New retail intelligent recommendation and customer behavior analysis method and system
By conducting multi-dimensional analysis of customers and products of the new retail intelligent recommendation system, the problem of poor recommendation results in the existing solutions is solved, more efficient personalized recommendation and display optimization is achieved, and the reliability of new retail intelligent recommendation is improved. and adaptability.
Patent Information
- Application Number
- CN202411114685.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing new retail intelligent recommendation and customer behavior analysis solutions fail to conduct dynamic analysis from the customer and product dimensions, resulting in poor recommendation results and insufficient reliability.
By monitoring and analyzing the entire shopping process of customers, obtaining customer behavior processing sequences, combining payment accounts and product purchase data, digital processing and matching, calculating the purchase impact status and appropriate display of products, and providing personalized recommendations and optimization tips.
It improves the diversity and reliability of new retail intelligent recommendations, and enhances the adaptability and regulatory effect of different product display statuses.
Smart Images

Figure CN119067748A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of behavior analysis technology, and in particular to a new retail intelligent recommendation and customer behavior analysis method and system. Background Art
[0002] The new retail intelligent recommendation system is an important part of new retail technology. It uses technologies such as big data, machine learning, and artificial intelligence to conduct in-depth analysis of consumers' shopping behaviors, preferences, historical purchase records and other data, thereby providing consumers with personalized product recommendations.
[0003] The existing new retail intelligent recommendation and customer behavior analysis solutions have certain defects in implementation. Most of them only monitor and analyze the customer behavior data dimension, and do not combine the monitoring and analysis with the purchased goods dimension. Furthermore, they cannot integrate the analysis results of different dimensions to adaptively and dynamically process and analyze the display status of different goods, resulting in poor supervision and reliability of new retail intelligent recommendations. Summary of the invention
[0004] The purpose of the present invention is to provide a new retail intelligent recommendation and customer behavior analysis method and system to solve the technical problem that the existing solutions cannot adaptively perform dynamic processing and analysis on the appropriate display status of different commodities from the dimensions of purchasing customers and purchasing commodities, resulting in poor supervision effect and poor reliability of new retail intelligent recommendation.
[0005] The purpose of the present invention can be achieved by the following technical solutions: New retail intelligent recommendation and customer behavior analysis methods, including: Monitor and count the entire shopping process of different target customers, monitor and process customer payment details, and obtain customer behavior processing sequences; Digitally process and analyze the purchase influence states corresponding to different commodities purchased by customers according to the customer behavior processing sequence, and determine the purchase experience state corresponding to the customer according to all the purchase influence states corresponding to all commodities purchased by the customer; Among them, the payment account in the customer behavior processing sequence is obtained, and the payment account is traversed and matched with the commodity purchase database. If the payment account exists in the commodity purchase database, a monitoring and verification instruction is generated, and all commodity type identifiers in the customer behavior processing sequence corresponding to the payment account are traversed and matched with its historical purchase commodity data according to the monitoring and verification instruction; If the commodity type identifier does not exist in the historical commodity purchase data corresponding to the customer, a commodity validity instruction is generated and the commodity type identifier is marked as a valid commodity type identifier, and the purchase influence status corresponding to all valid commodity type identifiers is digitally processed and analyzed; On the contrary, if the commodity type identifier exists in the historical commodity purchase data corresponding to the customer, an invalid commodity instruction is generated and the commodity type identifier is marked as an invalid commodity type identifier; If the payment account does not exist in the product purchase database, a monitoring continuous instruction is generated and all product type identifiers in the customer behavior processing sequence are marked as valid product type identifiers, and the purchase influence status corresponding to all valid product type identifiers is digitally processed and analyzed to determine the purchase influence status corresponding to different products and the purchase experience status corresponding to the customer; Calculate the display suitability data of different commodities to obtain the display suitability corresponding to different commodities, perform data analysis on the display suitability corresponding to different commodities and provide optimization tips for the display of commodities with abnormal display status.
[0006] Preferably, when a customer brings up a product basket, a supervision instruction is generated, and a purchase identifier corresponding to the customer is obtained based on the product basket, and subsequent behavior of the customer is tracked based on the purchase identifier; Monitor and count the timestamps of items that customers put into their shopping carts and the locations of the items, and obtain the timestamps of item selection and the corresponding item type identifiers. When the customer to whom the purchase identification belongs finishes checking out, an end instruction is generated, and the details of the goods paid by the customer are monitored according to the end instruction to obtain purchase monitoring data.
[0007] Preferably, when the purchase monitoring data is digitally processed, the customer's payment account, all product type identifiers and their corresponding selection timestamps, the working timestamp of the product basket and the checkout timestamp in the purchase monitoring data are sorted and combined to obtain a customer behavior processing sequence.
[0008] Preferably, by the formula Calculate the purchase processing time difference TC corresponding to different products i ; In the formula, The selected timestamp corresponding to the product; is the last selected timestamp corresponding to the product; i=1, 2, 3, ..., n; n is a positive integer; i is a different product; and the purchase processing time difference corresponding to different products is analyzed through the purchase influence identification piecewise function to obtain the corresponding purchase influence identifier, and the purchase influence status corresponding to the corresponding product is determined according to the purchase influence identifier; The purchase influence flag contains a value of 0 or 1, indicating whether the purchase influence status corresponding to the corresponding product is normal or abnormal; The purchase influence identifiers corresponding to all valid commodity type identifiers purchased by the customer are sorted and combined to obtain a purchase influence identifier sequence.
[0009] Preferably, the expression of the purchase influence identification piecewise function is: ; In the formula, GY is the purchase influence indicator corresponding to the product, TC i ´ is the standard purchase processing time difference corresponding to the product.
[0010] Preferably, when determining the purchase experience status corresponding to the customer according to the purchase influence identification sequence, the purchase influence identification sequence is traversed and the total number of elements N with a value of 1 is counted, and the formula Calculate the purchase experience GT corresponding to the customer's purchase behavior; where exp() is an exponential function with the natural constant e as the base; N' is the total number of all elements in the purchase influence identification sequence; A is the standard purchase experience value; If the purchase experience is greater than 1, the corresponding customer will be marked as a target customer and a prompt indicating an abnormal purchase status will be generated; if the purchase experience is not greater than 1, the corresponding customer will be marked as a normal customer and a prompt indicating a normal purchase status will be generated.
[0011] Preferably, periodic data calculations are performed on different commodities within a preset monitoring period.
[0012] Preferably, the total number of target customers is counted, and the customer influence identifiers corresponding to all target customers are obtained by performing data analysis through a periodic state recognition piecewise function; Among them, the expression of the periodic state identification piecewise function is: ; In the formula, ZZ is the customer impact identifier corresponding to all target customers; Ny is all target customers; Nz is all normal customers; the customer impact identifier contains 0 or non-zero values; Periodic data calculations are performed on different merchants based on customer influence flags with non-zero values.
[0013] Preferably, all commodities that meet the scheme conditions are counted and marked as target commodities, and all purchase influence identifiers corresponding to different target commodities are obtained. When periodic data calculation is performed on different target commodities, the total number N1 of all purchase influence identifiers corresponding to the target commodities and the total number N2 of all purchase influence identifiers with a value of 1 are counted, and the total number N2 of all purchase influence identifiers with a value of 1 is calculated, and the total number N1 of all purchase influence identifiers corresponding to the target commodities is calculated, and the total number N2 of all purchase influence identifiers with a value of 1 is calculated, and the total number N2 ... corresponding to the target commodities is calculated, and the total number N2 of all purchase influence identifiers with a value of 1 is calculated, and the total number N2 of all purchase Calculate and obtain the display suitability CH corresponding to the target product; where B is the standard display suitability; If the display suitability is less than or equal to 1, the target product is marked as display suitable; If the display suitability is greater than 1, the target product will be marked as inappropriate for display.
[0014] New retail intelligent recommendation and customer behavior analysis system, including: The data monitoring and statistical processing module is used to monitor and count the entire shopping process of different customers within the target, monitor and process the customer's payment details, and obtain the customer behavior processing sequence; The monitoring data processing and analysis module is used to digitally process and analyze the purchase influence status corresponding to different commodities purchased by the customer according to the customer behavior processing sequence, and determine the purchase experience status corresponding to the customer according to all the purchase influence status corresponding to all commodities purchased by the customer; The commodity monitoring data evaluation management module is used to calculate the display suitability data of different commodities, obtain the display suitability corresponding to different commodities, perform data analysis on the display suitability corresponding to different commodities, and provide optimization prompts for the display of commodities with abnormal display status.
[0015] Compared with the existing solutions, the present invention achieves the following beneficial effects: The present invention monitors and counts the entire shopping process of different customers within the target, and processes the monitored and counted data of different dimensions to obtain the customer behavior processing sequence corresponding to the customers. While realizing the statistical processing of data of different dimensions, it can also provide reliable multi-dimensional monitoring data support for the subsequent purchase influence status analysis corresponding to different commodities and the purchase experience status analysis corresponding to customers.
[0016] The present invention digitally processes and combines the purchase influence states corresponding to different commodities purchased by customers to obtain a purchase influence identification sequence, and performs extended analysis based on the purchase influence identification sequence to digitally process and analyze the purchase experience states of the corresponding customers, thereby realizing data processing and analysis utilization in different dimensions of the customer purchase process, and improving the diversity and reliability of new retail intelligent recommendation supervision.
[0017] The present invention obtains the display status of different commodities by calculating and analyzing the display suitability data of different commodities, and optimizes the display of commodities with abnormal display status, thereby improving the adaptability and diversity of new retail intelligent recommendations. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] The present invention will be further described below in conjunction with the accompanying drawings.
[0019] Figure 1 This is a flowchart of the new retail intelligent recommendation and customer behavior analysis method of the present invention.
[0020] Figure 2 This is a flowchart of the commodity type identification classification in the present invention.
[0021] Figure 3 This is a module block diagram of the new retail intelligent recommendation and customer behavior analysis system of the present invention. DETAILED DESCRIPTION
[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0023] Example 1: Figure 1 As shown, the present invention is a new retail intelligent recommendation and customer behavior analysis method, comprising: Monitor and count the entire shopping process of different target customers, monitor and process customer payment details, and obtain customer behavior processing sequences; including: When a customer brings up a product basket, a supervision instruction is generated, and the customer's corresponding purchase ID is obtained based on the product basket. The subsequent behavior of the customer is tracked based on the purchase ID. It should be noted that the target may be a retail store, and a positioning sensor, a pressure sensor and a timer are provided in the product basket. The positioning sensor is used to locate the real-time position of the product basket, and the pressure sensor is used to monitor the change in the quantity of products in the product basket; the timer is used to time different shopping actions; Monitor and count the timestamps of the items that customers put into the product basket and the locations of the items. The location of the item can be determined by the corresponding basket position when the item is put into the product basket, and obtain the selected timestamp and the corresponding product type identifier of the item. The units of different types of timestamps are accurate to minutes. It should be explained that when obtaining the position corresponding to the product through the product basket, although there are several layers of shelves, the products at the same position on different shelf layers are of the same type. For example, different brands of beverages are located on different shelf layers. Therefore, the product type identifier corresponding to the product can be determined according to the location of the product. In addition, the commodity type identifier is used to digitally and differentially represent the type of the commodity. The commodity name can be traversed and matched with all sample commodities pre-stored in the pre-built commodity database to obtain the corresponding commodity type identifier. Different sample commodities are pre-associated with a corresponding commodity type identifier. The specific value of the commodity type identifier can be determined based on the corresponding price and sales volume. When the customer to whom the purchase identifier belongs finishes checking out, an end instruction is generated. The end instruction can be generated when there is no commodity in the commodity basket, and the details of the commodity paid by the customer are monitored according to the end instruction to obtain purchase monitoring data; When the purchase monitoring data is digitally processed, the customer's payment account, all commodity type identifiers and their corresponding selection timestamps, the working timestamp of the commodity basket and the checkout timestamp in the purchase monitoring data are sorted and combined to obtain a customer behavior processing sequence; the purchase monitoring data is digitally processed, and the numbers in the purchase monitoring data can be identified and extracted and combined through the existing OCR recognition technology; the customer's payment account, all commodity type identifiers and their corresponding selection timestamps, the working timestamp of the commodity basket and the checkout timestamp in the purchase monitoring data are all numbers; in addition, the customer's payment account can be obtained based on the cash register software, and the cash register software obtains the user's payment account information by integrating the payment gateway or the API provided by the payment service provider (PSP); when the user chooses to use mobile payment methods (such as WeChat Pay, Alipay, etc.) to pay, the cash register software will interact with these payment platforms to complete the transaction process; In an embodiment of the present invention, by monitoring and counting the entire shopping process of different customers within the target, and processing the monitored and counted data of different dimensions to obtain the customer behavior processing sequence corresponding to the customer, statistical processing of data of different dimensions is realized, and reliable multi-dimensional monitoring data support can also be provided for subsequent purchase influence status analysis corresponding to different commodities and purchase experience status analysis corresponding to customers.
[0024] Digitally process and analyze the purchase influence states corresponding to different commodities purchased by customers according to the customer behavior processing sequence, and determine the purchase experience state corresponding to the customer according to all the purchase influence states corresponding to all commodities purchased by the customer; including: Obtain the payment account in the customer behavior processing sequence, which can be a membership number, and traverse and match the payment account with the commodity purchase database. If the payment account exists in the commodity purchase database, generate a monitoring and verification instruction, and traverse and match all commodity type identifiers in the customer behavior processing sequence corresponding to the payment account with its historical purchase commodity data according to the monitoring and verification instruction; like Figure 2 As shown, if the commodity type identifier does not exist in the historical commodity purchase data corresponding to the customer, a commodity valid instruction is generated and the commodity type identifier is marked as a valid commodity type identifier, and the purchase influence status corresponding to all valid commodity type identifiers is digitally processed and analyzed; On the contrary, if the commodity type identifier exists in the historical commodity purchase data corresponding to the customer, an invalid commodity instruction is generated and the commodity type identifier is marked as an invalid commodity type identifier; It should be explained that if a customer has purchased a corresponding product in the past, the monitoring data corresponding to the product is not valuable for analysis, which enables the screening of the effectiveness of different products purchased by different customers; If the payment account does not exist in the product purchase database, or the payment account is empty, a monitoring continuous instruction is generated and all product type identifiers in the customer behavior processing sequence are marked as valid product type identifiers, and the purchase influence status corresponding to all valid product type identifiers is digitally processed and analyzed; By formula Calculate the purchase processing time difference TC corresponding to different products i ; In the formula, The selected timestamp corresponding to the product; is the last selected timestamp of the product; i=1, 2, 3, ..., n; n is a positive integer; i is a different product; when the product is the first product put in, is the timestamp corresponding to when the product basket is picked up; and the purchase processing time differences corresponding to different products are analyzed through the purchase influence identification piecewise function to obtain the corresponding purchase influence identifier, and the purchase influence status corresponding to the corresponding product is determined according to the purchase influence identifier; Among them, the expression of the piecewise function for purchasing influence identification is: ; In the formula, GY is the purchase influence indicator corresponding to the product, TC i ´ is the standard purchase processing time difference corresponding to the product, which can be determined based on the median of all historical purchase processing time differences of the corresponding product, or based on previous test data; It should be explained that, in the embodiment of the present invention, by processing the purchase processing time difference data of different commodities to obtain the corresponding purchase influence identifier, it is possible to digitally represent the purchase influence status of different commodities, and provide reliable local commodity purchase analysis data support for subsequent customer corresponding purchase experience status analysis, thereby improving the diversity of commodity dimension monitoring processing and expanded utilization; The purchase influence flag contains a value of 0 or 1, indicating whether the purchase influence status corresponding to the corresponding product is normal or abnormal; Sort and combine the purchase influence identifiers corresponding to all valid commodity type identifiers purchased by the customer to obtain a purchase influence identifier sequence; Among them, the purchase impact identification sequence plays a role in integration and subsequent data analysis support; When determining the purchase experience status of a customer according to the purchase influence identification sequence, the purchase influence identification sequence is traversed and the total number of elements N with a value of 1 is counted, and the formula is used to calculate the purchase experience status of the customer. Calculate and obtain the purchase experience GT corresponding to the customer's purchase behavior; where exp() is an exponential function with the natural constant e as the base; N' is the total number of all elements in the purchase influence identification sequence; A is the standard purchase experience value, which is determined based on the median value of the purchase experience of all historical customers, or can be determined based on the actual operational requirements of the target; Conduct data analysis on the purchase experience to determine the customer's corresponding purchase experience status; If the purchase experience is greater than 1, the corresponding customer will be marked as a target customer and a prompt indicating abnormal purchase status will be generated; If the purchase experience is not greater than 1, the corresponding customer will be marked as a normal customer and a prompt indicating that the purchase status is normal will be generated; In the embodiment of the present invention, the purchase influence states corresponding to different commodities purchased by customers are digitally processed and combined to obtain a purchase influence identification sequence, and an extended analysis is performed based on the purchase influence identification sequence to digitally process and analyze the purchase experience states of the corresponding customers, thereby realizing data processing and analysis utilization of different dimensions of the customer purchase process, and improving the diversity and reliability of new retail intelligent recommendation supervision.
[0025] Calculate the display suitability data of different commodities to obtain the display suitability of different commodities, analyze the display suitability of different commodities and provide optimization tips for commodities with abnormal display status; including: Solution 1: Periodically calculate data for different commodities within a preset monitoring period; the monitoring period is in days, which can be seven days; Solution 2: Count the total number of target customers and perform data analysis through the periodic state recognition piecewise function to obtain the customer influence identifiers corresponding to all target customers; Among them, the expression of the periodic state identification piecewise function is: ; In the formula, ZZ is the customer influence identifier corresponding to all target customers; Ny is all target customers; Nz is all normal customers; The customer impact flag contains a 0 or non-zero value; Perform periodic data calculations on different merchants based on customer impact flags with non-zero values; It should be noted that, unlike the existing technical solutions which only use a single monitoring solution to monitor and analyze the status of different commodity displays, the embodiment of the present invention discloses two different monitoring solutions, which can be selected and implemented according to the actual needs of the actual application scenario, or can be implemented simultaneously, further improving the diversity of supervision; Count all products that meet the solution conditions and mark them as target products. Obtain all purchase influence identifiers corresponding to different target products. When periodically calculating data for different target products, count the total number of all purchase influence identifiers corresponding to the target products, N1, and the total number of purchase influence identifiers with a value of 1, N2, and use the formula Calculate and obtain the display suitability CH corresponding to the target product; where B is the standard display suitability, which can be determined based on the actual operational requirements of the target; When performing data analysis on display suitability to determine the display suitability status corresponding to the target product, if the display suitability is less than or equal to 1, the target product is marked as being in the display suitability status; If the display suitability is greater than 1, the target product is marked as being in an inappropriate display state; it can be understood that the inappropriate display state means that the display position corresponding to the product does not meet the purchase experience requirements of the corresponding product.
[0026] In the embodiment of the present invention, the display status of different commodities is obtained by calculating and analyzing the display suitability data of different commodities, and optimization prompts are given for the display of commodities with abnormal display status, thereby improving the adaptability and diversity of new retail intelligent recommendations.
[0027] Example 2: Figure 3 As shown, the present invention is a new retail intelligent recommendation and customer behavior analysis system, including: The data monitoring and statistical processing module is used to monitor and count the entire shopping process of different customers within the target, monitor and process the customer's payment details, and obtain the customer behavior processing sequence; The monitoring data processing and analysis module is used to digitally process and analyze the purchase influence status corresponding to different commodities purchased by the customer according to the customer behavior processing sequence, and determine the purchase experience status corresponding to the customer according to all the purchase influence status corresponding to all commodities purchased by the customer; The commodity monitoring data evaluation management module is used to calculate the display suitability data of different commodities, obtain the display suitability corresponding to different commodities, perform data analysis on the display suitability corresponding to different commodities, and provide optimization prompts for the display of commodities with abnormal display status.
[0028] In addition, the formulas involved in the above are all dimensionless and numerical calculations. They are a formula that is closest to the actual situation obtained by collecting a large amount of data and simulating it with simulation software.
[0029] In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the above-described embodiments of the invention are only illustrative, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation.
[0030] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0031] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of hardware plus software functional modules.
[0032] It is obvious to a person skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the essential characteristics of the present invention.
[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.
Claims
1. New retail intelligent recommendation and customer behavior analysis method, characterized by: include: Monitor and count the entire shopping process of different target customers, monitor and process customer payment details, and obtain customer behavior processing sequences; Digitally process and analyze the purchase influence states corresponding to different commodities purchased by customers according to the customer behavior processing sequence, and determine the purchase experience state corresponding to the customer according to all the purchase influence states corresponding to all commodities purchased by the customer; Among them, the payment account in the customer behavior processing sequence is obtained, and the payment account is traversed and matched with the commodity purchase database. If the payment account exists in the commodity purchase database, a monitoring and verification instruction is generated, and all commodity type identifiers in the customer behavior processing sequence corresponding to the payment account are traversed and matched with its historical purchase commodity data according to the monitoring and verification instruction; If the commodity type identifier does not exist in the historical commodity purchase data corresponding to the customer, a commodity validity instruction is generated and the commodity type identifier is marked as a valid commodity type identifier, and the purchase influence status corresponding to all valid commodity type identifiers is digitally processed and analyzed; On the contrary, if the commodity type identifier exists in the historical commodity purchase data corresponding to the customer, an invalid commodity instruction is generated and the commodity type identifier is marked as an invalid commodity type identifier; If the payment account does not exist in the product purchase database, a monitoring continuous instruction is generated and all product type identifiers in the customer behavior processing sequence are marked as valid product type identifiers, and the purchase influence status corresponding to all valid product type identifiers is digitally processed and analyzed to determine the purchase influence status corresponding to different products and the purchase experience status corresponding to the customer; Calculate the display suitability data of different commodities to obtain the display suitability corresponding to different commodities, perform data analysis on the display suitability corresponding to different commodities and provide optimization tips for the display of commodities with abnormal display status.
2. The new retail intelligent recommendation and customer behavior analysis method according to claim 1 is characterized in that: When a customer brings up a product basket, a supervision instruction is generated, and the customer's corresponding purchase ID is obtained based on the product basket. The subsequent behavior of the customer is tracked based on the purchase ID. Monitor and count the timestamps of items that customers put into their shopping carts and the locations of the items, and obtain the timestamps of item selection and the corresponding item type identifiers. When the customer to whom the purchase identification belongs finishes checking out, an end instruction is generated, and the details of the goods paid by the customer are monitored according to the end instruction to obtain purchase monitoring data.
3. The new retail intelligent recommendation and customer behavior analysis method according to claim 2 is characterized in that: When the purchase monitoring data is digitally processed, the customer's payment account, all product type identifiers and their corresponding selection timestamps, the working timestamp of the product basket and the checkout timestamp in the purchase monitoring data are sorted and combined to obtain a customer behavior processing sequence.
4. The new retail intelligent recommendation and customer behavior analysis method according to claim 1 is characterized in that: By formula Calculate the purchase processing time difference TC corresponding to different products i ; In the formula, The selected timestamp corresponding to the product; is the last selected timestamp corresponding to the product; i=1, 2, 3, ..., n; n is a positive integer; i is a different product; and the purchase processing time difference corresponding to different products is analyzed through the purchase influence identification piecewise function to obtain the corresponding purchase influence identifier, and the purchase influence status corresponding to the corresponding product is determined according to the purchase influence identifier; The purchase influence flag contains a value of 0 or 1, indicating whether the purchase influence status corresponding to the corresponding product is normal or abnormal; The purchase influence identifiers corresponding to all valid commodity type identifiers purchased by the customer are sorted and combined to obtain a purchase influence identifier sequence.
5. The new retail intelligent recommendation and customer behavior analysis method according to claim 4 is characterized in that: The expression of the piecewise function for purchasing influence identification is: ; In the formula, GY is the purchase influence indicator corresponding to the product, TC i ´ is the standard purchase processing time difference corresponding to the product.
6. The new retail intelligent recommendation and customer behavior analysis method according to claim 4 is characterized in that: When determining the purchase experience status of a customer according to the purchase influence identification sequence, the purchase influence identification sequence is traversed and the total number of elements N with a value of 1 is counted, and the formula is used to calculate the purchase experience status of the customer. Calculate the purchase experience GT corresponding to the customer's purchase behavior; where exp() is an exponential function with the natural constant e as the base; N' is the total number of all elements in the purchase influence identification sequence; A is the standard purchase experience value; If the purchase experience is greater than 1, the corresponding customer will be marked as a target customer and a prompt indicating abnormal purchase status will be generated; If the purchase experience is not greater than 1, the corresponding customer will be marked as a normal customer and a prompt indicating that the purchase status is normal will be generated.
7. The new retail intelligent recommendation and customer behavior analysis method according to claim 6, characterized in that: Perform periodic data calculations on different commodities within the preset monitoring period.
8. The new retail intelligent recommendation and customer behavior analysis method according to claim 6, characterized in that: Count the total number of target customers, and perform data analysis through the periodic state recognition piecewise function to obtain the customer influence identifiers corresponding to all target customers; Among them, the expression of the periodic state identification piecewise function is: ; In the formula, ZZ is the customer impact identifier corresponding to all target customers; Ny is all target customers; Nz is all normal customers; the customer impact identifier contains 0 or non-zero values; Periodic data calculations are performed on different merchants based on customer influence flags with non-zero values.
9. The new retail intelligent recommendation and customer behavior analysis method according to claim 7 or 8, characterized in that: Count all products that meet the solution conditions and mark them as target products. Obtain all purchase influence identifiers corresponding to different target products. When periodically calculating data for different target products, count the total number of all purchase influence identifiers corresponding to the target products, N1, and the total number of purchase influence identifiers with a value of 1, N2, and use the formula Calculate and obtain the display suitability CH corresponding to the target product; where B is the standard display suitability; If the display suitability is less than or equal to 1, the target product is marked as display suitable; If the display suitability is greater than 1, the target product will be marked as inappropriate for display.
10. A new retail intelligent recommendation and customer behavior analysis system, applying the new retail intelligent recommendation and customer behavior analysis method according to any one of claims 1 to 9, characterized in that: include: The data monitoring and statistical processing module is used to monitor and count the entire shopping process of different customers within the target, monitor and process the customer's payment details, and obtain the customer behavior processing sequence; The monitoring data processing and analysis module is used to digitally process and analyze the purchase influence states corresponding to different commodities purchased by the customer according to the customer behavior processing sequence, and determine the purchase experience state corresponding to the customer according to all the purchase influence states corresponding to all the commodities purchased by the customer; The commodity monitoring data evaluation management module is used to calculate the display suitability data of different commodities, obtain the display suitability corresponding to different commodities, perform data analysis on the display suitability corresponding to different commodities, and provide optimization prompts for the display of commodities with abnormal display status.
Citation Information
Patent Citations
Intelligent terminal shopping guide system and method
CN111652669A
Commodity information recommendation method, related device, equipment and storage medium
CN112017015A
Intelligent monitoring management system and method for retail operation
CN115796908A
Commodity display position determination method and self-service system
CN116342169A
Intelligent checkout processing method and system, POS cash register and storage medium
CN116665380A
