth aspect of this embodiment can be referred to the e-commerce product accurate recommendation method described in the first aspect or any possible design in the first aspect, and will not be repeated here.

[0168] In a fifth aspect, the present embodiment provides a computer program product, including a computer program or an instruction, wherein the computer program or the instruction, when executed by a computer, implements the e-commerce product accurate recommendation method as described in the first aspect or any possible design in the first aspect. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0169] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for accurately recommending e-commerce products, characterized in that: include: Collect e-commerce product browsing record data and e-commerce product purchase record data generated by target customers during each historical startup and operation of the e-commerce platform application; According to the e-commerce product browsing record data, the browsing event indication value or browsing time proportion value of each type of e-commerce product during each historical startup operation is obtained, and according to the e-commerce product purchase record data, the purchase event indication value of each type of e-commerce product during each historical startup operation is obtained, wherein the browsing event indication value is used to indicate whether an event of browsing the corresponding e-commerce product by the target customer occurs during the startup operation of the e-commerce platform application, the browsing time proportion value refers to the ratio of the time duration of browsing the corresponding e-commerce product by the target customer to the total time duration of browsing all e-commerce products during the startup operation of the e-commerce platform application, and the purchase event indication value is used to indicate whether an event of purchasing the corresponding e-commerce product by the target customer occurs during the startup operation of the e-commerce platform application; For each type of e-commerce product, the corresponding sample data is extracted in the following manner: the browsing event indication values ​​or browsing time proportion values ​​of all e-commerce products during the m consecutive historical startup runs are taken as model input items, and the purchase event indication value of the corresponding e-commerce product during the last historical startup run during the m consecutive historical startup runs is taken as the model output item, and then the model input item and the model output item are taken as a sample data, wherein m represents a positive integer greater than or equal to 2; For each type of e-commerce product, all the corresponding sample data are used to calibrate and validate the machine learning model based on the artificial intelligence algorithm to obtain the corresponding purchase event prediction model; For each type of e-commerce product, the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme of the e-commerce platform application during the current next startup operation are imported into the corresponding purchase event prediction model, and the confidence of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup operation is output, wherein the e-commerce product recommendation scheme includes the recommendation event indication values ​​or recommendation frequency proportion values ​​of each type of e-commerce product during the current next startup operation, the recommendation event indication value is used to indicate whether the event of the corresponding e-commerce product being recommended to the target customer for browsing is triggered during the current next startup operation, and the recommendation frequency proportion value refers to the ratio of the frequency of the corresponding e-commerce product being recommended to the target customer for browsing during the current next startup operation to the total frequency of all e-commerce products being recommended to the target customer for browsing; Determining whether the sum of the confidence levels of the various e-commerce products exceeds a first preset threshold; If so, during the next startup operation, the e-commerce product recommendation solution is executed through the e-commerce platform application.

2. The e-commerce product accurate recommendation method according to claim 1, characterized in that: The method further comprises: For each type of e-commerce product, based on the purchase event indication value of the corresponding e-commerce product during each of the historical startup and operation periods and the purchase event indication value of other types of e-commerce products during each of the historical startup and operation periods, respectively calculate the linkage purchase influence coefficient of the corresponding e-commerce product and other types of e-commerce products; Determining whether the sum of the confidences of the various e-commerce products exceeds a first preset threshold includes: For each type of e-commerce product, according to the corresponding confidence and the linkage purchase influence coefficient between the corresponding e-commerce product and other types of e-commerce products, the corresponding confidence correction value is calculated according to the following formula: In the formula, N represents the total number of categories of e-commerce products, i represents a positive integer less than or equal to N, and c i represents the confidence of the i-th e-commerce product, represents the confidence correction value of the i-th e-commerce product, i′ represents a positive integer less than or equal to N-1, c i,i′ represents the confidence of other i′th category e-commerce products relative to the i-th category e-commerce products, η i,i′ represents the linkage purchase influence coefficient between the i-th e-commerce product and the other i′-th e-commerce products; Determine whether the sum of the confidence correction values ​​of the various types of e-commerce products exceeds a first preset threshold.

3. The e-commerce product accurate recommendation method according to claim 2, characterized in that: For each type of e-commerce product, based on the purchase event indication value of the corresponding e-commerce product during each of the historical startup operations and the purchase event indication value of other types of e-commerce products during each of the historical startup operations, the linkage purchase influence coefficients of the corresponding e-commerce product and other types of e-commerce products are calculated respectively, including: For a certain type of e-commerce product, extract the purchase event indication value of the corresponding e-commerce product during the most recent M consecutive historical startup operations to form a first sample S1, where M represents a positive integer greater than m; For another type of e-commerce product relative to the type of e-commerce product, extract the purchase event indication value of the corresponding e-commerce product during the current most recent M consecutive historical startup operations to form a second sample S2; Performing a normal distribution KS check on the first sample S1 to calculate a first check statistic p value p1, and also performing the normal distribution KS check on the second sample S2 to calculate a second check statistic p value p2; If the first check statistic p value p1 and the second check statistic p value p2 are both greater than the second preset threshold, the correlation coefficient r between the first sample S1 and the sample S2 is calculated according to the following formula: q : In the formula, i represents a positive integer, x i represents the i-th sample value in the first sample S1, y i represents the i-th sample value in the second sample S2, represents the sample mean of the first sample S1, represents the sample mean of the second sample S2; The correlation coefficient r q As the linkage purchase influence coefficient between the certain type of e-commerce products and the other certain type of e-commerce products.

4. The e-commerce product accurate recommendation method according to claim 1, characterized in that: After determining whether the sum of the confidences of the various e-commerce products exceeds the first preset threshold, if it is determined that the sum of the confidences of the various e-commerce products does not exceed the first preset threshold, the method further includes the following steps S701 to S704: S701. Adjust the current e-commerce product recommendation plan to obtain a new e-commerce product recommendation plan, and then execute step S702; S702. For each type of e-commerce product, the browsing event indicator values ​​or browsing time proportion values ​​of all e-commerce products during the most recent m-1 consecutive historical startups and runs and the new e-commerce product recommendation scheme are imported into the corresponding purchase event prediction model, and a new confidence level of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup and run is obtained as an output, and then step S703 is executed; S703. Determine whether the sum of the new confidences of the various e-commerce products exceeds the first preset threshold. If so, execute step S704; otherwise, execute step S701; S704. During the current next startup operation, the new e-commerce product recommendation plan is executed through the e-commerce platform application.

5. The e-commerce product accurate recommendation method according to claim 1, characterized in that: When there are multiple e-commerce product recommendation solutions, the method further includes: For each of the various e-commerce products and each of the multiple e-commerce product recommendation schemes, the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the corresponding schemes are imported into the purchase event prediction model of the corresponding e-commerce product, and the event of the corresponding e-commerce product being purchased by the target customer during the current next startup operation and the corresponding confidence level are output; For each of the e-commerce product recommendation solutions, the corresponding confidences of each type of e-commerce product are cumulatively calculated to obtain a corresponding total confidence value; Determining whether the maximum confidence sum value exceeds the first preset threshold; If so, during the current next startup operation, an e-commerce product recommendation scheme among the multiple e-commerce product recommendation schemes and corresponding to the maximum confidence sum value is executed through the e-commerce platform application.

6. The e-commerce product accurate recommendation method according to claim 1, characterized in that: The method further comprises: Applying the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the most recent m-1 consecutive historical startup runs and the purchase event prediction model of each type of e-commerce product, the e-commerce product recommendation scheme is optimized based on an optimization algorithm to obtain the e-commerce product recommendation scheme and an optimized search result for maximizing the sum of confidences of each type of e-commerce product; During the current next startup operation, the e-commerce product recommendation solution based on the optimized search results is executed through the e-commerce platform application.

7. An e-commerce product accurate recommendation device, characterized in that: It includes a record data collection unit, a record data processing unit, a sample data extraction unit, a prediction model training unit, a prediction model application unit, a threshold comparison and judgment unit and a recommendation scheme execution unit which are sequentially connected in communication; The record data collection unit is used to collect the e-commerce product browsing record data and e-commerce product purchase record data generated by the target customer during each historical startup and operation of the e-commerce platform application; The record data processing unit is used to obtain browsing event indication values ​​or browsing time proportion values ​​of various types of e-commerce products during the various historical startups and operations according to the e-commerce product browsing record data, and to obtain purchase event indication values ​​of various types of e-commerce products during the various historical startups and operations according to the e-commerce product purchase record data, wherein the browsing event indication value is used to indicate whether an event of browsing the corresponding e-commerce product by the target customer occurs during the startup and operation of the e-commerce platform application, the browsing time proportion value refers to the ratio of the time duration of browsing the corresponding e-commerce product by the target customer to the total time duration of browsing all e-commerce products during the startup and operation of the e-commerce platform application, and the purchase event indication value is used to indicate whether an event of purchasing the corresponding e-commerce product by the target customer occurs during the startup and operation of the e-commerce platform application; The sample data extraction unit is used to extract corresponding sample data for each type of e-commerce product in the following manner: taking the browsing event indication value or browsing time proportion value of all e-commerce products during the m consecutive historical startup operation periods as a model input item, and taking the purchase event indication value of the corresponding e-commerce product during the last historical startup operation period during the m consecutive historical startup operation periods as a model output item, and then taking the model input item and the model output item as a sample data, wherein m represents a positive integer greater than or equal to 2; The prediction model training unit is used to apply all the corresponding sample data to calibrate and verify the machine learning model based on the artificial intelligence algorithm for each type of e-commerce product, so as to obtain a corresponding purchase event prediction model; The prediction model application unit is used to import the browsing event indication values ​​or browsing time proportion values ​​of all the e-commerce products during the current most recent m-1 consecutive historical startup operations and the e-commerce product recommendation scheme of the e-commerce platform application during the current next startup operation into the corresponding purchase event prediction model for each type of e-commerce product, and output the confidence of the event that the corresponding e-commerce product is purchased by the target customer during the current next startup operation, wherein the e-commerce product recommendation scheme includes the recommendation event indication values ​​or recommendation frequency proportion values ​​of each type of e-commerce product during the current next startup operation, the recommendation event indication value is used to indicate whether the event of the corresponding e-commerce product being recommended to the target customer for browsing is triggered during the current next startup operation, and the recommendation frequency proportion value refers to the ratio of the frequency of the corresponding e-commerce product being recommended to the target customer for browsing during the current next startup operation to the total frequency of all e-commerce products being recommended to the target customer for browsing; The threshold comparison and judgment unit is used to judge whether the sum of the confidences of the various types of e-commerce products exceeds a first preset threshold; The recommendation scheme execution unit is used to execute the e-commerce product recommendation scheme through the e-commerce platform application during the current next startup operation when it is determined that the total confidence of the various e-commerce products exceeds the first preset threshold.

8. A computer device, characterized in that: It includes a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the e-commerce product accurate recommendation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores instructions, and when the instructions are executed on the computer, the e-commerce product accurate recommendation method as described in any one of claims 1 to 6 is executed.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or the instruction is executed by a computer, the method for accurately recommending e-commerce products as described in any one of claims 1 to 6 is implemented.

Images