WO2021042541A1

WO2021042541A1 - Shopping guide method and apparatus in new retail model, device and storage medium

Info

Publication number: WO2021042541A1
Application number: PCT/CN2019/117714
Authority: WO
Inventors: 乐志能
Original assignee: 平安科技（深圳）有限公司
Priority date: 2019-09-06
Filing date: 2019-11-12
Publication date: 2021-03-11
Also published as: CN110751501A; CN110751501B

Abstract

A shopping guide method and apparatus in a new retail model, a device and a storage medium. Said method comprises: acquiring customer information; calculating a similarity between customer information and information about a customer having performed transaction, so as to obtain a plurality of similarities; determining a maximum similarity from the plurality of similarities, and when the maximum similarity is less than a preset threshold, inputting the customer information into a preset prediction model to obtain a product recommendation result; and acquiring a facial image, and sending the facial image and the product recommendation result to a sale terminal.

Description

Commodity shopping guide method, device, equipment and storage medium under new retail mode

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office, the application number is 201910844417.9, and the invention title is "Methods, devices, equipment, and storage media for shopping guides under the new retail model" on September 6, 2019. All of them The content is incorporated in this application by reference.

Technical field

The present invention relates to the field of artificial intelligence technology, and in particular to a method, device, equipment and storage medium for shopping guides in a new retail mode.

Background technique

The current sales method in a physical store is generally that the user enters the physical store, and then the sales consultant judges whether the user is a potential buying user based on experience, and if it is judged to be a potential buying customer, follow up the sales guide. This method relies on the empirical judgment of sales consultants, and requires a high level of ability of sales consultants. When there are many customers, the number of sales consultants is limited, which leads to the loss of potential consumer customers because they are not received by sales consultants, or the sales consultants are just new. At the beginning, if you don't understand any hobbies and characteristics of customers, you need to have too many conversations with customers, which arouses customers' disgust and leads to a lower sales success rate.

Summary of the invention

The main purpose of the present invention is to provide a method, device, equipment and storage medium for shopping guide under the new retail mode, aiming to solve the technical problem of low sales success rate in the prior art.

In order to achieve the above objectives, the present invention provides a shopping guide method for commodities under a new retail mode, including:

Acquiring the location information reported by the customer terminal, and determining whether the customer terminal is located in the target physical store according to the location information;

If the customer terminal is located in the target physical store, acquiring customer information reported by the customer terminal;

Calculate the similarity between the customer information and the transactional customer information stored in the historical transaction record to obtain multiple similarity calculation results;

Determine a maximum similarity calculation result from the multiple similarity calculation results, and determine whether the maximum similarity calculation result is less than a preset threshold;

If the maximum similarity calculation result is less than a preset threshold, input the customer information into a preset prediction model to obtain a product recommendation result for the customer information;

Acquire the face image reported by the customer terminal, and send the face image and the product recommendation result to the sales terminal corresponding to the target physical store.

In addition, in order to achieve the above object, the present invention also provides a commodity shopping guide device under the new retail mode, and the commodity shopping guide device under the new retail mode includes:

The location judgment module is used to obtain the location information reported by the customer terminal, and determine whether the customer terminal is located in the target physical store according to the location information;

A customer information acquisition module, configured to acquire customer information reported by the customer terminal if the customer terminal is located in a target physical store;

The calculation module is used to calculate the similarity between the customer information and the transactional customer information stored in the historical transaction records to obtain multiple similarity calculation results;

The detection module is configured to determine a maximum similarity calculation result from the multiple similarity calculation results, and determine whether the maximum similarity calculation result is less than a preset threshold;

A prediction module, configured to input the customer information into a preset prediction model if the maximum similarity calculation result is less than a preset threshold, to obtain a product recommendation result for the customer information;

The push module is configured to obtain the face image reported by the customer terminal, and send the face image and the product recommendation result to the sales terminal corresponding to the target physical store.

In addition, in order to achieve the above object, the present invention also provides a computer device, the computer device including a memory and a processor;

The memory is used to store a computer program;

The processor is configured to execute the computer program and, when executing the computer program, implement the above-mentioned method for shopping guide under the new retail mode.

In addition, in order to achieve the above-mentioned object, the present invention also provides a computer-readable storage medium on which is stored a commodity shopping guide program under the new retail mode, and the commodity shopping guide program under the new retail mode is processed When the device is executed, the steps of the commodity shopping guide method under the new retail mode as described above are realized.

In the present invention, when it is determined that the customer is located in the target physical store, the product that the customer is interested in is determined according to the customer's information, and the sales staff in the target physical store are notified, so that the sales staff can provide customers with more accurate shopping guide services and increase sales. Efficiency and success rate.

Description of the drawings

FIG. 1 is a schematic diagram of a computer device structure of a hardware operating environment involved in a solution of an embodiment of the present invention;

FIG. 2 is a schematic flowchart of an embodiment of a merchandise shopping guide method under the new retail mode of the present invention;

3 is a schematic diagram of a scene of an embodiment of a method for shopping guide in a new retail mode according to the present invention;

FIG. 4 is a schematic diagram of the detailed process of determining that the customer terminal is located in the target physical store in FIG. 2;

5 is a detailed flowchart of the steps of sending the face image and the product recommendation result to the sales terminal corresponding to the target physical store in FIG. 2;

Fig. 6 is a schematic diagram of functional modules of an embodiment of a merchandise shopping guide device in a new retail mode of the present invention.

The realization of the objectives, functional characteristics and advantages of the present invention will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

detailed description

It should be understood that the specific embodiments described here are only used to explain the present invention, but not to limit the present invention.

The flowchart shown in the drawings is only an example, and does not necessarily include all contents and operations/steps, nor does it have to be executed in the described order. For example, some operations/steps can also be decomposed, combined or partially combined, so the actual execution order may be changed according to actual conditions.

Hereinafter, some embodiments of the present application will be described in detail with reference to the accompanying drawings. In the case of no conflict, the following embodiments and features in the embodiments can be combined with each other.

As shown in Fig. 1, Fig. 1 is a schematic structural diagram of a computer device in a hardware operating environment involved in a solution of an embodiment of the present invention.

The computer device in the embodiment of the present invention may be a PC, or a terminal device with data processing capabilities, such as a smart phone or a server.

As shown in FIG. 1, the computer device may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is used to implement connection and communication between these components. The user interface 1003 may include a display screen (Display) and an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as a magnetic disk memory. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

Those skilled in the art can understand that the structure of the computer device shown in FIG. 1 does not constitute a limitation on the computer device, and may include more or less components than those shown in the figure, or a combination of certain components, or different component arrangements.

As shown in FIG. 1, a memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a network operation control application program.

In the computer device shown in Figure 1, the network interface 1004 is mainly used to connect to the back-end server and communicate with the back-end server; the user interface 1003 is mainly used to connect to the client (user side) and communicate with the client; and the processing The device 1001 can be used to call a computer program stored in the memory 1005 and perform the following operations:

In an embodiment, the processor 1001 may call a computer program stored in the memory 1005, and also perform the following operations:

Acquiring the location information reported by the customer terminal, calculating the distance between the location information and the location information of each physical store, and obtaining multiple distance calculation results;

Determine whether there is a target calculation result less than or equal to a preset distance among the multiple distance calculation results;

If there is a target calculation result less than or equal to the preset distance among the multiple distance calculation results, it is determined that the customer terminal is located in the target physical store corresponding to the target calculation result.

Obtain each customer's information that has been traded stored in historical transaction records;

The similarity between the customer information and the transaction customer information is calculated by the cosine similarity formula, and multiple similarity calculation results are obtained. The cosine similarity formula is as follows:

Among them, n(A) represents the number of types of information contained in the customer information, n(B) represents the number of types of information contained in the customer information that has been traded, and n(A∩B) represents that the customer information is the same as that of the customer information that has been traded The amount of information, K is the similarity calculation result.

Acquiring historical transaction records, the historical transaction records including the product name and the transaction customer information corresponding to the product name;

Calculate the feature value corresponding to the transaction customer information, and substitute the feature value into the prediction function formula to obtain multiple prediction functions;

Iteratively solving the multiple prediction functions to obtain a prediction model corresponding to the product name;

The prediction function formula is as follows:

among them,

θ _i is the weight value of the transaction customer information i, x _i is the characteristic value corresponding to the transaction customer information i, θ ^T =[θ ₁ ,θ ₂ ,...,θ _n ], x=[x ₁ ,x ₂ ,...,x _n ], e is a natural constant.

If the calculation result of the maximum acquaintance degree is less than the preset threshold, calculate the characteristic value corresponding to the customer information, and input the characteristic value into the prediction model corresponding to each product to obtain several output values;

A target output value greater than or equal to a preset probability value is selected, and a product recommendation result for the customer information is obtained based on the target output value.

Obtaining a product list of the target physical store;

Detecting whether the product recommendation result exists in the product list;

If the product recommendation result exists in the product list, execute the acquisition of the face image reported by the customer terminal, and send the face image and the product recommendation result to the target physical store. The steps of the point-of-sale.

Acquiring status information of each sales terminal corresponding to the target physical store;

Determining the target sales terminal whose status is idle based on the status information;

Sending the face image and the product recommendation result to the target sales terminal.

Referring to FIG. 2, FIG. 2 is a schematic flowchart of an embodiment of a merchandise shopping guide method under the new retail mode of the present invention.

As shown in Figure 2, the shopping guide methods under the new retail model include:

Step S10: Obtain the location information reported by the customer terminal, and determine whether the customer terminal is located in the target physical store according to the location information.

In this embodiment, the message pushing method is applied to a computer device, and the computer device may specifically be a server. The location information of multiple physical stores is stored on the server. For example, the location information of physical store 1 to physical store 5 are W1, W2, W3, W4, and W5, respectively. The distance between the location information W reported by the customer terminal and W1 to W5 is calculated respectively, and five calculation results are obtained, which are respectively L1 to L5. Then it is further judged whether there is a result less than or equal to the preset distance in L1 to L5, where the preset distance is set according to the actual situation, for example, set to 50 meters. Due to the specific distance between the physical store and the physical store, generally speaking, only one result from L1 to L5 is less than or equal to 50 meters. For example, if L3 is less than or equal to 50 meters, it is determined that the customer terminal is currently in the physical store 3.

Step S20: If the customer terminal is located in the target physical store, acquire customer information reported by the customer terminal.

In this embodiment, the customer needs to pre-register when using the client installed in the customer terminal. During the registration process, some personal information needs to be filled in, such as gender, age, occupation, income status, health status, family medical history, etc. (The type of personal information can be expanded or reduced according to actual needs, and there is no restriction here), and then the customer terminal reports the personal information to the server, and the server stores the personal information in association with the terminal identification of the customer terminal. When it is detected that the customer terminal is located in the target physical store, the personal information (that is, customer information) corresponding to the terminal identification of the customer terminal can be obtained from the stored data.

Step S30: Calculate the similarity between the customer information and each transaction customer information stored in the historical transaction record to obtain multiple similarity calculation results.

In an optional embodiment, step S30 includes:

Obtain each transaction customer information stored in historical transaction records; calculate the similarity between the customer information and the transaction customer information through a cosine similarity formula, and obtain multiple similarity calculation results, the cosine similarity formula is as follows :

In this embodiment, every time a sales transaction is completed, the transaction details are recorded in the historical transaction record. Transaction details include: product name, customer information. As shown in Table 1, Table 1 is a data table that stores historical transaction records.

Table 1

已交易顾客信息1Transaction customer information 1	产品名称1Product name 1
已交易顾客信息2Transaction customer information 2	产品名称2Product name 2
已交易顾客信息3Transaction customer information 3	产品名称3Product name 3
已交易顾客信息4Transaction customer information 4	产品名称4Product name 4

Calculate the similarity between customer information and transaction customer information 1 to transaction customer information 4 through the cosine similarity formula. The cosine similarity formula is as follows:

Among them, n(A) represents the type and quantity of information contained in the customer information. If the customer information includes gender, age, occupation, income status, health status, and family medical history, then n(A) is 6; n(B) represents The type and quantity of the information contained in the transaction customer information. If the transaction customer information includes gender, age, occupation, income status, health status, and family medical history, then n(B) is 6; n(A∩B) represents the customer information and The quantity of the same information in the transaction customer information; K represents the calculation result of the similarity between the customer information and the transaction customer information. In one embodiment, if the age, occupation, and income status of the customer information and the transaction customer information 1 are the same, then the similarity calculation result of the customer information and the transaction customer information 1 is K1=0.5. In the same way, the similarity between the current customer information and the stored transactional customer information can be calculated, and multiple similarity calculation results can be obtained.

Step S40: Determine a maximum similarity calculation result from the multiple similarity calculation results, and determine whether the maximum similarity calculation result is less than a preset threshold.

In this embodiment, if the transaction customer information includes transaction customer information 1 to transaction customer information 4, four similarity calculation results are calculated, namely K1 to K4, and the largest value is selected from K1 to K4 For example, if the maximum calculation result is K2, it is determined whether K2 is greater than or equal to the preset threshold. The preset threshold is set according to actual needs, for example, set to 0.8.

In this embodiment, if K2 is greater than or equal to the preset threshold, it means that the personal information of the customer is very similar to that of customer 2 who has already traded. The goods purchased by customer 2 who have already traded are likely to be goods of interest to the customer. Therefore, It is possible to directly use the product purchased by the customer 2 who has already traded as the product to be recommended. For example, if the products purchased by customer 2 include: product 1 and product 3, the product recommendation result is product 1 and product 3.

Step S50: If the maximum similarity calculation result is less than a preset threshold, input the customer information into a preset prediction model to obtain a product recommendation result for the customer information.

In this embodiment, if the maximum similarity calculation result is less than the preset threshold, it means that the current customer information is not the same as the stored transactional customer information, and it is impossible to predict the products of interest to the current customer information based on historical transaction records. The customer information is input into the predictive model, and the result of product recommendation based on the customer information is obtained.

In this embodiment, there may be multiple prediction models, such as an LR model with prediction information of product 1, an LR model with prediction information of product 2, and an LR model with prediction information of product 3. Then, the characteristic values corresponding to the customer information are input into the three prediction models, and three probability values P1, P2, and P3 are obtained. Select a target output value greater than a preset probability value (for example, 0.85) from P1, P2, and P3. If P2 and P3 are greater than or equal to the preset probability value, then P2 and P3 are the target output values, indicating that product 2 and product 3 fit the needs of the customer to whom the current customer information belongs. Therefore, the determined product recommendation result is product 2 and Product 3.

Step S60: Obtain the face image reported by the customer terminal, and send the face image and the product recommendation result to the sales terminal corresponding to the target physical store.

In this embodiment, after the product recommendation result is obtained, the face image reported by the customer terminal is further obtained, and the face image and the product recommendation result are generated to the sales terminal corresponding to the target physical store, so that the sales staff of the sales terminal can follow the person The face image finds the customer, and conducts a shopping guide for the customer according to the product recommendation result. In this embodiment, when a customer registers to use a customer terminal, he will be required to upload a face image. After the customer terminal receives the face image, it will report the face image to the server, and the server will associate the face image with the terminal identifier of the customer terminal. Associated storage.

Referring to FIG. 3, FIG. 3 is a schematic diagram of a scene of an embodiment of a method for shopping guide of goods in a new retail mode of the present invention. As shown in Figure 3, the server establishes a communication connection with the customer terminal and the sales terminal respectively. The server determines the target physical store where it is located according to the location information reported by the customer terminal, and determines the product recommendation result according to the customer information corresponding to the customer terminal, so that the product The recommendation result and the facial image reported by the customer terminal are pushed to the sales terminal corresponding to the target physical store.

In this embodiment, the location information reported by the customer terminal is acquired, and based on the location information, it is determined whether the customer terminal is located in the target physical store; if the customer terminal is located in the target physical store, the customer information reported by the customer terminal is acquired; Calculate the similarity between the customer information and the transactional customer information stored in the historical transaction records to obtain multiple similarity calculation results; determine the maximum similarity calculation result from the multiple similarity calculation results, and determine the Whether the maximum similarity calculation result is less than the preset threshold; if the maximum similarity calculation result is less than the preset threshold, the customer information is input into the preset prediction model to obtain the product recommendation result for the customer information; The face image reported by the customer terminal is sent, and the face image and the product recommendation result are sent to the sales terminal corresponding to the target physical store. Through this embodiment, the product that the customer is interested in is determined according to the customer's information, and the salesperson is notified, so that the salesperson can provide the customer with more accurate shopping guide services, and improve the sales efficiency and success rate.

Referring to Fig. 4, Fig. 4 is a schematic diagram of the detailed process of determining that the customer terminal is located in the target physical store in Fig. 2. In an embodiment of the merchandise shopping guide method under the new retail mode of the present invention, step S10 includes:

Step S101: Obtain the location information reported by the customer terminal, calculate the distance between the location information and the location information of each physical store, and obtain multiple distance calculation results.

The earth is a nearly standard ellipsoid with an equatorial radius of 6378.140 kilometers, a polar radius of 6356.755 kilometers, and an average radius of 6371.04 kilometers. If we assume that the earth is a perfect sphere, then its radius is the average radius of the earth, denoted as R. If the 0 degree longitude line is used as the reference, then the surface distance between these two points can be calculated based on the latitude and longitude of any two points on the surface of the earth (here, the error caused by the topography of the earth surface to the calculation is ignored, and it is only a theoretical estimate). Suppose the longitude and latitude of the first point A is (LonA, LatA), and the longitude and latitude of the second point B is (LonB, LatB). According to the 0 degree longitude, the east longitude is taken as the positive value (Longitude), and the west longitude is taken as the negative value (-Longitude), the north latitude takes 90-latitude value (90-Latitude), the south latitude takes 90+latitude value (90+Latitude), then the two points after the above processing are counted as (MLonA,MLatA) and (MLonB, MLatB). Then according to the trigonometric derivation, the following formula for calculating the distance between two points can be obtained:

Formula 1:

C=sin(MLatA)*sin(MLatB)*cos(MLonA-MLonB)+cos(MLatA)*cos(MLatB) Formula 2:

Distance=R*Arccos(C)*Pi/180

In this embodiment, the location information (the location information reported by the customer terminal and the location information of each physical store pre-stored in the server) may be latitude and longitude. According to the above formula 1 and formula 2, the distance between the customer terminal and each physical store can be obtained.

Step S102. It is determined whether there is a target calculation result less than or equal to a preset distance among the multiple distance calculation results.

In this embodiment, multiple distance calculation results can be obtained according to step S101, and then it is further determined whether there is a target calculation result less than or equal to a preset distance among the multiple distance calculation results. Among them, the preset distance is set according to actual needs, for example, set to 50 meters.

Step S103: If there is a target calculation result less than or equal to a preset distance among the multiple distance calculation results, it is determined that the customer terminal is located in the target physical store corresponding to the target calculation result.

In this embodiment, if the calculated result of the distance between the customer terminal and the physical store 7 is less than or equal to the preset distance, it is determined that the customer terminal is located in the physical store 7.

In an embodiment, in an embodiment of the merchandise shopping guide method in the new retail mode of the present invention, before step S10, the method further includes:

Obtain historical transaction records, the historical transaction records including the product name and the transaction customer information corresponding to the product name; calculate the characteristic value corresponding to the transaction customer information, and substitute the characteristic value into the prediction function formula to obtain A prediction function; the multiple prediction functions are iteratively solved to obtain the prediction model corresponding to the product name; the prediction function formula is as follows:

among them,

In this embodiment, if customer 1, customer 2, customer 3... customer 10 has purchased product 1, then the transactional customer information corresponding to product 1 in the historical transaction record includes customer information from customer 1 to customer 10; if customer 11 , Customer 12, Customer 13...Customer 20 has purchased product 2, then the transaction customer information corresponding to product 2 in the historical transaction record includes customer information from customer 11 to customer 20; if customer 21, customer 22, customer 23... The customer 30 once purchased the product 3, and the transaction customer information corresponding to the product 3 in the historical transaction record includes the customer information of the customer 21 to the customer 30. Among them, customer information includes: gender, age, income, personal health status (the type of information can be expanded or reduced according to actual needs). Calculate the characteristic value corresponding to the customer information according to the preset characteristic value conversion rule. For example, if the gender is male, take 1; if the female is 0, take 1 if the age is greater than or equal to 50, take 0 if the age is greater than or equal to 50, and take 1 if the income is greater than or equal to 200,000. , If it is less than 200,000, take 0, take 1 if there is a disease, take 0 if there is no disease. Based on the conversion rule, the characteristic value corresponding to the transaction customer information can be obtained. And substitute the eigenvalues into the formula:

among them,

In this embodiment, the essence of the prediction model is a specific algorithm. Specific algorithms can include: logistic regression algorithms or neural network algorithms. The following describes the logistic regression algorithm. Correspondingly, the prediction model is the LR (Logistic Regression) model. Logistic regression is a classification method, which is mainly used to solve two classification problems (that is, there are only two outputs, representing two classifications respectively). The logistic regression algorithm uses the Logistic function (or called the Sigmoid function), and the curve form of the function is S Type curve, the function form is:

For the case of a linear boundary, the boundary form is as follows:

Use formula (1) and formula (2) to construct the prediction function:

In this embodiment, the prediction model includes: the prediction information is the LR model of product 1, the prediction information is the LR model of product 2, and the prediction information is the LR model of product 3.

The way to obtain the prediction information of the LR model of product 1 is: substitute the feature value corresponding to the customer information of customer 1 as sample data 1 into the above formula (3) to obtain the prediction function 1; similarly, the feature corresponding to the customer information of customer 2 The value is substituted into the above formula (3) as the sample data 2 to obtain the prediction function 2, and the characteristic value corresponding to the customer information of the customer 3 to the customer 10 is substituted into the above formula (3) as the sample data 3 to the sample data 10, respectively, to obtain the prediction function 3 To 10, and prediction function 1 to prediction function 10 are all obtained from the information of customers who have purchased product 1, so the output values of prediction function 1 to prediction function 10 are all equal, thereby constructing multiple solution functions, namely prediction function 1. = Prediction function 2, Prediction function 1 = Prediction function 3, Prediction function 1 = Prediction function 4..., and then iteratively solve the aforementioned solution function, and calculate θ ^T =[θ ₁ ,θ ₂ ,...,θ _n ], ^{substituting the calculated θ T} into the above prediction function, you can get the LR model whose prediction information is product 1. The output value of the LR model is 0~1. The larger the output value, the customer willing The greater the probability of buying product 1.

In the same way, the way to obtain the LR model whose predicted information is product 2 is: substitute the feature value corresponding to customer information of customer 11 as sample data 11 into the above formula (3), and use the feature value corresponding to customer information of customer 12 as sample data 12 is substituted into the above formula (3), and the characteristic values corresponding to the customer information of customers 13 to 20 are substituted into the above formula (3) as sample data 13 to sample data 20 respectively to obtain 10 prediction functions, and then according to the 10 prediction functions Construct multiple solution functions, and then iteratively solve the aforementioned multiple solution functions, and calculate θ ^T =[θ ₁ ,θ ₂ ,...,θ _n ], and ^{substitute the calculated θ T} into the above prediction function. The LR model whose predicted information is product 2 is obtained. The output value of the LR model is 0 to 1. The larger the output value, the greater the probability that the customer corresponding to the input customer information is willing to purchase product 2.

According to the above training process, the LR model whose prediction information is product 3, the LR model whose prediction information is product 4, and the LR model whose prediction information is product n can be obtained.

It should be noted that the process of quantifying (converting) the transaction customer information (gender, age, income, personal health status) into a numerical value can be based on actual needs and using reasonable rules to convert the transaction customer information (gender, age, Income, personal health status) converted to appropriate values. In addition, after the LR model is obtained, when the LR model is used, the characteristic value corresponding to the current customer information is input to the LR model. The output of the LR model is a probability value. The probability value ranges from 0 to 1, and the probability value is higher. Large, indicating that the product corresponding to the LR model is more suitable for the needs of the customer to which the customer information belongs.

Further, on the basis of this embodiment, step S50 includes:

If the calculation result of the maximum acquaintance degree is less than the preset threshold, the characteristic value corresponding to the customer information is calculated, and the characteristic value is input into the prediction model corresponding to each product to obtain a number of output values; select greater than or equal to the preset probability The value of the target output value is based on the target output value to obtain the product recommendation result for the customer information.

In this embodiment, as described above, multiple training models are obtained by training, for example, the LR model whose prediction information is product 1, the LR model whose prediction information is product 2, and the LR model whose prediction information is product 3. Then, the characteristic values corresponding to the customer information are input into the three prediction models, and three probability values P1, P2, and P3 are obtained. Select a target output value greater than a preset probability value (for example, 0.85) from P1, P2, and P3. If P2 and P3 are greater than or equal to the preset probability value, then P2 and P3 are the target output values, indicating that product 2 and product 3 fit the needs of the customer to whom the current customer information belongs. Therefore, the determined product recommendation result is product 2 and Product 3.

Further, in an embodiment of the merchandise shopping guide method in the new retail mode of the present invention, before step S60, the method further includes:

Obtain the product list of the target physical store; detect whether the product recommendation result exists in the product list; if the product recommendation result exists in the product list, execute the acquisition of the report from the customer terminal The step of sending the face image and the product recommendation result to the sales terminal corresponding to the target physical store.

In this embodiment, after the product recommendation results are obtained based on step S50, for example, the product recommendation results are product 2 and product 3. It is necessary to determine whether there is product 2 and product 3 in the target physical store, that is, whether the product recommendation result exists there. In the product list of the target physical store, only when the product recommendation result exists in the product list, step S60 is executed, which avoids waste of sales resources.

In this embodiment, if the product recommendation results (such as product 2 and product 3) do not exist in the product list, the out-of-stock reminder can be sent to the customer terminal to remind the customer that the product 2 and product 3 are in the destination physical store. Out of stock status.

Referring to FIG. 5, FIG. 5 is a detailed flowchart of the steps of sending the face image and the product recommendation result to the sales terminal corresponding to the target physical store in FIG. 2.

In this embodiment, the step of sending the face image and the product recommendation result to the sales terminal corresponding to the target physical store includes:

Step S601: Obtain status information of each sales terminal corresponding to the target physical store.

In this embodiment, each salesperson is equipped with a sales terminal. When the salesperson is working, the status of the sales terminal is adjusted to busy. When the salesperson is idle, the status of the sales terminal is adjusted to idle, and the status of the sales terminal is reported To the server.

Step S602: Determine a target sales terminal whose status is idle based on the status information.

In this embodiment, according to the status information reported by each sales terminal, a target sales terminal whose status is idle can be determined from all sales terminals.

Step S603: Send the face image and the product recommendation result to the target sales terminal.

In this embodiment, when sending the face image and the product recommendation result, the target sales terminal in the idle state is first determined, and the face image and the product recommendation result are sent to the target sales terminal in the idle state, so that the target sales terminal belongs to Sales staff can immediately serve customers, realizing reasonable management of sales resources.

Referring to FIG. 6, FIG. 6 is a schematic diagram of functional modules of an embodiment of a merchandise shopping guide device in a new retail mode of the present invention.

In this embodiment, the commodity shopping guide device in the new retail mode includes:

The location judgment module 10 is configured to obtain location information reported by a customer terminal, and determine whether the customer terminal is located in a target physical store according to the location information;

The customer information obtaining module 20 is configured to obtain the customer information reported by the customer terminal if the customer terminal is located in the target physical store;

The calculation module 30 is used to calculate the similarity between the customer information and the transactional customer information stored in the historical transaction records to obtain multiple similarity calculation results;

The detection module 40 is configured to determine a maximum similarity calculation result from the multiple similarity calculation results, and determine whether the maximum similarity calculation result is less than a preset threshold;

The prediction module 50 is configured to, if the maximum similarity calculation result is less than a preset threshold, input the customer information into a preset prediction model to obtain a product recommendation result for the customer information;

The push module 60 is configured to obtain the face image reported by the customer terminal, and send the face image and the product recommendation result to the sales terminal corresponding to the target physical store.

The specific embodiments of the merchandise shopping guide device in the new retail mode of the present invention are basically the same as the various embodiments of the merchandise shopping guide method in the new retail mode described above, and will not be repeated here.

In addition, the embodiment of the present invention also provides a computer-readable storage medium, the computer-readable storage medium stores a commodity shopping guide program in the new retail mode, and when the commodity shopping guide program in the new retail mode is executed by a processor The steps of each embodiment of the commodity shopping guide method in the above new retail mode are realized.

The specific embodiments of the computer-readable storage medium of the present invention are basically the same as the various embodiments of the commodity shopping guide method under the new retail mode described above, and will not be repeated here.

It should be noted that in this article, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, It also includes other elements that are not explicitly listed, or elements inherent to the process, method, article, or system. Without more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, article, or system that includes the element.

The sequence numbers of the foregoing embodiments of the present invention are only for description, and do not represent the superiority or inferiority of the embodiments.

Through the description of the above implementation manners, those skilled in the art can clearly understand that the above-mentioned embodiment method can be implemented by means of software plus the necessary general hardware platform, of course, it can also be implemented by hardware, but in many cases the former is better.的实施方式。 Based on this understanding, the technical solution of the present invention essentially or the part that contributes to the existing technology can be embodied in the form of a software product, and the computer software product is stored in a storage medium (such as ROM/RAM) as described above. , Magnetic disk, optical disk), including several instructions to make a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) execute the method described in each embodiment of the present invention.

The above are only preferred embodiments of the present invention, and do not limit the scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the description and drawings of the present invention, or directly or indirectly applied to other related technical fields , The same reason is included in the scope of patent protection of the present invention.

Claims

A merchandise shopping guide method under the new retail mode, the method includes:

Acquiring the location information reported by the customer terminal, and determining whether the customer terminal is located in the target physical store according to the location information;

If the customer terminal is located in the target physical store, acquiring customer information reported by the customer terminal;

Calculate the similarity between the customer information and the transactional customer information stored in the historical transaction record to obtain multiple similarity calculation results;

Determine a maximum similarity calculation result from the multiple similarity calculation results, and determine whether the maximum similarity calculation result is less than a preset threshold;

If the maximum similarity calculation result is less than a preset threshold, input the customer information into a preset prediction model to obtain a product recommendation result for the customer information;

Acquire the face image reported by the customer terminal, and send the face image and the product recommendation result to the sales terminal corresponding to the target physical store.
The method of shopping guide for goods under the new retail mode according to claim 1, wherein the step of obtaining the location information reported by the customer terminal, and determining whether the customer terminal is located in the target physical store according to the location information comprises:

Acquiring the location information reported by the customer terminal, calculating the distance between the location information and the location information of each physical store, and obtaining multiple distance calculation results;

Determine whether there is a target calculation result less than or equal to a preset distance among the multiple distance calculation results;

If there is a target calculation result less than or equal to the preset distance among the multiple distance calculation results, it is determined that the customer terminal is located in the target physical store corresponding to the target calculation result.
The method of shopping guide for goods under the new retail mode according to claim 1, wherein the step of calculating the similarity between the customer information and each transaction customer information stored in historical transaction records to obtain multiple similarity calculation results include:

Obtain each customer's information that has been traded stored in historical transaction records;

The similarity between the customer information and the transaction customer information is calculated by the cosine similarity formula, and multiple similarity calculation results are obtained. The cosine similarity formula is as follows:

Among them, n(A) represents the number of types of information contained in the customer information, n(B) represents the number of types of information contained in the customer information that has been traded, and n(A∩B) represents that the customer information is the same as that of the customer information that has been traded The amount of information, K is the similarity calculation result.
The method of shopping guide for goods under the new retail mode according to claim 1, wherein, before the step of obtaining the location information reported by the customer terminal and determining whether the customer terminal is located in the target physical store according to the location information, the method further comprises :

Acquiring historical transaction records, the historical transaction records including the product name and the transaction customer information corresponding to the product name;

Calculate the feature value corresponding to the transaction customer information, and substitute the feature value into the prediction function formula to obtain multiple prediction functions;

Iteratively solving the multiple prediction functions to obtain a prediction model corresponding to the product name;

The prediction function formula is as follows:

among them,
θ i is the weight value of the transaction customer information i, x i is the characteristic value corresponding to the transaction customer information i, θ T =[θ 1 ,θ 2 ,...,θ n ], x=[x 1 ,x 2 ,...,x n ], e is a natural constant.
The method of shopping guide for goods under the new retail mode according to claim 4, wherein, if the calculation result of the maximum acquaintance degree is less than a preset threshold, the customer information is input into a preset prediction model to obtain The steps of product recommendation results for customer information include:

If the calculation result of the maximum acquaintance degree is less than the preset threshold, calculate the characteristic value corresponding to the customer information, and input the characteristic value into the prediction model corresponding to each product to obtain several output values;

A target output value greater than or equal to a preset probability value is selected, and a product recommendation result for the customer information is obtained based on the target output value.
The method of shopping guide for goods under the new retail mode according to claim 1, wherein the face image reported by the customer terminal is acquired, and the face image and the product recommendation result are sent to the target Before the steps of the sales terminal corresponding to the physical store, it also includes:

Obtaining a product list of the target physical store;

Detecting whether the product recommendation result exists in the product list;

If the product recommendation result exists in the product list, execute the acquisition of the face image reported by the customer terminal, and send the face image and the product recommendation result to the target physical store. The steps of the point-of-sale.
The merchandise shopping guide method under the new retail mode according to any one of claims 1 to 6, wherein said sending said face image and said product recommendation result to the sales terminal corresponding to said target physical store The steps include:

Acquiring status information of each sales terminal corresponding to the target physical store;

Determining the target sales terminal whose status is idle based on the status information;

Sending the face image and the product recommendation result to the target sales terminal.
A merchandise shopping guide device in a new retail mode. The merchandise shopping guide device in a new retail mode includes:

The location judgment module is used to obtain the location information reported by the customer terminal, and determine whether the customer terminal is located in the target physical store according to the location information;

A customer information acquisition module, configured to acquire customer information reported by the customer terminal if the customer terminal is located in a target physical store;

The calculation module is used to calculate the similarity between the customer information and the transactional customer information stored in the historical transaction records to obtain multiple similarity calculation results;

The detection module is configured to determine a maximum similarity calculation result from the multiple similarity calculation results, and determine whether the maximum similarity calculation result is less than a preset threshold;

A prediction module, configured to input the customer information into a preset prediction model if the maximum similarity calculation result is less than a preset threshold, to obtain a product recommendation result for the customer information;

The push module is configured to obtain the face image reported by the customer terminal, and send the face image and the product recommendation result to the sales terminal corresponding to the target physical store.
A computer device, wherein the computer device includes a memory and a processor;

The memory is used to store a computer program;

The processor is configured to execute the computer program and implement the following steps when the computer program is executed:

Acquiring the location information reported by the customer terminal, and determining whether the customer terminal is located in the target physical store according to the location information;

If the customer terminal is located in the target physical store, acquiring customer information reported by the customer terminal;

Calculate the similarity between the customer information and the transactional customer information stored in the historical transaction record to obtain multiple similarity calculation results;

Determine a maximum similarity calculation result from the multiple similarity calculation results, and determine whether the maximum similarity calculation result is less than a preset threshold;

If the maximum similarity calculation result is less than a preset threshold, input the customer information into a preset prediction model to obtain a product recommendation result for the customer information;

Acquire the face image reported by the customer terminal, and send the face image and the product recommendation result to the sales terminal corresponding to the target physical store.
8. The computer device according to claim 9, wherein the processor implements the acquisition of the location information reported by the customer terminal, and determines whether the customer terminal is located in the target physical store according to the location information, and specifically implements:

Acquiring the location information reported by the customer terminal, calculating the distance between the location information and the location information of each physical store, and obtaining multiple distance calculation results;

Determine whether there is a target calculation result less than or equal to a preset distance among the multiple distance calculation results;

If there is a target calculation result less than or equal to the preset distance among the multiple distance calculation results, it is determined that the customer terminal is located in the target physical store corresponding to the target calculation result.
9. The computer device according to claim 9, wherein the processor realizes the calculation of the similarity between the customer information and the transactional customer information stored in the historical transaction record to obtain multiple similarity calculation results, Implementation:

Obtain each customer's information that has been traded stored in historical transaction records;

The similarity between the customer information and the transaction customer information is calculated by the cosine similarity formula, and multiple similarity calculation results are obtained. The cosine similarity formula is as follows:

Among them, n(A) represents the number of types of information contained in the customer information, n(B) represents the number of types of information contained in the customer information that has been traded, and n(A∩B) represents that the customer information is the same as that of the customer information that has been traded The amount of information, K is the similarity calculation result.
9. The computer device according to claim 9, wherein the processor is configured to implement the acquisition of the location information reported by the customer terminal, and determine whether the customer terminal is located in front of the target physical store according to the location information, to achieve:

Acquiring historical transaction records, the historical transaction records including the product name and the transaction customer information corresponding to the product name;

Calculate the feature value corresponding to the transaction customer information, and substitute the feature value into the prediction function formula to obtain multiple prediction functions;

Iteratively solving the multiple prediction functions to obtain a prediction model corresponding to the product name;

The prediction function formula is as follows:

among them,
θ i is the weight value of the transaction customer information i, x i is the characteristic value corresponding to the transaction customer information i, θ T =[θ 1 ,θ 2 ,...,θ n ], x=[x 1 ,x 2 ,...,x n ], e is a natural constant.
The computer device according to claim 12, wherein the processor realizes that if the calculation result of the maximum acquaintance degree is less than a preset threshold, the customer information is input into a preset prediction model to obtain When the result of product recommendation based on customer information, the specific realization:

If the calculation result of the maximum acquaintance degree is less than the preset threshold, calculate the characteristic value corresponding to the customer information, and input the characteristic value into the prediction model corresponding to each product to obtain several output values;

A target output value greater than or equal to a preset probability value is selected, and a product recommendation result for the customer information is obtained based on the target output value.
The computer device according to claim 9, wherein the processor implements the acquisition of the face image reported by the customer terminal, and sends the face image and the product recommendation result to the target entity Before the sales terminal corresponding to the store, it is used to realize:

Obtaining a product list of the target physical store;

Detecting whether the product recommendation result exists in the product list;

If the product recommendation result exists in the product list, execute the acquisition of the face image reported by the customer terminal, and send the face image and the product recommendation result to the target physical store. The steps of the point-of-sale.
The computer device according to any one of claims 9 to 14, wherein when the processor realizes the sending of the face image and the product recommendation result to the sales terminal corresponding to the target physical store, Implementation:

Acquiring status information of each sales terminal corresponding to the target physical store;

Determining the target sales terminal whose status is idle based on the status information;

Sending the face image and the product recommendation result to the target sales terminal.
A computer-readable storage medium, wherein a product shopping guide program in a new retail mode is stored on the computer-readable storage medium, and the following steps are implemented when the product shopping guide program in a new retail mode is executed by a processor:

Acquiring the location information reported by the customer terminal, and determining whether the customer terminal is located in the target physical store according to the location information;

If the customer terminal is located in the target physical store, acquiring customer information reported by the customer terminal;

Calculate the similarity between the customer information and the transactional customer information stored in the historical transaction record to obtain multiple similarity calculation results;

Determine a maximum similarity calculation result from the multiple similarity calculation results, and determine whether the maximum similarity calculation result is less than a preset threshold;

If the maximum similarity calculation result is less than a preset threshold, input the customer information into a preset prediction model to obtain a product recommendation result for the customer information;

Acquire the face image reported by the customer terminal, and send the face image and the product recommendation result to the sales terminal corresponding to the target physical store.
The computer-readable storage medium according to claim 16, wherein the processor implements the acquisition of the location information reported by the customer terminal and determines whether the customer terminal is located in the target physical store according to the location information. :

Acquiring the location information reported by the customer terminal, calculating the distance between the location information and the location information of each physical store, and obtaining multiple distance calculation results;

Determine whether there is a target calculation result less than or equal to a preset distance among the multiple distance calculation results;

If there is a target calculation result less than or equal to the preset distance among the multiple distance calculation results, it is determined that the customer terminal is located in the target physical store corresponding to the target calculation result.
The computer-readable storage medium according to claim 16, wherein the processor obtains a plurality of similarity calculations after calculating the similarity between the customer information and each transaction customer information stored in the historical transaction record. As a result, the specific realization:

Obtain each customer's information that has been traded stored in historical transaction records;

The similarity between the customer information and the transaction customer information is calculated by the cosine similarity formula, and multiple similarity calculation results are obtained. The cosine similarity formula is as follows:

Among them, n(A) represents the number of types of information contained in the customer information, n(B) represents the number of types of information contained in the customer information that has been traded, and n(A∩B) represents that the customer information is the same as that of the customer information that has been traded The amount of information, K is the similarity calculation result.
The computer-readable storage medium according to claim 16, wherein the processor is configured to obtain the location information reported by the customer terminal and determine whether the customer terminal is located before the target physical store according to the location information. achieve:

Acquiring historical transaction records, the historical transaction records including the product name and the transaction customer information corresponding to the product name;

Calculate the feature value corresponding to the transaction customer information, and substitute the feature value into the prediction function formula to obtain multiple prediction functions;

Iteratively solving the multiple prediction functions to obtain a prediction model corresponding to the product name;

The prediction function formula is as follows:

among them,
θ i is the weight value of the transaction customer information i, x i is the characteristic value corresponding to the transaction customer information i, θ T =[θ 1 ,θ 2 ,...,θ n ], x=[x 1 ,x 2 ,...,x n ], e is a natural constant.
The computer-readable storage medium according to claim 19, wherein the processor, after realizing that if the calculation result of the maximum acquaintance degree is less than a preset threshold, inputs the customer information into a preset prediction model to obtain When the result of product recommendation based on the customer information, the specific realization is as follows:

If the calculation result of the maximum acquaintance degree is less than the preset threshold, calculate the characteristic value corresponding to the customer information, and input the characteristic value into the prediction model corresponding to each product to obtain several output values;

A target output value greater than or equal to a preset probability value is selected, and a product recommendation result for the customer information is obtained based on the target output value.