Intelligent control management method and system for bag making machine

Abstract

The invention discloses an intelligent control management method and system for a bag making machine. The method comprises the following steps: obtaining the return rate of the bag making machine, client field data, bag making machine information and the goods price of the bag making machine; obtaining goods popularity data based on the customer field data, so as to obtain the goods preference degree of the bag machine according to the goods popularity data and the bag machine information; obtaining goods propaganda information based on the goods preference degree of the Fujiang bag machine, so as to obtain the push content of the Fujiang bag machine according to the goods propaganda information; on the basis of a return rate updating formula, updating the return rate of the bag-making machine to obtain an updated return rate, and obtaining goods configuration data of the bag-making machine according to the updated return rate, the goods preference degree of the bag-making machine and the goods price of the bag-making machine; and obtaining extraction probability configuration data according to the goods configuration data of the bag machine and the updated return rate. According to big data processing and neural network processing, the sales experience and feeding precision of the fossil bag machine are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control, and particularly relates to an intelligent control management method and system for a lucky bag machine. Background Art

[0002] The rise of the new retail model has promoted the rapid development of the vending machine industry. Among them, the lucky bag machine has become an innovative product that has attracted much attention in the market with its unique shopping experience and rich gift selections. With the continuous progress of technologies such as the Internet of Things, big data, and AI, the intelligent control and management level of the lucky bag machine has been significantly improved, laying a solid foundation for its market expansion.

[0003] Globally, the lucky bag machine market shows a booming development trend, and the lucky bag machine has become a popular shopping method among consumers. However, with the rapid growth of the market, the lucky bag machine market also faces multiple challenges. First, the pre-sale publicity of the lucky bag machine is insufficient; second, the after-sale experience of the lucky bag machine is poor; finally, the intelligent feeding accuracy of the lucky bag machine is insufficient, ultimately resulting in a relatively high user churn rate. Summary of the Invention

[0004] To solve the above problems existing in the prior art, the present invention provides an intelligent control management method and system for a lucky bag machine. The technical problems to be solved by the present invention are realized through the following technical solutions:

[0005] An intelligent control management method for a lucky bag machine, comprising:

[0006] Obtaining the yield rate of the lucky bag machine, customer on-site data, lucky bag machine information, and the prices of lucky bag machine goods;

[0007] Obtaining popularity data of goods based on the customer on-site data, and obtaining the preference degree of lucky bag machine goods according to the popularity data of goods and the lucky bag machine information;

[0008] Obtaining goods promotion information based on the preference degree of lucky bag machine goods, and obtaining the push content of the lucky bag machine according to the goods promotion information;

[0009] Updating the yield rate of the lucky bag machine based on a yield rate update formula to obtain an updated yield rate, and obtaining lucky bag machine goods configuration data according to the updated yield rate, the preference degree of lucky bag machine goods, and the prices of lucky bag machine goods;

[0010] Obtaining extraction probability configuration data according to the lucky bag machine goods configuration data and the updated yield rate; wherein, the yield rate of the lucky bag machine includes the preset yield rate of the lucky bag machine and the current yield rate of the lucky bag machine.

[0011] In a specific embodiment, obtaining goods promotion information based on the preference degree of lucky bag machine goods includes:

[0012] Input the preference degree of the goods in the lucky bag machine into a preset decision neural network model to output goods promotion information. Among them, the preset decision neural network model includes 3 parallel probability sub-processing units and 1 decision-making processing unit. The probability sub-processing unit includes 2 1*5 convolutional kernels, 2 5*3 convolutional kernels, 1 ReLU activation function, and 1 BN layer connected in sequence; the decision-making processing unit includes a fully connected layer, 1 MAX activation function, 1 5*5 convolutional kernel, 1 fully connected layer, 1 MIN activation function, 1 3*3 convolutional kernel, 1 fully connected layer, and 1 Sigmoid activation function connected in sequence.

[0013] In a specific embodiment, obtaining the push content of the lucky bag machine according to the goods promotion information includes:

[0014] Obtain the goods push content according to the goods promotion information,

[0015] Use big data analysis on the goods push content and the lucky bag machine information to obtain the push content of the lucky bag machine.

[0016] In a specific embodiment, update the yield rate of the lucky bag machine based on the yield rate update formula to obtain the updated yield rate, and obtain the goods configuration data of the lucky bag machine according to the updated yield rate, the preference degree of the goods in the lucky bag machine, and the price of the goods in the lucky bag machine, including:

[0017] Obtain the updated yield rate according to the preset yield rate of the lucky bag machine and the current yield rate of the lucky bag machine;

[0018] Use iterative calculation on the preset yield rate of the lucky bag machine, the preference degree of the goods in the lucky bag machine, and the price of the goods in the lucky bag machine to obtain the goods configuration data of the lucky bag machine;

[0019] In a specific embodiment, obtain the extraction probability configuration data according to the goods configuration data of the lucky bag machine and the updated yield rate, including:

[0020] Use a probability neural network on the goods configuration data of the lucky bag machine and the updated yield rate to obtain the extraction probability configuration data;

[0021] The probability neural network includes: a convolutional processing unit and a fully connected processing unit;

[0022] Among them, the convolutional processing unit includes 5 5*5 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 3 3*3 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 1 3*3 convolutional kernel, and 1 ReLU activation function connected in sequence; the fully connected processing unit includes a fully connected layer, 1 ReLU activation function, 1 fully connected layer, 1 ReLU activation function, 1 fully connected layer, and 1 softmax activation function connected in sequence.

[0023] The present invention discloses an intelligent control and management system for a lucky bag machine, including:

[0024] An acquisition unit, configured to obtain the yield rate of the lucky bag machine, customer on-site data, lucky bag machine information, and the price of goods in the lucky bag machine;

[0025] A calculation unit for the popularity of goods in the lucky bag machine, configured to obtain popularity data of goods based on the customer on-site data, and obtain the popularity of goods in the lucky bag machine according to the popularity data of goods and the lucky bag machine information;

[0026] A calculation unit for the push content of the lucky bag machine, configured to obtain product promotion information based on the popularity of goods in the lucky bag machine, and obtain the push content of the lucky bag machine according to the product promotion information;

[0027] A calculation unit for the configuration data of goods in the lucky bag machine, configured to update the yield rate of the lucky bag machine based on a yield rate update formula to obtain an updated yield rate, and obtain the configuration data of goods in the lucky bag machine according to the updated yield rate, the popularity of goods in the lucky bag machine, and the price of goods in the lucky bag machine;

[0028] A calculation unit for the extraction probability configuration data, configured to obtain the extraction probability configuration data according to the configuration data of goods in the lucky bag machine and the updated yield rate; wherein, the yield rate of the lucky bag machine includes a preset yield rate of the lucky bag machine and the current yield rate of the lucky bag machine.

[0029] In a specific embodiment, obtaining product promotion information based on the popularity of goods in the lucky bag machine includes:

[0030] Inputting the popularity of goods in the lucky bag machine into a preset decision neural network model to output product promotion information, wherein the preset decision neural network model includes 3 parallel probability sub-processing units and 1 decision-making processing unit, and the probability sub-processing unit includes 2 1*5 convolutional kernels, 2 5*3 convolutional kernels, 1 ReLU activation function, and 1 BN layer connected in sequence; the decision-making processing unit includes 1 fully connected layer, 1 MAX activation function, 1 5*5 convolutional kernel, 1 fully connected layer, 1 MIN activation function, 1 3*3 convolutional kernel, 1 fully connected layer, and 1 Sigmoid activation function connected in sequence.

[0031] In a specific embodiment, obtaining the push content of the lucky bag machine according to the product promotion information includes:

[0032] Obtaining the product push content according to the product promotion information,

[0033] Obtaining the push content of the lucky bag machine by using big data analysis according to the product push content and the lucky bag machine information.

[0034] In a specific embodiment, the yield rate of the lucky bag machine is updated based on the yield rate update formula to obtain an updated yield rate, and the lucky bag machine item configuration data is obtained according to the updated yield rate, the item preference degree of the lucky bag machine, and the item price of the lucky bag machine, including:

[0035] The updated yield rate is obtained according to the preset yield rate of the lucky bag machine and the current yield rate of the lucky bag machine;

[0036] The item configuration data of the lucky bag machine is obtained by iterative calculation for the preset yield rate of the lucky bag machine, the item preference degree of the lucky bag machine, and the item price of the lucky bag machine;

[0037] In a specific embodiment, the extraction probability configuration data is obtained according to the item configuration data of the lucky bag machine and the updated yield rate, including:

[0038] The extraction probability configuration data is obtained by using a probabilistic neural network for the item configuration data of the lucky bag machine and the updated yield rate;

[0039] The probabilistic neural network includes: a convolutional processing unit and a fully connected processing unit;

[0040] Among them, the convolutional processing unit includes 5 consecutive 5*5 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 3 consecutive 3*3 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 1 3*3 convolutional kernel, and 1 ReLU activation function; the fully connected processing unit includes 1 consecutive fully connected layer, 1 ReLU activation function, 1 fully connected layer, 1 ReLU activation function, 1 fully connected layer, and 1 softmax activation function.

[0041] Advantages of the present invention:

[0042] An intelligent control and management method and system for a lucky bag machine according to the present invention obtain accurate pre-sales and after-sales interaction content, reasonable item configuration data, and extraction probability configuration data based on big data processing and neural network processing, and finally improve the sales experience and feeding accuracy of the lucky bag machine.

[0043] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Description of the Drawings

[0044] Figure 1 is a flowchart of an intelligent control and management method for a lucky bag machine provided by an embodiment of the present invention;

[0045] Figure 2 is a schematic diagram of a decision neural network of an intelligent control and management method for a lucky bag machine provided by an embodiment of the present invention;

[0046] Figure 3It is a schematic diagram of a probabilistic neural network for an intelligent control and management method of a lucky bag machine provided by an embodiment of the present invention;

[0047] Figure 4 It is a block diagram of modules of an intelligent control and management system of a lucky bag machine provided by an embodiment of the present invention. Detailed implementation manners

[0048] The present invention will be further described in detail below in conjunction with specific embodiments, but the implementation manners of the present invention are not limited thereto.

[0049] Embodiment 1

[0050] In a specific implementation manner, please refer to Figure 1 , Figure 1 It is a flowchart of an intelligent control and management method of a lucky bag machine, and the specific steps are as follows:

[0051] S1: As a common marketing method, the operation effect of the lucky bag machine is directly related to the merchant's income and customer satisfaction. In order to optimize the operation strategy of the lucky bag machine, comprehensive data and information need to be collected. By collecting this data and information, a basis can be provided for subsequent decisions to ensure that the operation strategy of the lucky bag machine is more accurate and effective. Therefore, obtain the yield rate of the lucky bag machine, customer on-site data, lucky bag machine information, and the price of lucky bag machine goods. In a specific implementation manner,

[0052] The yield rate of the lucky bag machine includes the preset yield rate of the lucky bag machine and the current yield rate of the lucky bag machine;

[0053] The customer on-site data includes customer population classification, the stay time before and after customer purchase, the expressions of customers before and after purchase, the customer purchase time period, the types of lucky bags purchased by customers, the types of sold lucky bags, and the prices of sold lucky bags;

[0054] The lucky bag machine information includes the geographical location of the lucky bag machine, the type of the lucky bag machine, and the goods of the lucky bag machine;

[0055] The price of lucky bag machine goods includes the cost price and the selling price of each kind of goods of the lucky bag machine.

[0056] S2: The on-site customer data contains a large amount of customer behavior information, which can reflect customers' preferences for goods. At the same time, the claw machine information can also provide some valuable clues, such as the usage frequency of the device, the foot traffic at the location, etc. Through big data statistics and the processing of a judgment neural network, potential rules and trends can be mined from these data. By using the technical means of big data statistics and the judgment neural network, in-depth analysis is carried out on the on-site customer data and the claw machine information, so as to obtain a more accurate preference for claw machine goods and push content, providing a scientific basis for subsequent goods configuration and push strategies. Therefore, the on-site customer data and the claw machine information are processed by big data statistics and a judgment neural network to obtain the preference for claw machine goods and the push content of the claw machine; in a specific embodiment,

[0057] S21: To deeply understand consumers' preferences and optimize the operation strategy of the claw machine, in-depth big data statistics and analysis are carried out on the on-site customer data and the claw machine information. Based on this, the preference for claw machine goods is calculated, so as to formulate personalized marketing strategies and improve the user experience. Therefore, goods popularity data is obtained based on the on-site customer data, and the preference for claw machine goods is obtained according to the goods popularity data and the claw machine information.

[0058] S211: Big data statistics is used for the customer cycle data to obtain the average price of goods, the shipment volume of goods, the repurchase volume of goods, and the on-site happiness degree; in a specific embodiment,

[0059] To understand the market pricing trend and consumers' purchasing power, the average price of each good is obtained through big data statistics and calculation; since goods with a large shipment volume mean high market demand or successful marketing strategies, the shipment volume of each good is statistically obtained through big data; and since goods with a high repurchase volume indicate high customer satisfaction and loyalty, the number of times consumers repeat the purchase of each good is statistically analyzed through big data; finally, to evaluate consumers' emotional experience during the purchase process, the on-site happiness degree is obtained through multi-dimensional evaluation of customer feedback and on-site interaction data, etc.

[0060] S212: The goods popularity is obtained according to the average price of goods, the shipment volume of goods, the repurchase volume of goods, and the on-site happiness degree. The formula for calculating the goods popularity is

[0061] t = log((num1 * val + e num2 ) * h),

[0062] where t is the goods popularity, num1 and num2 are the shipment volume of goods and the repurchase volume of goods respectively, val is the average price of goods, and h is the on-site happiness degree.

[0063] S213: Obtain the preference degree of the lucky bag machine for goods through big data analysis based on the popularity of the goods and the information of the lucky bag machine. In a specific embodiment, the specific steps are as follows:

[0064] Sort the popularity of all inventory goods;

[0065] Calculate the preference degree of the lucky bag machine according to the popularity of the goods in the lucky bag machine and the geographical location weight of the lucky bag machine. The preference degree calculation formula is:

[0066] L = t * wt,

[0067] where t is the popularity of the goods, wt is the geographical location weight of the lucky bag machine, and L is the preference degree of the lucky bag machine for the goods.

[0068] S22: Obtain the push content of the lucky bag machine through a decision neural network and big data analysis based on the preference degree of the lucky bag machine for the goods and the information of the lucky bag machine.

[0069] S221: Obtain the goods promotion information through a decision neural network according to the preference degree of the lucky bag machine for the goods. The decision neural network includes: 3 parallel probability sub - processing units and 1 decision - making processing unit. In a specific embodiment, please refer to Figure 2 , Figure 2 which is a schematic diagram of the probability neural network of an intelligent control and management method for a lucky bag machine. The preset decision neural network model includes 3 parallel probability sub - processing units and 1 decision - making processing unit. The probability sub - processing unit includes 2 1*5 convolutional kernels, 2 5*3 convolutional kernels, 1 ReLU activation function, and 1 BN layer connected in sequence; the decision - making processing unit includes 1 fully - connected layer, 1 MAX activation function, 1 5*5 convolutional kernel, 1 fully - connected layer, 1 MIN activation function, 1 3*3 convolutional kernel, 1 fully - connected layer, and 1 Sigmoid activation function connected in sequence.

[0070] S222: Obtain the goods push content according to the goods promotion information. In a specific embodiment, the specific steps are as follows:

[0071] Plan the specific form and style of the push content according to the goods promotion information output by the decision neural network;

[0072] Obtain the goods push content according to the planned specific form and style of the push content. The goods push content includes text, pictures, and videos.

[0073] S223: Obtain the push content of the lucky bag machine through big data analysis based on the goods push content and the information of the lucky bag machine. In a specific embodiment, the specific steps are as follows:

[0074] Collect geographical location data and device information from the lucky bag machine operation system;

[0075] Clean and integrate the collected geographical location data and device information for big data analysis;

[0076] Use big data analysis tools to deeply mine and analyze the claw machine information to provide data support for optimizing the pushed content;

[0077] Optimize and adjust the pushed content of goods according to the results of big data analysis to obtain the pushed content of the claw machine.

[0078] S3: The yield rate of the claw machine is directly affected by the goods configuration and the extraction probability. The goods configuration determines the types and quantities of goods in the claw machine, while the extraction probability is related to the possibility for customers to obtain their desired goods. By comprehensively considering the yield rate, the goods preference degree, and the goods price, formulate an optimal goods configuration and extraction probability configuration strategy to improve the overall profitability and customer satisfaction of the claw machine. Therefore, update the yield rate of the claw machine based on the yield rate update formula to obtain the updated yield rate, and obtain the goods configuration data of the claw machine according to the updated yield rate, the goods preference degree of the claw machine, and the goods price of the claw machine;

[0079] S31: Obtain the updated yield rate according to the preset yield rate and the current yield rate of the claw machine. In a specific implementation manner, the specific steps are as follows:

[0080] Read the preset yield rate of the claw machine, which is usually set based on historical data or industry experience;

[0081] Obtain the current yield rate of the claw machine from the operation data, which reflects the recent operation effect;

[0082] Obtain the updated yield rate by using the update formula, and the update formula is:

[0083] rto_new = k * rto_set - rto_cur,

[0084] where rto_new is the updated yield rate, rto_set is the preset yield rate, rto_cur is the current yield rate, and k is a manually set parameter, which defaults to 1;

[0085] The preset yield rate is the expected yield rate preset according to the investment goal or strategy, and it can be determined based on various factors such as historical data, market conditions, and risk assessment. The preset yield rate is the benchmark for investment decisions, used to guide investment behaviors and evaluate investment effects;

[0086] The current yield rate is the yield rate actually achieved by the investment portfolio or asset currently, which reflects the actual performance of the investment and is a key indicator for measuring the investment effect. The current yield rate can be calculated by comparing the investment income and the principal;

[0087] The manually set parameter is an adjustment coefficient used to balance the relationship between the preset rate of return and the current rate of return. It can be adjusted according to the conservatism of the investment strategy or the sensitivity to market changes. When the value of k is large, the updated rate of return tends to be closer to the preset rate of return, reflecting the pursuit of the preset goal. When the value of k is small, the updated rate of return is closer to the current rate of return, reflecting the adaptation to the current market situation.

[0088] S32: For the preset rate of return of the lucky bag machine, the preference degree of the goods in the lucky bag machine, and the price of the goods in the lucky bag machine, iterative calculation is used to obtain the goods configuration data of the lucky bag machine. In a specific embodiment, the specific steps are as follows:

[0089] Initialize parameters: According to historical data or experience, set an initial configuration ratio or quantity for each type of goods, and set the parameters of the iterative algorithm, that is, the number of iterations and the convergence condition.

[0090] Construct a linear programming model: Use the maximization of the total return as the objective function, and consider the preference degree of the goods, the price of the goods, and the updated rate of return as the constraint conditions.

[0091] Execute iterative calculation: In each iteration, calculate the expected return according to the current goods configuration, adjust the goods configuration according to the expected return and the constraint conditions. If the convergence condition is met, stop the iteration and output the final goods configuration data.

[0092] S33: Use a probabilistic neural network for the goods configuration data of the lucky bag machine and the updated rate of return to obtain the extraction probability configuration data. The probabilistic neural network includes: a convolutional processing unit and a fully connected processing unit. In a specific embodiment, please refer to Figure 3 , Figure 3 is a schematic diagram of a probabilistic neural network for an intelligent control and management method of a lucky bag machine. The probabilistic neural network includes: a convolutional processing unit and a fully connected processing unit;

[0093] Among them, the convolutional processing unit includes 5 consecutive 5*5 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 3 consecutive 3*3 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 1 3*3 convolutional kernel, and 1 ReLU activation function; the fully connected processing unit includes 1 consecutive fully connected layer, 1 ReLU activation function, 1 fully connected layer, 1 ReLU activation function, 1 fully connected layer, and 1 softmax activation function.

[0094] S4: Pre-sales interaction is one of the important means to attract customers. By formulating appropriate pre-sales interaction content, the purchasing desire of customers can be stimulated, and the sales effect of the lucky bag machine can be improved. Therefore, the pre-sales interaction content is obtained according to the lucky bag machine product configuration data, the lucky bag machine information, and the lucky bag machine push content;

[0095] S41: Obtain the pre-sales push index according to the lucky bag machine product configuration data and the lucky bag machine information. In a specific embodiment, the specific steps are as follows:

[0096] Obtain the lucky bag machine product configuration data to identify popular products, new products, and promotional products, so as to provide a basis for formulating the push strategy;

[0097] Combined with the geographical location information of the lucky bag machine, determine the key products to be pushed from the popular products, new products, and promotional products;

[0098] Generate a pre-sales push index based on the key products.

[0099] S42: Obtain the pre-sales interaction content according to the pre-sales push index and the lucky bag machine push content. In a specific embodiment, the specific steps are as follows:

[0100] According to the pre-sales push index, screen the push content in the lucky bag machine push content. The push content includes product introduction videos, preferential information, and limited-time activity previews.

[0101] Integrate the screened push content to form pre-sales interaction content, and scroll and display it on the display screen of the lucky bag machine.

[0102] S5: After-sales interaction is an important link to consolidate customer relationships and enhance brand loyalty. By collecting customer on-site data and analyzing the feedback of the lucky bag machine push content, the real needs and opinions of customers can be understood, so as to formulate more effective after-sales interaction strategies. Therefore, the after-sales interaction content and after-sales feedback are obtained according to the customer on-site data and the lucky bag machine push content;

[0103] S51: Obtain the after-sales interaction content according to the after-sales push index and the lucky bag machine push content. In a specific embodiment, the specific steps are as follows:

[0104] Analyze the customer on-site data, including user expressions, purchase records, user feedback, and usage evaluations to identify user needs and potential problems;

[0105] Formulate an after-sales push index based on user needs and potential problems;

[0106] In the content pushed by the lucky bag machine, after-sales interaction content is obtained according to the after-sales push index and is scrolled and displayed on the display screen of the lucky bag machine. The after-sales interaction content includes rewards, sending personalized usage tutorials, and inviting participation in product improvement discussions.

[0107] S52: Obtain after-sales feedback based on the after-sales interaction content. In a specific embodiment, the specific steps are as follows:

[0108] Implement the after-sales interaction content and interact with users through the lucky bag machine or an online platform.

[0109] Collect user feedback, including direct replies on the lucky bag machine screen, questionnaires, social media comments, etc., to ensure comprehensive acquisition of user opinions.

[0110] Analyze the after-sales feedback and identify the key points of user satisfaction, including product quality, service attitude, and after-sales response speed.

[0111] According to the analysis results, formulate improvement measures, such as optimizing product design, improving service quality, strengthening user communication, etc., to provide guidance for future operations.

[0112] S6: According to the optimal product configuration and extraction probability configuration data formulated in the previous steps, perform actual product configuration and extraction probability configuration on the lucky bag machine to ensure that the operation strategy of the lucky bag machine can be effectively implemented and achieve the expected results. Therefore, perform product configuration and extraction probability configuration on the lucky bag machine using the lucky bag machine product configuration data and the extraction probability configuration data. In a specific embodiment, the specific steps are as follows:

[0113] The delivery personnel configure and shelve the products of the lucky bag machine at a fixed time with the pre-prepared lucky bag machine product configuration data, and at the same time update the product information of the lucky bag machine;

[0114] Replace the data of the lucky bag machine with the pre-prepared extraction probability configuration data on-site or remotely;

[0115] After the extraction probability configuration data is replaced, the management system of the lucky bag machine automatically performs data processing and verification to ensure the accuracy of the data.

[0116] After the product information of the lucky bag machine and the extraction probability configuration data are correctly configured, the lucky bag machine restarts and runs.

[0117] In a specific embodiment, please refer to Figure 4 , Figure 4 is a block diagram of the intelligent control management system of a lucky bag machine, including:

[0118] An acquisition unit for obtaining the yield rate of the lucky bag machine, on-site customer data, lucky bag machine information, and the price of lucky bag machine products;

[0119] A lucky bag machine product preference calculation unit, which is used to obtain product popularity data based on the customer on-site data, so as to obtain the lucky bag machine product preference according to the product popularity data and the lucky bag machine information;

[0120] A lucky bag machine push content calculation unit, which is used to obtain product promotion information based on the lucky bag machine product preference, so as to obtain the lucky bag machine push content according to the product promotion information;

[0121] A lucky bag machine product configuration data calculation unit, which is used to update the lucky bag machine yield rate based on the yield rate update formula to obtain an updated yield rate, so as to obtain the lucky bag machine product configuration data according to the updated yield rate, the lucky bag machine product preference, and the lucky bag machine product price;

[0122] An extraction probability configuration data calculation unit, which is used to obtain the extraction probability configuration data according to the lucky bag machine product configuration data and the updated yield rate; wherein, the lucky bag machine yield rate includes the lucky bag machine preset yield rate and the lucky bag machine current yield rate.

[0123] Preferably, obtaining product promotion information based on the lucky bag machine product preference includes:

[0124] Inputting the lucky bag machine product preference into a preset decision neural network model to output product promotion information, wherein the preset decision neural network model includes 3 parallel probability sub-processing units and 1 decision-making processing unit, and the probability sub-processing unit includes 2 1*5 convolutional kernels, 2 5*3 convolutional kernels, 1 ReLU activation function, and 1 BN layer connected in sequence; the decision-making processing unit includes 1 fully connected layer, 1 MAX activation function, 1 5*5 convolutional kernel, 1 fully connected layer, 1 MIN activation function, 1 3*3 convolutional kernel, 1 fully connected layer, and 1 Sigmoid activation function connected in sequence.

[0125] Preferably, obtaining the lucky bag machine push content according to the product promotion information includes:

[0126] Obtaining product push content according to the product promotion information,

[0127] Obtaining the lucky bag machine push content by using big data analysis according to the product push content and the lucky bag machine information.

[0128] Preferably, updating the lucky bag machine yield rate based on the yield rate update formula to obtain an updated yield rate, so as to obtain the lucky bag machine product configuration data according to the updated yield rate, the lucky bag machine product preference, and the lucky bag machine product price, includes:

[0129] Obtaining the updated yield rate according to the lucky bag machine preset yield rate and the lucky bag machine current yield rate;

[0130] The preset rate of return, the preference for the goods of the lucky bag machine, and the price of the goods of the lucky bag machine are used to obtain the goods configuration data of the lucky bag machine through iterative calculation;

[0131] Preferably, the extraction probability configuration data is obtained according to the goods configuration data of the lucky bag machine and the updated rate of return, including:

[0132] The probability neural network is used for the goods configuration data of the lucky bag machine and the updated rate of return to obtain the extraction probability configuration data;

[0133] The probability neural network includes: a convolutional processing unit and a fully connected processing unit;

[0134] Among them, the convolutional processing unit includes 5 consecutive 5*5 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 3 consecutive 3*3 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 1 3*3 convolutional kernel, and 1 ReLU activation function; the fully connected processing unit includes 1 consecutive fully connected layer, 1 ReLU activation function, 1 fully connected layer, 1 ReLU activation function, 1 fully connected layer, and 1 softmax activation function.

[0135] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality" means two or more, unless otherwise specifically defined.

[0136] Although the present application has been described in conjunction with various embodiments herein, however, in the process of implementing the claimed present application, those skilled in the art can understand and achieve other variations of the disclosed embodiments by viewing the accompanying drawings, the disclosure content, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "one" does not exclude a plurality of situations.

[0137] The above content is a further detailed description of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without departing from the concept of the present invention, several simple deductions or substitutions can be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. An intelligent control and management method for a lucky bag machine, characterized in that, Including: Obtaining the yield rate of the lucky bag machine, customer on-site data, lucky bag machine information, and the prices of the goods in the lucky bag machine; Obtaining popularity data of the goods based on the customer on-site data, and obtaining the preference degree of the goods in the lucky bag machine according to the popularity data of the goods and the lucky bag machine information; Obtaining goods promotion information based on the preference degree of the goods in the lucky bag machine, and obtaining the push content of the lucky bag machine according to the goods promotion information; Updating the yield rate of the lucky bag machine based on the yield rate update formula to obtain the updated yield rate, and obtaining the goods configuration data of the lucky bag machine according to the updated yield rate, the preference degree of the goods in the lucky bag machine, and the prices of the goods in the lucky bag machine; Obtaining the extraction probability configuration data according to the goods configuration data of the lucky bag machine and the updated yield rate; wherein, the yield rate of the lucky bag machine includes the preset yield rate of the lucky bag machine and the current yield rate of the lucky bag machine.

2. The intelligent control and management method of a lucky bag machine according to claim 1, wherein Obtaining goods promotion information based on the preference degree of the goods in the lucky bag machine, including: Inputting the preference degree of the goods in the lucky bag machine into a preset decision neural network model to output goods promotion information, wherein the preset decision neural network model includes 3 parallel probability sub-processing units and 1 decision-making processing unit, and the probability sub-processing unit includes 2 1*5 convolutional kernels connected in sequence, 2 5*3 convolutional kernels, 1 ReLU activation function, and 1 BN layer; the decision-making processing unit includes 1 fully connected layer, 1 MAX activation function, 1 5*5 convolutional kernel, 1 fully connected layer, 1 MIN activation function, 1 3*3 convolutional kernel, 1 fully connected layer, and 1 Sigmoid activation function connected in sequence.

3. The intelligent control and management method of a lucky bag machine according to claim 1, characterized in that Obtaining the push content of the lucky bag machine according to the goods promotion information, including: Obtaining the push content of the goods according to the goods promotion information, Obtaining the push content of the lucky bag machine by using big data analysis according to the push content of the goods and the lucky bag machine information.

4. The intelligent control and management method of a lucky bag machine according to claim 1, characterized in that, Updating the yield rate of the lucky bag machine based on the yield rate update formula to obtain the updated yield rate, and obtaining the goods configuration data of the lucky bag machine according to the updated yield rate, the preference degree of the goods in the lucky bag machine, and the prices of the goods in the lucky bag machine, including: Obtaining the updated yield rate according to the preset yield rate of the lucky bag machine and the current yield rate of the lucky bag machine; Performing iterative calculation on the preset yield rate of the lucky bag machine, the preference degree of the goods in the lucky bag machine, and the prices of the goods in the lucky bag machine to obtain the goods configuration data of the lucky bag machine.

5. The intelligent control and management method of a lucky bag machine according to claim 4, characterized in that, Obtaining the extraction probability configuration data according to the goods configuration data of the lucky bag machine and the updated yield rate, including: Obtaining the extraction probability configuration data by using a probability neural network for the goods configuration data of the lucky bag machine and the updated yield rate; The probability neural network includes: a convolutional processing unit and a fully connected processing unit; Wherein, the convolutional processing unit includes 5 5*5 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 3 3*3 convolutional kernels, 1 ReLU activation function, 1 average pooling layer, 1 3*3 convolutional kernel, and 1 ReLU activation function connected in sequence; the fully connected processing unit includes 1 fully connected layer, 1 ReLU activation function, 1 fully connected layer, 1 ReLU activation function, 1 fully connected layer, and 1 softmax activation function connected in sequence.

6. An intelligent control and management system for a lucky bag machine, characterized in that, Including: The acquisition unit is used to obtain the yield rate of the lucky bag machine, customer on-site data, lucky bag machine information, and the prices of the goods in the lucky bag machine; The lucky bag machine goods preference calculation unit is used to obtain goods popularity data based on the customer on-site data, so as to obtain the lucky bag machine goods preference according to the goods popularity data and the lucky bag machine information; The lucky bag machine push content calculation unit is used to obtain goods promotion information based on the lucky bag machine goods preference, so as to obtain the lucky bag machine push content according to the goods promotion information; The lucky bag machine goods configuration data calculation unit is used to update the yield rate of the lucky bag machine based on the yield rate update formula to obtain the updated yield rate, so as to obtain the lucky bag machine goods configuration data according to the updated yield rate, the lucky bag machine goods preference, and the prices of the goods in the lucky bag machine; The extraction probability configuration data calculation unit is used to obtain the extraction probability configuration data according to the lucky bag machine goods configuration data and the updated yield rate; wherein, the yield rate of the lucky bag machine includes the preset yield rate of the lucky bag machine and the current yield rate of the lucky bag machine.

7. An intelligent control and management system for a lucky bag machine according to claim 6, characterized in that, Obtaining goods promotion information based on the lucky bag machine goods preference includes: Inputting the lucky bag machine goods preference into a preset decision neural network model to output goods promotion information, wherein the preset decision neural network model includes 3 parallel probability sub-processing units and 1 decision-making processing unit, and the probability sub-processing unit includes 2 1*5 convolutional kernels, 2 5*3 convolutional kernels, 1 ReLU activation function, and 1 BN layer connected in sequence; the decision-making processing unit includes 1 fully connected layer, 1 MAX activation function, 1 5*5 convolutional kernel, 1 fully connected layer, 1 MIN activation function, 1 3*3 convolutional kernel, 1 fully connected layer, and 1 Sigmoid activation function connected in sequence.

8. The intelligent control and management system of a lucky bag machine according to claim 6, characterized in that, Obtaining the lucky bag machine push content according to the goods promotion information includes: Obtaining the goods push content according to the goods promotion information, Obtaining the lucky bag machine push content by using big data analysis according to the goods push content and the lucky bag machine information.

9. The intelligent control and management system of a lucky bag machine according to claim 6, characterized in that, Updating the yield rate of the lucky bag machine based on the yield rate update formula to obtain the updated yield rate, so as to obtain the lucky bag machine goods configuration data according to the updated yield rate, the lucky bag machine goods preference, and the prices of the goods in the lucky bag machine includes: Obtaining the updated yield rate according to the preset yield rate of the lucky bag machine and the current yield rate of the lucky bag machine; Performing iterative calculation on the preset yield rate of the lucky bag machine, the lucky bag machine goods preference, and the prices of the goods in the lucky bag machine to obtain the lucky bag machine goods configuration data.

10. The intelligent control and management system of a lucky bag machine according to claim 9, characterized in that, Obtaining the extraction probability configuration data according to the lucky bag machine goods configuration data and the updated yield rate includes: Obtaining the extraction probability configuration data by using a probability neural network for the lucky bag machine goods configuration data and the updated yield rate; The probability neural network includes: a convolutional processing unit and a fully connected processing unit; Among them, the convolution processing unit includes 5 5*5 convolution kernels connected in sequence, 1 ReLU activation function, 1 average pooling layer, 3 3*3 convolution kernels, 1 ReLU activation function, 1 average pooling layer, 1 3*3 convolution kernel, and 1 ReLU activation function; the fully connected processing unit includes 1 fully connected layer, 1 ReLU activation function, 1 fully connected layer, 1 ReLU activation function, 1 fully connected layer, and 1 softmax activation function connected in sequence.

Images