Method and apparatus for generating information

By analyzing the order and traffic data of the e-commerce platform, generating price elasticity level information, it solves the accuracy difficulties of store operators when pricing, and optimizes inventory management and revenue.

CN110348922BActive Publication Date: 2024-11-26BEIJING JINGDONG SHANGKE INFORMATION TECH CO LTD +1
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN201810295280.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2018-04-04
Publication Date
2024-11-26
Estimated Expiration
2038-04-04

AI Technical Summary

Technical Problem

Store operators of e-commerce platforms find it difficult to accurately determine whether the product price should be increased or lowered when pricing, resulting in difficulties in inventory management and profit optimization.

Method used

By obtaining order data of target products and their products of the same category, integrating and analyzing sales and traffic data, generating price elasticity level information, and adjusting product prices based on traffic level information.

Benefits of technology

It improves the accuracy of product price elasticity level information, helps store operators to price scientifically, and optimize inventory management and revenue.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN110348922B_ABST
    Figure CN110348922B_ABST
Patent Text Reader

Abstract

The embodiment of the present application discloses a method and device for generating information. A specific implementation of the method includes: obtaining order data of a target product and at least one product belonging to the same category as the target product within a first preset time period, the first preset time period including at least one first sub-time period; merging order data of the same product within the same first sub-time period into first order data; based on the order data within the first preset time period, determining information on sales volume changes over time for products with the same brand logo as the target product; based on the first order data within the first preset time period and the determined information on sales volume changes over time, generating price elasticity level information of the target product. This implementation can improve the accuracy of product price elasticity level information.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computer technology, and in particular to a method and device for generating information. Background Art

[0002] With the rapid rise of Internet technology, e-commerce has gradually become a mainstream business transaction model. As a new shopping concept, e-commerce has become an important application field of economic globalization with its advantages of convenience and speed.

[0003] When setting prices for products, store operators on e-commerce platforms usually need to judge based on experience whether the prices of products should be raised or lowered to increase profits. In practice, store operators usually manage a large number of products, and in order to avoid unsalable products and reduce inventory resources, they often need to adjust product prices. Summary of the invention

[0004] The embodiments of the present application provide a method and device for generating information.

[0005] In a first aspect, an embodiment of the present application provides a method for generating information, the method comprising: obtaining order data of a target product and at least one product belonging to the same category as the target product within a first preset time period, the first preset time period including at least one first sub-time period; merging the order data of the same product within the same first sub-time period into first order data; based on the order data within the first preset time period, determining information on sales volume changes over time for a product having the same brand logo as the target product; based on the first order data within the first preset time period and the determined information on sales volume changes over time, generating price elasticity level information of the target product.

[0006] In some embodiments, the method also includes: obtaining traffic data of the target store within a second preset time period, the traffic data including browsing records of the target store; based on the traffic data, determining the number of times the target store is browsed and the number of times the target product is browsed; based on the number of times the target store is browsed and the number of times the target product is browsed, generating traffic level information of the target product in the target store; based on the price elasticity level information of the target product and the traffic level information of the target product in the target store, generating price adjustment information of the target product in the target store.

[0007] In some embodiments, order data of the same product in the same first sub-time period are merged into first order data, including: filtering order data in a first preset time period, and deleting order data with a transaction price less than a preset price threshold; based on the filtered order data, performing robust regression on sales volume and transaction price, and deleting order data with residuals greater than the residual threshold; for each first sub-time period in at least one first sub-time period, based on the order data after robust regression, the order data of the same product in the first sub-time period are merged to generate first order data.

[0008] In some embodiments, the first preset time period includes at least one second sub-time period, the second sub-time period includes at least one first sub-time period, and based on the order data within the first preset time period, the information on how the sales volume of a product with the same brand logo as the target product changes over time is determined, including: for each second sub-time period in at least one second sub-time period, the order data of the product with the same brand logo as the target product within the second sub-time period is merged into second order data; based on at least one second sub-time period and the second order data within the corresponding second sub-time period, the information on how the sales volume of a product with the same brand logo as the target product changes over time is determined.

[0009] In some embodiments, before the order data of products with the same brand identification as the target product in the second sub-time period are merged into the second order data, the sales volume changes over time of the products with the same brand identification as the target product are determined based on the order data in the first preset time period. The determined sales volume changes over time also include: among the brand identifications of products with the same category identification as the target product, merging the brand identifications of brands with a product quantity less than or equal to a preset product quantity.

[0010] In some embodiments, based on the first order data within a first preset time period and the information on the sales volume changes over time, price elasticity level information of the target product is generated, including: selecting the first order data of products whose price change times and sales volume change times exceed the preset change times threshold from the first order data within the first preset time period to generate a first sample set; based on the first sample set and the information on the sales volume changes over time, determining the first price elasticity information of the target product and the first price elasticity information of the category represented by the category identifier of the target product to generate a price elasticity information set; selecting the first order data of products with the same brand identifier as the target product from the first sample set to generate a second sample set; based on the second sample set and the information on the sales volume changes over time, determining the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand identifier of the target product, and adding the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand identifier of the target product to the price elasticity information set; based on the price elasticity information set, generating price elasticity level information of the target product.

[0011] In some embodiments, the first preset time period includes at least one second sub-time period, the second sub-time period includes at least one first sub-time period, and before generating the price elasticity level information of the target product, based on the first order data in the first preset time period and the determined information on sales volume changes over time, the price elasticity level information of the target product is generated, further comprising: for each second sub-time period in at least one second sub-time period, determining the first order data of a product whose number of first order data in the second sub-time period is greater than a preset number, and merging the first order data determined for the same product in the second sub-time period into third order data; selecting the third order data of the product whose price change times and sales volume change times exceed the preset change times threshold from the third order data to generate a third sample set; based on the first Based on the three sample sets and the information on sales volume changes over time, determine the third price elasticity information of the target product and the second price elasticity information of the category represented by the category identifier of the target product, and add the third price elasticity information of the target product and the second price elasticity information of the category represented by the category identifier of the target product to the price elasticity information set; select the third order data of the product with the same brand identifier as the target product from the third sample set to generate a fourth sample set; based on the fourth sample set and the information on sales volume changes over time, determine the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand identifier of the target product, and add the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand identifier of the target product to the price elasticity information set.

[0012] In some embodiments, before generating the second sample set, price elasticity level information of the target product is generated based on the first order data within a first preset time period and the information on the determined sales volume changes over time, and also includes: among the products and brand logos involved in the first sample set, the brand logos of brands whose product quantity is less than or equal to the preset product quantity are merging.

[0013] In some embodiments, price elasticity grade information of a target product is generated based on a price elasticity information set, including: determining at least one price elasticity information having a coefficient of determination greater than a coefficient of determination threshold from the price elasticity information set; determining the price elasticity information having a significance test statistic greater than the significance test statistic threshold and the highest significance test statistic among the at least one price elasticity information as the price elasticity information of the target product; and generating price elasticity grade information of the target product based on a size relationship between the determined price elasticity information and the price elasticity information threshold.

[0014] In some embodiments, price elasticity level information of a target product is generated based on first order data within a first preset time period and information on sales volume changes over time, including: extracting a value of at least one attribute of the target product from the first order data within the first preset time period; for each of the at least one attribute, determining the quartile and / or standard deviation of the value extracted for the attribute as the attribute value of the attribute; inputting the attribute value of at least one attribute into a pre-trained random forest model to obtain price elasticity information of the target product, the random forest model being used to characterize the correspondence between attribute values ​​and price elasticity information; and generating price elasticity level information of the target product based on the size relationship between the price elasticity information of the target product and a price elasticity information threshold.

[0015] In some embodiments, the random forest model is trained by the following steps: obtaining multiple training samples, each training sample including attribute values ​​of a product having the same category identifier as the target product and price elasticity information predetermined based on the attribute values ​​of the product; taking the attribute value of each training sample as input and the price elasticity information of each training sample as output, and training the random forest model.

[0016] In some embodiments, based on the number of times the target store is browsed and the number of times the target product is browsed, traffic level information of the target product in the target store is generated, including: determining the number of products in the target store whose sales volume is greater than a preset sales volume and / or the number of times they are browsed is greater than a preset number of times; determining a browsing count threshold based on the number of times the target store is browsed and the determined number of products; generating traffic level information of the target product in the target store based on the relationship between the number of times the target product is browsed and the browsing count threshold.

[0017] In a second aspect, an embodiment of the present application provides a device for generating information, the device comprising: an order acquisition unit, configured to acquire order data of a target product and at least one product belonging to the same category as the target product within a first preset time period, the first preset time period including at least one first sub-time period; an order fusion unit, configured to fuse order data of the same product within the same first sub-time period into first order data; a trend determination unit, configured to determine information on sales volume changes over time of a product having the same brand logo as the target product based on the order data within the first preset time period; and an elasticity generation unit, configured to generate price elasticity level information of the target product based on the first order data within the first preset time period and the determined information on sales volume changes over time.

[0018] In some embodiments, the device also includes: a traffic acquisition unit, configured to acquire traffic data of the target store within a second preset time period, the traffic data including browsing records of the target store; a number determination unit, configured to determine the number of times the target store is browsed and the number of times the target product is browsed based on the traffic data; a traffic generation unit, configured to generate traffic level information of the target product in the target store based on the number of times the target store is browsed and the number of times the target product is browsed; a price adjustment unit, configured to generate price adjustment information of the target product in the target store based on the price elasticity level information of the target product and the traffic level information of the target product in the target store.

[0019] In some embodiments, the order fusion unit includes: a filtering module, configured to filter order data within a first preset time period, and delete order data with a transaction price less than a preset price threshold; a regression module, configured to perform robust regression on sales volume and transaction price based on the filtered order data, and delete order data with residuals greater than the residual threshold; a fusion module, configured to fuse the order data of the same product within each first sub-time period in at least one first sub-time period based on the order data after robust regression to generate first order data.

[0020] In some embodiments, the first preset time period includes at least one second sub-time period, the second sub-time period includes at least one first sub-time period, and the trend determination unit includes: a fusion module, configured to, for each second sub-time period in at least one second sub-time period, fuse the order data of products with the same brand logo as the target product in the second sub-time period into second order data; and a determination module, configured to determine information on changes in sales volume of products with the same brand logo as the target product over time based on at least one second sub-time period and the second order data in the corresponding second sub-time period.

[0021] In some embodiments, the trend determination unit further includes: a brand merging module configured to merge brand logos of brands containing a number of products less than or equal to a preset number of products among the brand logos of products having the same category logo as the target product.

[0022] In some embodiments, the elasticity generation unit includes: a first selection module, configured to select the first order data of products whose price change times and sales volume change times exceed a preset change number threshold from the first order data within a first preset time period, and generate a first sample set; a first determination module, configured to determine the first price elasticity information of the target product and the first price elasticity information of the category represented by the category identifier of the target product based on the first sample set and the determined information on sales volume changes over time, and generate a price elasticity information set; a second selection module, configured to select the first order data of products with the same brand identifier as the target product from the first sample set, and generate a second sample set; a second determination module, configured to determine the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand identifier of the target product based on the second sample set and the determined information on sales volume changes over time, and add the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand identifier of the target product to the price elasticity information set; an elasticity generation module, configured to generate price elasticity level information of the target product based on the price elasticity information set.

[0023] In some embodiments, the first preset time period includes at least one second sub-time period, the second sub-time period includes at least one first sub-time period, and the elastic generation unit further includes: a fusion module, configured to determine, for each second sub-time period in at least one second sub-time period, first order data of products whose number of first order data in the second sub-time period is greater than a preset number, and fuse the first order data of the same product determined in the second sub-time period into third order data; a third selection module, configured to select third order data of products whose price change times and sales change times exceed a preset change times threshold from the third order data to generate a third sample set; a third determination module, configured to determine the target product based on the third sample set and the determined information on sales changes over time. The third price elasticity information of the target product and the second price elasticity information of the category represented by the category identifier of the target product are added to the price elasticity information set; a fourth selection module is configured to select the third order data of the product with the same brand identifier as the target product from the third sample set to generate a fourth sample set; a fourth determination module is configured to determine the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand identifier of the target product based on the fourth sample set and the determined information on sales volume changes over time, and add the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand identifier of the target product to the price elasticity information set.

[0024] In some embodiments, the elasticity generation unit further includes: a brand merging module configured to merge brand logos of brands whose product quantity is less than or equal to a preset product quantity among the products and brand logos involved in the first sample set.

[0025] In some embodiments, the elasticity generation module also includes: an elasticity screening submodule, configured to determine at least one price elasticity information whose coefficient of determination is greater than a coefficient of determination threshold from the price elasticity information set; an elasticity determination submodule, configured to determine the price elasticity information whose significance test statistic is greater than the significance test statistic threshold and has the highest significance test statistic among the at least one price elasticity information as the price elasticity information of the target product; and an elasticity generation submodule, configured to generate price elasticity grade information of the target product based on the size relationship between the determined price elasticity information and the price elasticity information threshold.

[0026] In some embodiments, the elasticity generation unit includes: an attribute extraction module, configured to extract the value of at least one attribute of the target product from the first order data within a first preset time period; an attribute value determination module, configured to determine, for each of the at least one attribute, the quartile and / or standard deviation of the value extracted for the attribute as the attribute value of the attribute; an elasticity determination module, configured to input the attribute value of at least one attribute into a pre-trained random forest model to obtain price elasticity information of the target product, the random forest model being used to characterize the correspondence between the attribute value and the price elasticity information; an elasticity generation module, configured to generate price elasticity level information of the target product based on the size relationship between the price elasticity information of the target product and a price elasticity information threshold.

[0027] In some embodiments, the random forest model is trained by the following steps: obtaining multiple training samples, each training sample including attribute values ​​of a product having the same category identifier as the target product and price elasticity information predetermined based on the attribute values ​​of the product; taking the attribute value of each training sample as input and the price elasticity information of each training sample as output, and training the random forest model.

[0028] In some embodiments, the traffic generation unit includes: a product determination module, configured to determine the number of products in the target store whose sales volume is greater than a preset sales volume and / or whose browsing times are greater than a preset number of times; a threshold determination module, configured to determine a browsing times threshold based on the number of times the target store is browsed and the determined number of products; and a traffic generation module, configured to generate traffic level information of the target product in the target store based on the relationship between the number of times the target product is browsed and the browsing times threshold.

[0029] In a third aspect, an embodiment of the present application provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by one or more processors, the one or more processors implement the method described in any implementation method in the first aspect.

[0030] In a fourth aspect, an embodiment of the present application provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any implementation manner in the first aspect.

[0031] The method and device for generating information provided in the embodiments of the present application obtain order data of a target product and at least one product belonging to the same category as the target product within a preset time period, and then merge the order data of the same product within the same sub-time period into first order data, and then determine the sales volume change over time of products with the same brand logo as the target product based on the order data within the preset time period, and finally generate price elasticity level information of the target product based on the first order data within the preset time period and the determined sales volume change over time information, thereby improving the accuracy of the product price elasticity level information. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0033] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0034] Figure 2 is a flow chart of an embodiment of a method for generating information according to the present application;

[0035] Figure 3 yes Figure 2 A flow chart of one implementation of the illustrated embodiment;

[0036] Figure 4 yes Figure 2 A flow chart of another implementation of the illustrated embodiment;

[0037] Figure 5 is a flow chart of another embodiment of a method for generating information according to the present application;

[0038] Figure 6 is a schematic diagram of an application scenario of the method for generating information according to the present application;

[0039] Figure 7 is a structural schematic diagram of an embodiment of a device for generating information according to the present application;

[0040] Figure 8 It is a structural diagram of a computer system suitable for implementing a server of an embodiment of the present application. DETAILED DESCRIPTION

[0041] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It should also be noted that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0042] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0043] Figure 1 An exemplary system architecture 100 is shown to which embodiments of the method for generating information or the apparatus for generating information of the present application can be applied.

[0044] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0045] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as product management applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0046] Terminal devices 101, 102, 103 can be hardware or software. When terminal devices 101, 102, 103 are hardware, they can be various electronic devices with display screens, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, etc. When terminal devices 101, 102, 103 are software, they can be installed in the electronic devices listed above. They can be implemented as multiple software or software modules (for example, to provide distributed services), or they can be implemented as a single software or software module. No specific limitation is made here.

[0047] The server 105 may be a server that provides various services, such as a background server that provides support for product management applications running on the terminal devices 101, 102, and 103. The background server may analyze and process the received product management request and other data, and generate processing results (e.g., price elasticity level information) for the terminal device or other server to obtain.

[0048] It should be noted that the method for generating information provided in the embodiment of the present application is generally executed by the server 105 , and accordingly, the device for generating information is generally set in the server 105 .

[0049] It should be noted that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, used to provide distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0050] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0051] Continue to refer Figure 2 , shows a process 200 of an embodiment of a method for generating information according to the present application. The method for generating information comprises the following steps:

[0052] Step 201, obtaining order data of a target product and at least one product belonging to the same category as the target product within a first preset time period, where the first preset time period includes at least one first sub-time period.

[0053] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in the figure can obtain the order data of the target product and at least one product belonging to the same category as the target product (for example, having the same category identification) within a first preset time period (for example, two years) from the database through a wired connection or a wireless connection. The first preset time period includes at least one first sub-time period (for example, the first sub-time period is one day, one week, one month, etc.). The database stores the order data of the products, and the order data may include data information related to the products and orders, including but not limited to: product identification, category identification, brand identification, order time, page price, transaction price, sales volume, inventory, store identification, etc. For example, when the target product is a mobile phone of model XX, the above-mentioned execution entity can obtain the order data of the target product and at least one product with the category identification of "mobile phone" within two years. Here, the database can be deployed on a server, for example, it can be deployed on Figure 1 It can be deployed on the server shown, or on other servers.

[0054] Here, the same category may mean that the target product has the same category identifier as the at least one product mentioned above, or it may mean that the target product has a different category identifier than the at least one product mentioned above, but both products substantially belong to the same category.

[0055] It should be noted that the above-mentioned wireless connection methods may include but are not limited to 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0056] Step 202: merge the order data of the same product in the same first sub-time period into first order data.

[0057] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in the figure can use the first sub-time period as the time unit to merge the order data obtained in step 201 according to the product (for example, summing the sales volume of the same product in the same sub-time period, weighted averaging the transaction price of the same product in the same sub-time period, and other data conversion processing). That is, the order data of the same product in the same first sub-time period is merged into the first order data (for example, the order data of product A in the same day is merged into the first order data). By merging the order data, the amount of order data is effectively reduced, the speed of generating information is improved, and the price elasticity analysis is performed using the first sub-time period (day) as the time unit, which can better reflect the price elasticity of the product.

[0058] In practice, due to price discount activities such as full discount, gifts, sets, shopping vouchers, etc., there are usually abnormal order data in the order data whose transaction price and / or sales volume exceed the preset range. Therefore, in order to reduce the impact of abnormal order data on price elasticity information, it is necessary to filter the order data within the first preset time period.

[0059] In some optional implementations of this embodiment, step 202 may specifically include:

[0060] In the first step, the order data within the first preset time period is filtered, and the order data with a transaction price less than the preset price threshold is deleted. Here, the preset price threshold can be a relative value, such as 30%, 50%, 60% of the page price, etc., and the preset price threshold can also be an absolute value, such as 1 yuan, etc. By eliminating the order data with a transaction price less than the preset price threshold, the impact of price discount activities on the price elasticity information of the product can be reduced.

[0061] The second step is to perform robust regression on sales volume and transaction price based on the filtered order data, and delete the order data with residuals greater than the residual threshold (for example, 2.5 times the residual standard deviation). In addition to the price discount factor, there are other factors that are not easy to find directly (including but not limited to natural disasters, emergencies, etc.) that cause a sudden change in the quantity-price relationship within a first sub-time period (for example, a day or a week, etc.). Therefore, the above-mentioned execution entity can perform robust regression (i.e., Robust Regression, which is to use robust estimation methods for regression models to fit the structure of most data, so as to identify potential outliers, strong influence points, and structures that deviate from model assumptions) on sales volume (or the value after taking the logarithm of sales volume) and transaction price (or the value after taking the logarithm of transaction price) based on the filtered order data to eliminate order data that constitute outliers, strong influence points, etc.

[0062] Specifically, the execution subject can use the robust regression model to fit the relationship between sales volume and transaction price to obtain a fitting function. Then, the residual between the actual sales volume value and the model fitting value of each point (i.e., order data) is determined, so that the order data with a residual greater than the residual threshold can be eliminated.

[0063] As an example, the residual ε and residual standard deviation σ can be calculated by the following formula:

[0064]

[0065]

[0066] Among them, y is the true value of sales volume, is the fitted value of sales volume, M is the number of order data, and i is the order data number.

[0067] In this embodiment, the residual threshold may be a relative value, such as 2 times or 2.5 times the residual standard deviation, or an absolute value, which may be set by a person skilled in the art based on the needs of the actual application scenario.

[0068] In the third step, for each first sub-time period in at least one first sub-time period, based on the order data after robust regression, the order data in the first sub-time period is fused by product by, for example, summing the sales volume, weighted averaging the transaction price, etc., to generate first order data. For example, one piece of first order data is generated for product A, one piece of first order data is generated for product B, and so on. Through the above data fusion processing, any product has at most one piece of first order data in each first sub-time period (when there is no order data for a product in the first sub-time period, there is no first order data for the product).

[0069] Although the above implementation describes filtering and robust regression of the order data within the first preset time period and then fusing the order data into the first order data, the present embodiment is not limited thereto.

[0070] In some optional implementations of this embodiment, step 202 may include:

[0071] In the first step, for each first sub-time period in at least one first sub-time period, the order data in the first sub-time period is integrated according to products by methods such as summing sales volume and weighted averaging transaction prices to generate candidate first order data.

[0072] In the second step, the candidate first order data within the first preset time period are filtered, and the candidate first order data with a transaction price less than a preset price threshold are deleted.

[0073] In the third step, based on the filtered candidate first order data, robust regression is performed on the sales volume and the transaction price, and the candidate first order data with a residual less than or equal to a residual threshold (for example, 2.5 times the residual standard deviation) is determined as the first order data.

[0074] Step 203: Based on the order data in the first preset time period, determine the information on the change in sales volume of the product with the same brand logo as the target product over time.

[0075] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in FIG. 1 may first determine products with the same brand logo as the target product. Then, based on the order data of the determined products (including the target product), determine information on changes in sales volume of products with the same brand logo as the target product over time.

[0076] In addition to being affected by price, product sales will also naturally grow or fluctuate periodically over time. It can usually be assumed that the sales information of products of the same brand is the same or similar over time. By determining the sales information of products with the same brand identity as the target product over time, the impact of time factors on product sales can be eliminated (for example, for any order data, the sales volume caused by time factors can be determined based on the order date and the information on the sales volume of the product corresponding to the product identity in the order data over time, and then the sales volume determined to be caused by time factors can be eliminated from the sales volume of the order data (for example, the two are subtracted)), which can improve the accuracy of price elasticity information.

[0077] As an example, information on the change in sales volume over time of a product with the same brand logo as the target product can be obtained by the following steps: first, order data containing the brand logo of the target product is selected from order data within a first preset time period; then, based on the selected order data, a trend line is fitted with the order date as an independent variable and the sales volume as a dependent variable (for example, by fitting the trend line by means of an exponential function, a power function, a polynomial, etc.), thereby obtaining information on the change in sales volume over time of the product with the brand logo of the target product. Here, any information that can reflect the change in the sales volume of a product over time can be used as information on the change in sales volume over time, for example, the fitted trend line (or the functional expression of the trend line).

[0078] In some optional implementations of this embodiment, the first preset time period includes at least one second sub-time period, and the second sub-time period includes at least one first sub-time period (for example, the first preset time period is two years, the second sub-time period is one week, and the first sub-time period is one day).

[0079] Step 203 may specifically include:

[0080] In the first step, for each second sub-time period in the at least one second sub-time period, the order data of products with the same brand logo as the target product in the second sub-time period are merged according to products by methods such as summing up the sales volume and weighted averaging the transaction prices, so as to generate second order data.

[0081] The second step is to determine the information on the change of sales volume of the product with the same brand logo as the target product over time (with the second sub-time period as the time unit) based on the at least one second sub-time period and the second order data in the corresponding second sub-time period. Specifically, the above-mentioned execution subject performs LOWESS regression (Locally Weighted Scatterplot Smoothing, abbreviated as LOWESS or LOESS, local weighted regression scatter plot smoothing method, which is a tool for viewing the relationship between two-dimensional variables. The main idea of ​​LOWESS regression is to take a certain proportion of local data, fit a polynomial regression curve in the selected local data, and determine the laws and trends of the data displayed locally) on the sales volume of the product with the same brand logo as the target product over time (with the second sub-time period as the time unit).

[0082] Optionally, before generating the second order data, step 203 further includes: in the brand identification of the product having the same category identification as the target product, merging the brand identifications of brands whose product quantity is less than or equal to the preset product quantity. Here, merging the brand identifications means using a new brand identification (the new brand identification may be a fictitious brand identification and does not represent a real brand) to represent each brand identification to be merged, so as to achieve the effect of "brand merging". For example, in the category represented by the category identification of the target product, the number of products with the brand identification "Brand A" is 10, the number of products with the brand identification "Brand B" is 2, the number of products with the brand identification "Brand C" is 3, and the number of products with the brand identification "Brand D" is 5. When the preset product quantity is 3, "Brand B" and "Brand C" are replaced by "Brand X" (new brand identification), which is equivalent to the number of products with the brand identification "Brand X" being 5.

[0083] By merging the brand logos of brands with a small number of products, when the number of products with the same brand logo as the target product is small, more order data can be obtained for order data fusion and LOWESS regression analysis, thereby obtaining more accurate information on sales changes over time, thereby avoiding the problem that the information on sales changes over time cannot be fitted due to the small number of products or the fitted information on sales changes over time is not accurate enough.

[0084] Although the above implementation describes that the order data of the product having the same brand identification as the target product in the second sub-time period is merged into the second order data, the present embodiment is not limited thereto.

[0085] Optionally, the first order data of products having the same brand logo as the target product in the second sub-time period may also be merged into the second order data, thereby improving the fusion efficiency of the second order data.

[0086] Step 204: Generate price elasticity level information of the target product based on the first order data within the first preset time period and the determined information on sales volume changes over time.

[0087] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in FIG. 200 may determine the price elasticity information of the target product through a price elasticity model (e.g., a mixed effect model, a random forest model, etc.) based on the first order data generated in step 202 and the information on the change in sales volume over time determined in step 203. Then, the price elasticity level information of the target product is generated according to the relationship between the price elasticity information and the price elasticity information threshold.

[0088] Price elasticity information refers to the dependence of supply and demand on price changes, reflecting the response rate of supply and demand caused by price changes, that is, the sensitivity of supply and demand to price information, also known as supply and demand price elasticity information. Price elasticity information γ can be calculated by the following formula:

[0089]

[0090] Among them, Q0 is the initial sales volume, ΔQ is the change in sales volume, p0 is the initial price, and Δp is the change in price.

[0091] Price elasticity information is usually a negative value, that is, an increase in price will reduce sales. Those skilled in the art can set the value range of price elasticity information according to business needs. For example, it can be set to a value between -8 and -0.25. For example, the price elasticity information can be -1.5, -5, etc. The price elasticity information threshold is used to determine the price elasticity level information. The price elasticity information threshold can also be set to a value between -8 and -0.25. For example, the price elasticity information threshold is -2.5. Price elasticity level information (for example, "high price elasticity" and "low price elasticity") is used to characterize the price elasticity level of the product. For example, "high price elasticity" (the price elasticity information of the product is greater than or equal to the price elasticity information threshold) indicates that the sales volume of the product changes significantly with price, while "low price elasticity" (the price elasticity information of the product is less than the price elasticity information threshold) indicates that the sales volume of the product does not change significantly with price.

[0092] Although the present embodiment describes a situation where there is one price elasticity information threshold and two price elasticity level information, the present application is not limited thereto. It should be understood that the price elasticity information threshold may include multiple, and accordingly, the price elasticity level information may also include multiple. For example, when the price elasticity information threshold includes two, the price elasticity level information includes three, and those skilled in the art may set it according to the needs of the actual application scenario.

[0093] Continue to refer Figure 3 , shows a process 300 of an implementation of the method for generating information in this embodiment. Figure 3 As shown, based on the first order data within the first preset time period and the information on the determined sales volume changing over time, the price elasticity level information of the target product is generated (ie, Figure 2 Step 204) may include:

[0094] Step 311, select the first order data of products whose transaction price change times and sales volume change times exceed a preset change times threshold (for example, 8 times, a sufficient number of change points can improve the fitting success rate) from the first order data 301 (for example, the first order data generated by step 202) within a first preset time period (for example, two years) to generate a first sample set.

[0095] Step 312, based on the above-mentioned first sample set and the information on sales volume changes over time (for example, the information on sales volume changes over time determined by step 203), determine the first price elasticity information of the target product and the first price elasticity information of the category represented by the category identifier of the target product, and generate a price elasticity information set 302.

[0096] Since the price elasticity information of products with the same category identifier is similar, the price elasticity information γ of the i-th product with the same category identifier as the target product is i It can be expressed as:

[0097] γ i =γ′+dev i , dev i ~N(0,σ 2 ) (4)

[0098] Among them, γ is the price elasticity information, γ' is the price elasticity information of the category represented by the category identifier of the target product, and dev i is the value of the price elasticity information deviation γ' of the i-th product, N(0,σ 2 ) means the mathematical expectation is 0 and the variance is σ 2 Normal distribution of , i is the product number.

[0099] For each first order data in the first sample set, the execution entity may first determine the sales volume of the order product due to the time factor (for example, the sales volume of the point on the fitted trend line corresponding to the order date), and then eliminate the sales volume due to the time factor from the sales volume of the first order data (for example, subtracting the two), and determine the first order data after eliminating the time factor as a training sample.

[0100] Using the above training samples, the above execution entity can establish a mixed effect model, then:

[0101] log2(Q ij )=(γ′+dev i )×log2(P ij )+C′+C i (5)

[0102] Among them, P is the transaction price of the product, Q is the sales volume of the product, and Pij is the transaction price of the i-th product in the j-th order data with the same category ID as the target product, Q ij is the sales volume of the i-th product in the j-th order data of the i-th product with the same category identifier as the target product, C' is the intercept of the category represented by the category identifier of the target product, and C i is the value of the intercept deviation of the i-th product itself from C', and j is the order data number.

[0103] Through the mixed effect model, the execution subject can fit the first price elasticity information of the target product and the first price elasticity information of the category represented by the category identifier of the target product. Then, a price elasticity information set 302 is generated based on the first price elasticity information of the target product and the first price elasticity information of the category represented by the category identifier of the target product.

[0104] Step 313: Select the first order data of the product having the same brand logo as the target product from the first sample set to generate a second sample set.

[0105] Step 314, based on the second sample set and the information on sales volume changes over time, determine the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand logo of the target product, and add the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand logo of the target product to the price elasticity information set 302.

[0106] Since the price elasticity information of products with the same brand logo is similar, the execution subject can establish a mixed effect model based on the second sample set, so as to fit the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand logo of the target product. Then, the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand logo of the target product are added to the price elasticity information set 302.

[0107] The process of the execution subject fitting the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand logo of the target product based on the second sample set can be referred to the specific description of step 312, which is not repeated here.

[0108] Step 330: Generate price elasticity level information of the target product based on the price elasticity information set.

[0109] Optionally, step 330 may include:

[0110] Step 331: Determine at least one price elasticity information whose coefficient of determination is greater than a threshold value of the coefficient of determination from the price elasticity information set.

[0111] The coefficient of determination is a statistical indicator used to reflect the reliability of the dependent variable in the regression model, generally represented by the symbol "R 2 "It can be defined as the proportion of the total variation of the dependent variable that is explained by all the independent variables in the model (variation refers to the deviation of each value of a variable from its mean, usually expressed as the sum of squared deviations, while dispersion refers to the deviation of a particular value from its mean).

[0112] In this implementation, the coefficient of determination can be used to reflect the statistical index of the reliability of the mixed effect model (for example, the mixed effect model established by steps 312 and 314), which can be expressed as the traded price (for example, log2(P ij )) describes the sales volume (e.g., log2(Q ij ))'s total variance. The range of the coefficient of determination is between 0 and 1. The closer the coefficient of determination is to 1, the better the regression equation (e.g., formula (5)) fits the sales volume. On the contrary, the closer the coefficient of determination is to 0, the worse the regression equation fits the sales volume.

[0113] The determination coefficient threshold is a preset value, for example, 0.3, which is used to ensure that the established mixed effect model can fit reliable sales. The above-mentioned execution subject compares the determination coefficient of the established mixed effect model with the determination coefficient threshold, and determines at least one price elasticity information determined by the mixed effect model whose determination coefficient is greater than the determination coefficient threshold.

[0114] Step 332: determine the price elasticity information whose significance test statistic is greater than the significance test statistic threshold and has the highest significance test statistic among the at least one price elasticity information as the price elasticity information of the target product.

[0115] In this implementation, the execution subject performs a significance test (e.g., t-test, f-test, etc.) on the at least one price elasticity information determined in step 331 to determine whether there is price elasticity information whose significance test statistic (e.g., t value in t-test) is greater than a significance test statistic threshold (e.g., -1.64, -1.96, etc.) in the at least one price elasticity information. In response to the existence of at least one price elasticity information that meets the above conditions, the price elasticity information with the highest significance test statistic is determined as the price elasticity information of the target product.

[0116] For example, through a t-test, it can be determined that there are two price elasticity information in the at least one price elasticity information mentioned above (for example, the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand logo of the target product), and the price elasticity information with the highest significance test statistic (for example, the second price elasticity information of the target product) is determined as the price elasticity information of the target product.

[0117] Step 333: Generate price elasticity level information of the target product based on the determined relationship between the price elasticity information and the price elasticity information threshold.

[0118] Generally speaking, product categories can include multiple levels, for example, primary categories, secondary categories, tertiary categories, etc. Among them, the primary category includes at least one secondary category, the secondary category includes multiple tertiary categories, and so on. Generally speaking, the price elasticity information of products in the same tertiary category can be compared, while the price elasticity information of products in different tertiary categories cannot be compared. For example, the absolute value of the price elasticity information of fast-moving consumer goods will be higher than the absolute value of the price elasticity information of durable consumer goods. Therefore, the size of the price elasticity information is for the comparison between products with similar attributes.

[0119] In this implementation, the three-level category can be used as the comparison range (in actual application, it can be adjusted accordingly according to actual needs). The price elasticity threshold can be obtained by calculating the median, mean, mode, etc. of the price elasticity information of the target three-level category products. The price elasticity information of the target product is compared with the price elasticity threshold to determine the price elasticity level information of the target product.

[0120] Taking the median as an example, for a certain three-level category, the price elasticity information of product A, product B, product C, and product D with the three-level category logo is -1.5, -2, -2.5, and -2.5 respectively, then the price elasticity information threshold can be determined to be -2.25. In this case, if the target product is product A or product B, the price elasticity level information of the target product is "low price elasticity"; if the target product is product C or product D, the price elasticity level information of the target product is "high price elasticity".

[0121] The benefits of using mixed effects models are:

[0122] First, by using the hierarchical structure of "product-brand" and "product-category", the coverage of price elasticity information fitting can be improved. For example, for products whose price elasticity information cannot be fitted due to a small amount of data, the price elasticity information of the brand or the price elasticity information of the category can be used instead.

[0123] Secondly, the fitted price elasticity information comprehensively reflects the attributes of the product itself and the attributes of the brand / category.

[0124] Finally, it is possible to fit products of one brand or category at the same time, with high fitting efficiency.

[0125] Optionally, the first preset time period includes at least one second sub-time period, and the second sub-time period includes at least one first sub-time period (for example, the first preset time period is two years, the second sub-time period is one week, and the first sub-time period is one day). The price elasticity level information of the target product is generated based on the first order data within the first preset time period and the information on the determined sales volume changing over time (i.e., Figure 2 Step 204) shown may also include:

[0126] Step 320, for each second sub-time period in at least one second sub-time period, determine the first order data of the product whose number of first order data in the second sub-time period is greater than a preset number (for example, 3), and merge the first order data of the same product determined in the second sub-time period into third order data (for example, merge the first order data of product A in each week into third order data by summing up the sales volume, weighted averaging the transaction price, etc.).

[0127] Step 321 , selecting the third order data of products whose price change times and sales volume change times exceed a preset change times threshold (eg, 8 times) from the third order data to generate a third sample set.

[0128] Step 322: Based on the third sample set and the information on sales volume changes over time (e.g., the information on sales volume changes over time determined in step 203), determine the third price elasticity information of the target product and the second price elasticity information of the category represented by the category identifier of the target product. Then, the third price elasticity information of the target product and the second price elasticity information of the category represented by the category identifier of the target product are added to the price elasticity information set 302.

[0129] Step 323: Select the third order data of the product having the same brand logo as the target product from the third sample set to generate a fourth sample set.

[0130] Step 324: Based on the fourth sample set and the determined information about the sales volume changing over time, determine the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand logo of the target product. Then, add the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand logo of the target product to the price elasticity information set 302.

[0131] It should be noted that the process of generating price elasticity information in step 322 and step 324 can refer to the specific description of step 312, which will not be repeated here.

[0132] Optionally, before step 313, price elasticity level information of the target product is generated based on the first order data within the first preset time period and the information on the determined sales volume change over time (ie, Figure 2 Step 204) shown may also include: among the products and brand identifiers involved in the first sample set, merging brand identifiers of brands whose product quantity is less than or equal to the preset product quantity. Since among the brands represented by the brand identifiers in the first order data in the first sample set, there may be brands whose product quantity is less than the preset product quantity, when fitting the price elasticity information of the brands, there may be a problem that there are too few order data to fit. By merging the brand identifiers of brands whose product quantity is less than the preset product quantity, the above problem can be effectively avoided.

[0133] Similarly, before step 323, the price elasticity level information of the target product is generated based on the first order data within the first preset time period and the information on the determined sales volume change over time (ie, Figure 2 Step 204) shown may also include: among the products and brand logos involved in the third sample set, merging brand logos of brands whose product quantity is less than or equal to a preset product quantity.

[0134] It should be noted that the brand logo can be merged by referring to Figure 2 The specific description of step 203 is omitted here.

[0135] In this embodiment, through the above steps 320 to 324, the second sub-time period having a time span greater than the first sub-time period is used as the time unit, the order data is integrated and a mixed effect model is established to fit the price elasticity information. This can not only avoid the risk that the price elasticity information cannot be fitted due to the first order data having the first sub-time period as the time unit, thereby increasing the probability of successful fitting of the price elasticity information, but also increase the amount of price elasticity information in the price elasticity information set used for screening, thereby improving the accuracy of the price elasticity information.

[0136] Continue to refer Figure 4 , shows a process 400 of another implementation of the method for generating information in this embodiment. Figure 4 As shown, based on the first order data within the first preset time period and the information on the determined sales volume changing over time, the price elasticity level information of the target product is generated (ie, Figure 2 Step 204) may include:

[0137] Step 401, extracting the value of at least one attribute of the target product from the first order data within a first preset time period (e.g., two years). Here, the extracted attributes may include, but are not limited to: at least one of: transaction price (or the value after taking the logarithm of the transaction price), sales volume (or the value after taking the logarithm of the sales volume), the number of orders within the first sub-time period (e.g., the number of daily orders), and the number of views.

[0138] Step 402, for each of the at least one attribute mentioned above, extract the value corresponding to the attribute from the first order data of the target product, and determine the quartile (in statistics, all values ​​are arranged from small to large and divided into four equal parts, and the values ​​at the three dividing points are called quartiles) and / or the standard deviation of the extracted value as the attribute value of the attribute.

[0139] Compared with the mean, using quartiles as attribute values ​​can better reflect the changing trend of the data, thereby improving the accuracy of price elasticity information.

[0140] Step 403: Input the attribute value of at least one attribute into a pre-trained random forest model to obtain price elasticity information of the target product. The random forest model is used to characterize the corresponding relationship between the attribute value and the price elasticity information.

[0141] Random forest refers to building a forest composed of many decision trees in a random way. There is no correlation between each decision tree in the random forest. Random forest can be used for classification (when the decision tree in the random forest is a classification decision tree) and regression (when the decision tree in the random forest is a regression decision tree).

[0142] In this embodiment, the random forest model can be a price elasticity prediction model including many regression decision trees, and its prediction value can be the weighted average of the price elasticity information on the leaf nodes. Regression decision tree is a machine learning algorithm that makes decisions (i.e., predictions) based on the structure of the tree. Generally, a regression decision tree includes a root node, multiple internal nodes (also called non-leaf nodes) and multiple leaf nodes. The regression decision tree divides the input space into different regions, each region has independent parameters, each node of the regression decision tree is associated with a region of the input space, and the internal nodes continue to divide the region into sub-regions under the sub-nodes (usually using the coordinate axis to split the region). The space is thus subdivided into non-overlapping regions, and a one-to-one correspondence is formed between the leaf nodes and the input regions. When branching, the regression decision tree exhaustively enumerates each threshold of each feature (e.g., at least one of the above attributes) to find the best split point, and its measurement standard (i.e., loss function) can be the residual sum of squares, and the most reliable branching basis can be found by minimizing the residual sum of squares.

[0143] The benefit of using the random forest model is that it uses the product's own attribute values ​​as input, so that the output price elasticity information can fully reflect the characteristics of the product itself.

[0144] Optionally, the above-mentioned random forest model can be trained through the following steps: obtaining multiple training samples, each training sample including attribute values ​​of a product having the same category identification as the target product and price elasticity information predetermined based on the attribute values ​​of the product; taking the attribute value of each training sample as input and the price elasticity information of each training sample as output, and training to obtain the above-mentioned random forest model.

[0145] Optionally, the above random forest model can be trained by the following steps: Figure 3 The implementation method shown determines the price elasticity information of multiple products having the same category identification as the target product; for each of the above multiple products, the attribute value of at least one attribute of the product is determined by the method shown in steps 401 and 402; the attribute value of at least one attribute of each product is used as input, and the price elasticity information of each product is used as output, and the above random forest model is trained.

[0146] By Figure 3 The implementation shown is similar to Figure 4 The implementation method shown in the figure is combined, that is, first using the mixed effect model to obtain training samples to train the random forest model, and then using the trained random forest model to fit the price elasticity of the product. The advantages are: on the one hand, it is very convenient to obtain training samples; on the other hand, it improves the fitting success rate of the price elasticity information and solves the problem that the mixed effect model cannot fit the price elasticity information of individual products or the fitting value does not meet the screening conditions.

[0147] Step 404: Based on the relationship between the obtained price elasticity information and the price elasticity information threshold, price elasticity level information of the target product is generated. The process of generating price elasticity level information can be referred to Figure 2 The specific description of step 204 and Figure 3 The specific description of step 330 is omitted here.

[0148] The method provided by the above-mentioned embodiment of the present application obtains the order data of the target product and at least one product belonging to the same category as the target product within a preset time period, and then merges the order data of the same product in the same sub-time period into the first order data, and then determines the sales volume change information of the product with the same brand logo as the target product over time based on the order data within the preset time period, and finally generates the price elasticity level information of the target product based on the first order data within the preset time period and the determined sales volume change information over time, which can avoid outliers, strong influence points and the influence of time on sales, thereby improving the accuracy of the product price elasticity level information.

[0149] Further references Figure 5 , which shows a process 500 of another embodiment of a method for generating information. The process 500 of the method for generating information includes the following steps:

[0150] Step 501: Obtain order data of a target product and at least one product belonging to the same category as the target product within a first preset time period, where the first preset time period includes at least one first sub-time period.

[0151] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in the figure can obtain the order data of the target product and at least one product belonging to the same category as the target product (for example, having the same category identification) within a first preset time period (for example, two years) from the database through a wired connection or a wireless connection. The first preset time period includes at least one first sub-time period (for example, the first sub-time period is one day, one week, one month, etc.). The database stores the order data of the products, and the order data may include data information related to the products and orders, including but not limited to: product identification, category identification, brand identification, order time, page price, transaction price, sales volume, inventory, store identification, etc. For example, when the target product is a mobile phone of model XX, the above-mentioned execution entity can obtain the order data of the target product and at least one product with the category identification of "mobile phone" within two years. Here, the database can be deployed on a server, for example, it can be deployed on Figure 1 It can be deployed on the server shown, or on other servers.

[0152] Step 502: merge the order data of the same product in the same first sub-time period into first order data.

[0153] In this embodiment, the execution subject of the method for generating information (eg Figure 1The server shown in FIG. 5 can use the first sub-time period as the time unit to merge the order data obtained in step 501 according to the product (for example, summing the sales volume of the same product in the same sub-time period, performing weighted average on the transaction price of the same product in the same sub-time period, and other data conversion processing). That is, the order data of the same product in the same first sub-time period is merged into the first order data (for example, the order data of product A in the same day is merged into the first order data). By merging the order data, the amount of order data is effectively reduced, the speed of generating information is improved, and the price elasticity analysis is performed using the first sub-time period (day) as the time unit, which can better reflect the price elasticity of the product.

[0154] Step 503: Based on the order data in the first preset time period, determine the information on the change in sales volume over time of the product with the same brand logo as the target product.

[0155] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in FIG. 1 may first determine products with the same brand identity as the target product. Then, based on the order data of the determined products (including the target product), determine information on changes in sales volume over time for products with the same brand identity as the target product. The process of determining information on changes in sales volume over time may refer to Figure 2 The specific description of step 203 is omitted here.

[0156] Step 504: Generate price elasticity level information of the target product based on the first order data within the first preset time period and the determined information on sales volume changes over time.

[0157] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in FIG. 5 can determine the price elasticity information of the target product through a price elasticity model (e.g., a mixed effect model, a random forest model, etc.) based on the first order data generated in step 502 and the information on the change in sales volume over time determined in step 503. Then, according to the relationship between the price elasticity information and the price elasticity information threshold, the price elasticity level information of the target product is generated. The process of generating the price elasticity level information can refer to Figure 2 The specific description of step 204 and Figure 3 The specific description of step 330 is omitted here.

[0158] Step 505, obtaining traffic data of the target store in a second preset time period, where the traffic data includes browsing records of the target store.

[0159] In this embodiment, the execution subject of the method for generating information (for example, Figure 1The server shown in the figure) can obtain the traffic data of the target store in the second preset time period (for example, one month) from the database through a wired connection or a wireless connection. The traffic data includes the browsing records of the target store, for example, the time of browsing the target store, the products browsed, the browser's identification, the browsing time, etc. Here, the target store can be a physical store, that is, a commercial organization with a relatively fixed location established on the basis of certain hardware facilities (such as business premises), such as supermarkets, shopping malls, retail stores, etc.; the target store can also be a virtual store (also called an online store, a typical organizational form of e-commerce, a store established on the Internet), such as a website, APP (application), etc.

[0160] Step 506: Based on the above traffic data, determine the number of times the target store is browsed and the number of times the target product is browsed.

[0161] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in FIG. 5 may analyze (e.g., summarize, count, etc.) the traffic data obtained in step 505 to determine the number of times the target store is browsed and the number of times the target product is browsed. Here, the number of times browsed may be accumulated by the number of times the viewer repeatedly browses (e.g., the user repeatedly browses the same product page), or may not be accumulated. Those skilled in the art may set this according to the needs of the actual application scenario.

[0162] Step 507: Generate traffic level information of the target product in the target store based on the number of times the target store is browsed and the number of times the target product is browsed.

[0163] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in the figure can determine the browsing count threshold (for example, calculate the average) based on the number of times the target store is browsed and the number of products in the target store. Then, based on the relationship between the number of times the target product is browsed and the browsing count threshold, the traffic level information of the target product in the target store is generated. The browsing count threshold is used to characterize the average attention of the products in the target store, and the traffic level information is used to characterize the attention of the products. For example, if the traffic count threshold is 300 and product A is browsed 10 times, the traffic level information of product A is "low traffic", that is, the attention of product A is low; and if product B is browsed 1000 times, the traffic level information of product B is "high traffic", that is, the attention of product B is high.

[0164] In some optional implementations of this embodiment, step 507 may include:

[0165] In the first step, the execution entity can determine the number of valid products in the target store. The valid products here refer to products that have sales and / or browsing records in the second preset time period. Specifically, the execution entity can determine the number of products in the target store whose sales are greater than the preset sales (for example, 0, 1, etc.) based on the first order data in the second preset time period, and / or determine the number of products in the target store whose browsing times are greater than the preset times (for example, 0, 1, etc.) based on the traffic data, so as to determine the number of valid products in the target store.

[0166] In the second step, the execution entity may determine the browsing threshold based on the number of times the target store is browsed and the number of valid products. For example, the ratio of the number of times the target store is browsed to the number of valid products. By using the number of valid products to determine the browsing threshold, the influence of products with no sales and browsing records is eliminated, which can more accurately and objectively reflect the average attention of products in the target store and improve the accuracy of the browsing threshold.

[0167] In the third step, the execution subject may generate the traffic level information of the target product in the target store based on the relationship between the number of times the target product is browsed and the browsing number threshold. For example, if the number of times the target product is browsed is greater than the browsing number threshold, the traffic level information of the target product is "high traffic"; if the number of times the target product is browsed is less than or equal to the browsing number threshold, the traffic level information of the target product is "low traffic".

[0168] In practice, traffic levels of different stores may vary greatly. In more popular stores, each product is viewed a lot, while in niche stores (i.e., stores popular with a small number of people), even the products with the best sales / highest popularity are not viewed a lot. Therefore, traffic level information can be determined differently according to the specific situation of the store.

[0169] In some optional implementations of this embodiment, the traffic level information of the target product can be determined by the following steps: in response to the above-mentioned number of valid products being greater than the product quantity threshold (e.g., 10) and the browsing times threshold being greater than a first value (e.g., 300), the traffic level threshold of the target product is determined according to the relationship between the ratio of the number of times the target product is browsed to the browsing times threshold and a first ratio (e.g., 1.2). For example, when the ratio of the number of times the target product is browsed to the browsing times threshold is greater than the first ratio, the traffic level information of the target product is "high traffic"; when the ratio of the number of times the target product is browsed to the browsing times threshold is less than or equal to the first ratio, the traffic level information of the target product is "low traffic".

[0170] In some optional implementations of this embodiment, the traffic level information of the target product can also be determined by the following steps: in response to the above-mentioned number of valid products being greater than the product quantity threshold (e.g., 10) and the browsing times threshold being less than or equal to a first value (e.g., 300), the traffic level threshold of the target product is determined according to the relationship between the ratio of the number of times the target product is browsed to the browsing times threshold and a second ratio (e.g., 1.5). For example, when the ratio of the number of times the target product is browsed to the browsing times threshold is greater than the second ratio, the traffic level information of the target product is "high traffic"; when the ratio of the number of times the target product is browsed to the browsing times threshold is less than or equal to the second ratio, the traffic level information of the target product is "low traffic".

[0171] According to the popularity of the store, the store can be divided into multiple types, for example, niche stores, medium stores, mainstream stores, etc. Differentially determining the level of traffic information through the first ratio and / or the second ratio can more objectively reflect the attention of the products in the store.

[0172] Step 508: determining price adjustment information of the target product based on the price elasticity level information of the target product and the traffic level information of the target product in the target store.

[0173] In this embodiment, the execution subject of the method for generating information (for example, Figure 1 The server shown in FIG. 5 may determine the price adjustment information of the target product based on the price elasticity level information generated in step 504 and the traffic level information generated in step 507, according to the correspondence between the price elasticity level information, the traffic level information and the price adjustment information. As an example, Table 1 shows a correspondence table of the price elasticity level information, the traffic level information and the price adjustment information.

[0174] Table 1 Correspondence between price elasticity level, traffic level and price adjustment information

[0175]

[0176] It should be noted that those skilled in the art can set the generation frequency of the price elasticity level information and the flow level information considering their own characteristics, computing power and other factors. For example, the price elasticity level information of the product is updated weekly, and the flow level information of the product is updated daily.

[0177] Continue to see Figure 6 , Figure 6 FIG. 6 is a schematic diagram of an application scenario 600 of the method for generating information according to this embodiment. Figure 6 In the application scenario, a user (for example, the manager of the target store "XX store") can select the category logo and brand logo in the backend product management interface through the client; then, the backend server (for example, Figure 1 The server) can determine the qualified products in the target store (for example, products A, B, C, D, etc.) according to the category identification and brand identification selected by the user, and obtain the order data of the above products within two years and the traffic data of the target store within one month from the database; then, the above-mentioned background server respectively determines the price elasticity level information and traffic level information of each of the above products, and generates corresponding price adjustment information for each product; finally, the above-mentioned client displays the management page of the product with the selected category identification and brand identification based on the price adjustment information returned by the above-mentioned background server, for example, the above-mentioned products are divided into four types: traffic maintenance type (the number of products is 10), profit growth type (the number of products is 30), traffic growth type (the number of products is 25) and profit maintenance type (the number of products is 10), and specifically displays the corresponding price adjustment information of each product (it can also display the product's price, sales volume, number of views, price elasticity level information, browsing level information, etc. at the same time).

[0178] from Figure 5 It can be seen that Figure 2 Compared with the corresponding embodiment, the process 500 of the method for generating information in this embodiment highlights the steps of determining the traffic level information of the target product in the target store and the step of generating price adjustment information. Therefore, the scheme described in this embodiment can combine the price elasticity level information of the target product with the traffic level information of the target product in the target store, so as to provide the price adjustment information of the target product for the target store from multiple dimensions, and can provide accurate product price adjustment information.

[0179] Further references Figure 7 As an implementation of the methods shown in the above figures, the present application provides an embodiment of a device for generating information, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied in a server.

[0180] like Figure 7As shown, the device 700 for generating information in this embodiment includes: an order acquisition unit 701, an order fusion unit 702, a trend determination unit 703 and an elasticity generation unit 704. The order acquisition unit 701 is configured to acquire order data of a target product and at least one product belonging to the same category as the target product within a first preset time period, and the first preset time period includes at least one first sub-time period; the order fusion unit 702 is configured to fuse the order data of the same product within the same first sub-time period into the first order data; the trend determination unit 703 is configured to determine the information of the sales volume change over time of the product with the same brand logo as the target product based on the order data within the first preset time period; and the elasticity generation unit 704 is configured to generate the price elasticity level information of the target product based on the first order data within the first preset time period and the determined information of the sales volume change over time.

[0181] In this embodiment, the order acquisition unit 701 of the device 700 for generating information can obtain the order data of the target product and at least one product belonging to the same category as the target product (for example, having the same category identification) within a first preset time period (for example, two years) from the database through a wired connection or a wireless connection. The first preset time period includes at least one first sub-time period (for example, the first sub-time period is one day, one week, or one month, etc.). The database stores the order data of the product, and the order data may include data information related to the product and the order, including but not limited to: product identification, category identification, brand identification, order time, page price, transaction price, sales volume, inventory, store identification, etc. For example, when the target product is a mobile phone of model XX, the execution entity can obtain the order data of the target product and at least one product with a category identification of "mobile phone" within two years. Here, the database can be deployed on a server, for example, it can be deployed on Figure 1 It can be deployed on the server shown, or on other servers.

[0182] In this embodiment, the order fusion unit 702 can fuse the order data acquired by the order acquisition unit 701 according to the product with the first sub-time period as the time unit (for example, summing the sales volume of the same product in the same sub-time period, weighted averaging the transaction price of the same product in the same sub-time period, and other data conversion processing). That is, the order data of the same product in the same first sub-time period is fused into the first order data (for example, the order data of product A in the same day is fused into the first order data). By fusing the order data, the amount of order data is effectively reduced, the speed of generating information is improved, and the price elasticity analysis with the first sub-time period (day) as the time unit can better reflect the price elasticity of the product.

[0183] In some optional implementations of this embodiment, the order fusion unit 702 may include a filtering module, a regression module, and a fusion module. The filtering module is configured to filter the order data within the first preset time period, and delete the order data whose transaction price is less than the preset price threshold (for example, a relative value such as 30%, 50%, 60% of the page price, or an absolute value such as 1 yuan); the regression module is configured to perform a robust regression on the sales volume and the transaction price based on the filtered order data, and delete the order data whose residual is greater than the residual threshold (for example, 2.5 times the residual standard deviation); and the fusion module is configured to fuse the order data of the same product within the first sub-time period based on the order data after the robust regression (for example, by summing the sales volume, weighted averaging the transaction price, etc.) for each first sub-time period in at least one first sub-time period, and generate the first order data.

[0184] In this embodiment, the trend determination unit 703 may first determine products with the same brand logo as the target product, and then determine information on changes in sales volume of products with the same brand logo as the target product over time based on order data of the determined products (including the target product).

[0185] In some optional implementations of this embodiment, the first preset time period includes at least one second sub-time period, and the second sub-time period includes at least one first sub-time period (for example, the first preset time period is two years, the second sub-time period is one week, and the first sub-time period is one day), and the above-mentioned trend determination unit 703 may include a fusion module and a determination module. The fusion module is configured to fuse the order data of the product with the same brand logo as the target product in the second sub-time period into the second order data for each second sub-time period in at least one second sub-time period; the determination module is configured to determine the information on the change of the sales volume of the product with the same brand logo as the target product over time (with the second sub-time period as the time unit) based on at least one second sub-time period and the second order data in the corresponding second sub-time period.

[0186] In some optional implementations of this embodiment, the trend determination unit 703 may further include a brand merging module, wherein the brand merging module is configured to merge brand logos of brands whose product quantity is less than or equal to a preset product quantity among brand logos of products having the same category logo as the target product.

[0187] In this embodiment, the elasticity generating unit 704 may determine the price elasticity information of the target product through a price elasticity model (e.g., a mixed effect model, a random forest model, etc.) based on the first order data generated by the order fusion unit 702 and the information on the change in sales volume over time determined by the trend determining unit 703. Then, the price elasticity level information of the target product is generated according to the relationship between the price elasticity information and the price elasticity information threshold.

[0188] In some optional implementations of this embodiment, the elasticity generation unit 704 may include a first selection module, a first determination module, a second selection module, a second determination module and an elasticity generation module. Among them, the first selection module is configured to select the first order data of products whose price change times and sales change times exceed a preset change times threshold (for example, 8 times, a sufficient number of change points can improve the fitting success rate) from the first order data within the first preset time period, and generate a first sample set; the first determination module is configured to determine the first price elasticity information of the target product and the first price elasticity information of the category represented by the category identifier of the target product based on the first sample set and the determined information on sales changes over time, and generate a price elasticity information set; the second selection module is configured to select the first order data of products with the same brand identifier as the target product from the first sample set, and generate a second sample set; the second determination module is configured to determine the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand identifier of the target product based on the second sample set and the determined information on sales changes over time, and add the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand identifier of the target product to the price elasticity information set; and the elasticity generation module is configured to generate price elasticity level information of the target product based on the price elasticity information set.

[0189] In some optional implementations of this embodiment, the elasticity generation module may include an elasticity screening submodule, an elasticity determination submodule, and an elasticity generation submodule. The elasticity screening submodule is configured to determine at least one price elasticity information whose coefficient of determination is greater than a coefficient of determination threshold (e.g., 0.3) from the price elasticity information set; the elasticity determination submodule is configured to determine the price elasticity information whose significance test statistic (e.g., t value in t test) is greater than a significance test statistic threshold (e.g., -1.64, -1.96, etc.) and has the highest significance test statistic as the price elasticity information of the target product; the elasticity generation submodule is configured to generate price elasticity level information (e.g., "low price elasticity" and "high price elasticity") of the target product based on the size relationship between the determined price elasticity information and the price elasticity information threshold (e.g., obtained by median, mean, mode, etc.).

[0190] In some optional implementations of this embodiment, the first preset time period includes at least one second sub-time period, the second sub-time period includes at least one first sub-time period (for example, the first preset time period is two years, the second sub-time period is one week, and the first sub-time period is one day), and the elasticity generation unit 704 also includes a fusion module, a third selection module, a third determination module, a fourth selection module, and a fourth determination module. Among them, the fusion module is configured to determine, for each second sub-time period in at least one second sub-time period, the first order data of the product whose number of first order data in the second sub-time period is greater than a preset number (for example, 3), and fuse the first order data of the same product determined in the second sub-time period into the third order data; the third selection module is configured to select the third order data of the product whose price change times and sales change times exceed the preset change times threshold (for example, 8 times) from the third order data to generate a third sample set; the third determination module is configured to determine the third price elasticity information of the target product and the category identification table of the target product based on the third sample set and the determined information on sales changes over time. The fourth selection module is configured to select the third order data of the product with the same brand logo as the target product from the third sample set to generate a fourth sample set; the fourth determination module is configured to determine the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand logo of the target product based on the fourth sample set and the information on the determined sales volume changing over time, and add the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand logo of the target product to the price elasticity information set.

[0191] In some optional implementations of this embodiment, the elasticity generation unit 704 further includes a brand merging module, wherein the brand merging module is configured to merge brand logos of brands whose product quantity is less than or equal to a preset product quantity among the products and brand logos involved in the first sample set.

[0192] In some optional implementations of this embodiment, the brand merging module is further configured to merge brand logos of brands whose number of products is less than or equal to a preset number of products among the products and brand logos involved in the third sample set.

[0193] In some optional implementations of this embodiment, the elasticity generation unit 704 includes an attribute extraction module, an attribute value determination module, an elasticity determination module, and an elasticity generation module. Among them, the attribute extraction module is configured to extract the value of at least one attribute of the target product from the first order data within the first preset time period; the attribute value determination module is configured to determine the quartile and / or standard deviation of the value extracted for each attribute of the at least one attribute as the attribute value of the attribute; the elasticity determination module is configured to input the attribute value of at least one attribute into a pre-trained random forest model (for example, a price elasticity prediction model containing many regression decision trees) to obtain the price elasticity information of the target product, and the random forest model is used to characterize the corresponding relationship between the attribute value and the price elasticity information; the elasticity generation module is configured to generate the price elasticity level information of the target product based on the size relationship between the price elasticity information of the target product and the price elasticity information threshold.

[0194] In some optional implementations of the present embodiment, the above-mentioned random forest model can be trained through the following steps: obtaining multiple training samples, each training sample including attribute values ​​of a product having the same category identification as the target product and price elasticity information predetermined based on the attribute values ​​of the product; taking the attribute value of each training sample as input and the price elasticity information of each training sample as output, and training to obtain the above-mentioned random forest model.

[0195] In some optional implementations of this embodiment, the random forest model can be trained by the following steps: Figure 3 Specific description of ) determine the price elasticity information of multiple products with the same category identifier as the target product; for each of the above multiple products, determine the attribute value of at least one attribute of the product (refer to Figure 4 Detailed description of step 401 and step 402 in ); taking the attribute value of at least one attribute of each product as input, taking the price elasticity information of each product as output, and training to obtain the above-mentioned random forest model.

[0196] In some optional implementations of this embodiment, the device 700 for generating information further includes a flow acquisition unit, a frequency determination unit, a flow generation unit, and a price adjustment unit. The flow acquisition unit is configured to acquire flow data of the target store within a second preset time period, and the flow data includes browsing records of the target store; the frequency determination unit is configured to determine the number of times the target store is browsed and the number of times the target product is browsed based on the flow data; the flow generation unit is configured to generate flow level information of the target product in the target store based on the number of times the target store is browsed and the number of times the target product is browsed; and the price adjustment unit is configured to generate price adjustment information of the target product in the target store based on the price elasticity level information of the target product and the flow level information of the target product in the target store.

[0197] In this embodiment, the above-mentioned traffic acquisition unit of the device 700 for generating information can obtain the traffic data of the target store in the second preset time period (for example, one month) from the database through a wired connection or a wireless connection, wherein the traffic data includes the browsing record of the target store, for example, the time of browsing the target store, the products browsed, the identification of the browser, the browsing time, etc. Here, the target store can be a physical store, that is, a commercial organization with a relatively fixed location established on the basis of certain hardware facilities (such as business premises), such as a supermarket, a shopping mall, a retail store, etc.; the target store can also be a virtual store (also called an online store, a typical organizational form of e-commerce, a store established on the Internet), such as a website, an APP (application), etc.

[0198] In this embodiment, the number determination unit may analyze (e.g., summarize, count, etc.) the traffic data acquired by the traffic acquisition unit to determine the number of times the target store has been browsed and the number of times the target product has been browsed. Here, the number of times browsed may be accumulated by the number of times the viewer has repeatedly browsed (e.g., the user has repeatedly browsed the same product page), or may not be accumulated. Those skilled in the art may set this according to the needs of the actual application scenario.

[0199] In this embodiment, the above-mentioned traffic generation unit can determine the browsing times threshold (for example, calculate the average) based on the number of times the target store is browsed and the number of products in the target store. Then, based on the relationship between the number of times the target product is browsed and the browsing times threshold, the traffic level information of the target product in the target store is generated. The browsing times threshold is used to characterize the average attention of the products in the target store, and the traffic level information is used to characterize the attention of the products. For example, if the traffic times threshold is 300 and product A is browsed 10 times, the traffic level information of product A is "low traffic", that is, the attention of product A is low; and if product B is browsed 1000 times, the traffic level information of product B is "high traffic", that is, the attention of product B is high.

[0200] In some optional implementations of this embodiment, the above-mentioned traffic generation unit may include a product determination module, a threshold determination module and a traffic generation module. The product determination module is configured to determine the number of products (i.e., the number of valid products) in the target store whose sales volume is greater than a preset sales volume (e.g., 0, 1, etc.) and / or the number of times browsed is greater than a preset number of times (e.g., 0, 1, etc.); the threshold determination module is configured to determine the number of browse times threshold based on the number of times the target store is browsed and the determined number of products, for example, the ratio of the number of times the target store is browsed to the number of valid products; the traffic generation module is configured to generate traffic level information of the target product in the target store based on the relationship between the number of times the target product is browsed and the number of browse times threshold.

[0201] In some optional implementations of this embodiment, the traffic generation module is specifically used to: in response to the number of valid products being greater than the product quantity threshold (e.g., 10) and the browsing times threshold being greater than a first value (e.g., 300), determine the traffic level threshold of the target product according to the relationship between the ratio of the number of times the target product is browsed to the browsing times threshold and a first ratio (e.g., 1.2). For example, when the ratio of the number of times the target product is browsed to the browsing times threshold is greater than the first ratio, the traffic level information of the target product is "high traffic"; when the ratio of the number of times the target product is browsed to the browsing times threshold is less than or equal to the first ratio, the traffic level information of the target product is "low traffic".

[0202] In some optional implementations of this embodiment, the traffic generation module is further used to: in response to the number of valid products being greater than the product quantity threshold (e.g., 10) and the browsing times threshold being less than or equal to a first value (e.g., 300), determine the traffic level threshold of the target product according to the relationship between the ratio of the number of times the target product is browsed to the browsing times threshold and a second ratio (e.g., 1.5). For example, when the ratio of the number of times the target product is browsed to the browsing times threshold is greater than the second ratio, the traffic level information of the target product is "high traffic"; when the ratio of the number of times the target product is browsed to the browsing times threshold is less than or equal to the second ratio, the traffic level information of the target product is "low traffic".

[0203] In this embodiment, the price adjustment unit can determine the price adjustment information of the target product based on the price elasticity level information generated by the elasticity generation unit 704 and the traffic level information generated by the traffic generation unit, according to the correspondence between the price elasticity level information, the traffic level information and the price adjustment information (for example, as shown in Table 1).

[0204] The device provided by the above-mentioned embodiment of the present application obtains the order data of the target product and at least one product belonging to the same category as the target product within a preset time period, and then merges the order data of the same product in the same sub-time period into the first order data, and then determines the sales volume change information of the product with the same brand logo as the target product over time based on the order data within the preset time period, and finally generates the price elasticity level information of the target product based on the first order data within the preset time period and the determined sales volume change information over time, which can avoid outliers, strong influence points and the influence of time on sales, thereby improving the accuracy of the product price elasticity level information.

[0205] Reference below Figure 8 , which shows a schematic diagram of the structure of a computer system 800 suitable for implementing a server of an embodiment of the present application. Figure 8 The server shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0206] like Figure 8 As shown, the computer system 800 includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage part 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the system 800 are also stored. The CPU 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0207] The following components are connected to the I / O interface 805: an input section 806 including a keyboard, a mouse, etc.; an output section 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card, a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as needed. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as needed, so that a computer program read therefrom is installed into the storage section 808 as needed.

[0208] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part 809, and / or installed from a removable medium 811. When the computer program is executed by a central processing unit (CPU) 801, the above-mentioned functions defined in the method of the present application are executed.

[0209] It should be noted that the computer-readable medium described in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0210] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0211] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0212] The units involved in the embodiments described in the present application may be implemented by software or by hardware. The described units may also be set in a processor, for example, it may be described as: a processor includes an order acquisition unit, an order fusion unit, a trend determination unit, and an elasticity generation unit. Among them, the names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the order acquisition unit may also be described as "a unit for acquiring order data of a target product and at least one product belonging to the same category as the target product within a first preset time period".

[0213] As another aspect, the present application also provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by the device, the device: obtains order data of a target product and at least one product belonging to the same category as the target product within a first preset time period, and the first preset time period includes at least one first sub-time period; merges the order data of the same product within the same first sub-time period into the first order data; based on the order data within the first preset time period, determines the information on the sales volume changes over time of the product with the same brand logo as the target product; based on the first order data within the first preset time period and the determined information on the sales volume changes over time, generates the price elasticity level information of the target product.

[0214] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A method for generating information, comprising: Acquire order data of a target product and at least one product belonging to the same category as the target product within a first preset time period, where the first preset time period includes at least one first sub-time period; Merge order data of the same product in the same first sub-time period into first order data; Determining information on changes in sales volume of products having the same brand identity as the target product over time based on the order data within the first preset time period; Generate price elasticity level information of the target product based on the first order data within the first preset time period and the determined information on sales volume changes over time; Acquiring traffic data of a target store within a second preset time period, wherein the traffic data includes browsing records of the target store; Based on the traffic data, determine the number of times the target store is browsed and the number of times the target product is browsed; Based on the number of times the target store is browsed and the number of times the target product is browsed, generating traffic level information of the target product in the target store; Price adjustment information of the target product in the target store is generated based on the price elasticity level information of the target product and the traffic level information of the target product in the target store.

2. The method according to claim 1, wherein: The step of fusing the order data of the same product in the same first sub-time period into the first order data includes: Filtering the order data within the first preset time period, and deleting the order data with a transaction price less than a preset price threshold; Based on the filtered order data, robust regression is performed on sales volume and transaction price, and order data with residuals greater than the residual threshold is deleted; For each first sub-time period in the at least one first sub-time period, based on the order data after robust regression, the order data of the same product in the first sub-time period are fused to generate first order data.

3. The method according to claim 1, wherein: The first preset time period includes at least one second sub-time period, the second sub-time period includes at least one first sub-time period, and The determining, based on the order data within the first preset time period, information on changes in sales volume of a product having the same brand identification as the target product over time includes: For each second sub-time period in the at least one second sub-time period, merging order data of products having the same brand identification as the target product in the second sub-time period into second order data; Based on the at least one second sub-time period and the second order data in the corresponding second sub-time period, information on changes in sales volume over time of a product having the same brand identification as the target product is determined.

4. The method according to claim 3, wherein: Before fusing the order data of the product with the same brand identifier as the target product in the second sub-time period into the second order data, determining the information on the change in sales volume of the product with the same brand identifier as the target product over time based on the order data in the first preset time period further includes: Among the brand identifications of products having the same category identification as the target product, brand identifications of brands including a number of products less than or equal to a preset number of products are merged.

5. The method according to claim 1, wherein: The step of generating the price elasticity level information of the target product based on the first order data within the first preset time period and the determined information on sales volume changing over time includes: Selecting first order data of products whose price change times and sales volume change times exceed a preset change times threshold from the first order data within the first preset time period to generate a first sample set; Based on the first sample set and the determined information on sales volume changes over time, determine the first price elasticity information of the target product and the first price elasticity information of the category represented by the category identifier of the target product, and generate a price elasticity information set; Selecting first order data of a product having the same brand logo as the target product from the first sample set to generate a second sample set; Determine, based on the second sample set and the determined information about sales volume changes over time, second price elasticity information of the target product and first price elasticity information of the brand represented by the brand logo of the target product, and add the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand logo of the target product to a price elasticity information set; Based on the price elasticity information set, price elasticity level information of the target product is generated.

6. The method according to claim 5, wherein: The first preset time period includes at least one second sub-time period, the second sub-time period includes at least one first sub-time period, and Before generating the price elasticity level information of the target product, the generating the price elasticity level information of the target product based on the first order data within the first preset time period and the determined information on sales volume changing over time further includes: For each second sub-time period in the at least one second sub-time period, determine first order data of a product whose number of first order data in the second sub-time period is greater than a preset number, and merge the first order data of the same product determined in the second sub-time period into third order data; Selecting the third order data of products whose price change times and sales volume change times exceed the preset change times threshold from the third order data to generate a third sample set; Determine, based on the third sample set and the determined information about sales volume changes over time, third price elasticity information of the target product and second price elasticity information of the category represented by the category identifier of the target product, and add the third price elasticity information of the target product and the second price elasticity information of the category represented by the category identifier of the target product to a price elasticity information set; Selecting third order data of products having the same brand logo as the target product from the third sample set to generate a fourth sample set; Based on the fourth sample set and the determined information on sales volume changes over time, the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand logo of the target product are determined, and the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand logo of the target product are added to the price elasticity information set.

7. The method according to claim 5, wherein: Before generating the second sample set, generating the price elasticity level information of the target product based on the first order data within the first preset time period and the determined information on sales volume changes over time also includes: Among the products and brand logos involved in the first sample set, brand logos of brands whose number of products is less than or equal to a preset number of products are merged.

8. The method according to claim 5, wherein: The step of generating the price elasticity level information of the target product based on the price elasticity information set includes: Determine at least one price elasticity information having a determination coefficient greater than a determination coefficient threshold from the price elasticity information set; Determine the price elasticity information whose significance test statistic is greater than the significance test statistic threshold and has the highest significance test statistic among the at least one price elasticity information as the price elasticity information of the target product; Based on the determined relationship between the price elasticity information and the price elasticity information threshold, the price elasticity level information of the target product is generated.

9. The method according to claim 1, wherein: The step of generating the price elasticity level information of the target product based on the first order data within the first preset time period and the determined information on sales volume changing over time includes: Extracting a value of at least one attribute of the target product from the first order data within the first preset time period; For each of the at least one attribute, determining a quartile and / or a standard deviation of the values ​​extracted for the attribute as an attribute value of the attribute; Inputting the attribute value of the at least one attribute into a pre-trained random forest model to obtain the price elasticity information of the target product, wherein the random forest model is used to characterize the corresponding relationship between the attribute value and the price elasticity information; Price elasticity level information of the target product is generated based on the magnitude relationship between the price elasticity information of the target product and the price elasticity information threshold.

10. The method according to claim 9, wherein: The random forest model is trained by the following steps: Acquire multiple training samples, each training sample including attribute values ​​of a product having the same category identifier as the target product and price elasticity information predetermined based on the attribute values ​​of the product; The attribute value of each training sample is taken as input, and the price elasticity information of each training sample is taken as output, and the random forest model is obtained through training.

11. The method according to claim 1, wherein: The generating, based on the number of times the target store is browsed and the number of times the target product is browsed, traffic level information of the target product in the target store includes: Determine the number of products in the target store whose sales volume is greater than a preset sales volume and / or whose browsing times are greater than a preset number; Determining a browsing count threshold based on the number of times the target store is browsed and the determined number of products; Based on the relationship between the number of times the target product is browsed and the browsing number threshold, traffic level information of the target product in the target store is generated.

12. A device for generating information, comprising: An order acquisition unit, configured to acquire order data of a target product and at least one product belonging to the same category as the target product within a first preset time period, wherein the first preset time period includes at least one first sub-time period; An order fusion unit, configured to fuse order data of the same product in the same first sub-time period into first order data; a trend determination unit configured to determine information on changes in sales volume of a product having the same brand identity as the target product over time based on the order data within the first preset time period; an elasticity generating unit, configured to generate price elasticity level information of the target product based on the first order data within the first preset time period and the determined information on sales volume changes over time; A traffic acquisition unit configured to acquire traffic data of a target store within a second preset time period, wherein the traffic data includes browsing records of the target store; A number determination unit, configured to determine the number of times the target store is browsed and the number of times the target product is browsed based on the traffic data; a traffic generation unit configured to generate traffic level information of the target product in the target store based on the number of times the target store is browsed and the number of times the target product is browsed; The price adjustment unit is configured to generate price adjustment information of the target product in the target store based on the price elasticity level information of the target product and the traffic level information of the target product in the target store.

13. The device according to claim 12, wherein: The order fusion unit comprises: A filtering module configured to filter the order data within the first preset time period and delete the order data with a transaction price less than a preset price threshold; A regression module is configured to perform robust regression on sales volume and transaction price based on the filtered order data, and delete order data with residuals greater than the residual threshold; The fusion module is configured to fuse the order data of the same product in each first sub-time period of the at least one first sub-time period based on the order data after robust regression to generate first order data.

14. The device according to claim 12, wherein: The first preset time period includes at least one second sub-time period, the second sub-time period includes at least one first sub-time period, and the trend determination unit includes: a fusion module configured to fuse, for each second sub-time period in the at least one second sub-time period, order data of products having the same brand identification as the target product in the second sub-time period into second order data; The determination module is configured to determine information on changes in sales volume over time of a product having the same brand identification as the target product based on the at least one second sub-time period and second order data in the corresponding second sub-time period.

15. The device according to claim 14, wherein: The trend determination unit also includes: The brand merging module is configured to merge the brand identifications of brands whose product quantity is less than or equal to a preset product quantity among the brand identifications of products having the same category identification as the target product.

16. The device according to claim 12, wherein: The elasticity generating unit comprises: A first selection module is configured to select first order data of products whose price change times and sales volume change times exceed a preset change times threshold from the first order data within the first preset time period, and generate a first sample set; A first determination module is configured to determine, based on the first sample set and the determined information about sales volume changes over time, the first price elasticity information of the target product and the first price elasticity information of the category represented by the category identifier of the target product, and generate a price elasticity information set; A second selection module is configured to select first order data of a product having the same brand logo as the target product from the first sample set to generate a second sample set; A second determination module is configured to determine, based on the second sample set and the determined information about sales volume changes over time, the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand logo of the target product, and add the second price elasticity information of the target product and the first price elasticity information of the brand represented by the brand logo of the target product to a price elasticity information set; The elasticity generation module is configured to generate price elasticity level information of the target product based on the price elasticity information set.

17. The device according to claim 16, wherein: The first preset time period includes at least one second sub-time period, the second sub-time period includes at least one first sub-time period, and the elasticity generating unit further includes: a fusion module configured to determine, for each second sub-time period in the at least one second sub-time period, first order data of a product whose number of first order data in the second sub-time period is greater than a preset number, and fuse the first order data of the same product determined in the second sub-time period into third order data; A third selection module is configured to select, from the third order data, the third order data of products whose price change times and sales volume change times exceed the preset change times threshold, to generate a third sample set; a third determination module configured to determine, based on the third sample set and the determined information on sales volume changes over time, the third price elasticity information of the target product and the second price elasticity information of the category represented by the category identifier of the target product, and add the third price elasticity information of the target product and the second price elasticity information of the category represented by the category identifier of the target product to the price elasticity information set; a fourth selection module configured to select third order data of a product having the same brand logo as the target product from the third sample set to generate a fourth sample set; A fourth determination module is configured to determine the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand logo of the target product based on the fourth sample set and the determined information on sales volume changes over time, and add the fourth price elasticity information of the target product and the second price elasticity information of the brand represented by the brand logo of the target product to the price elasticity information set.

18. The device according to claim 16, wherein: The elasticity generating unit further comprises: The brand merging module is configured to merge the brand logos of brands whose product quantity is less than or equal to a preset product quantity among the products and brand logos involved in the first sample set.

19. The device according to claim 16, wherein: The elasticity generation module also includes: an elasticity screening submodule, configured to determine at least one price elasticity information having a determination coefficient greater than a determination coefficient threshold from the price elasticity information set; an elasticity determination submodule, configured to determine the price elasticity information whose significance test statistic is greater than the significance test statistic threshold and has the highest significance test statistic among the at least one price elasticity information as the price elasticity information of the target product; The elasticity generation submodule is configured to generate price elasticity level information of the target product based on the size relationship between the determined price elasticity information and the price elasticity information threshold.

20. The device according to claim 12, wherein: The elasticity generating unit comprises: an attribute extraction module, configured to extract a value of at least one attribute of the target product from the first order data within the first preset time period; an attribute value determination module configured to determine, for each attribute of the at least one attribute, a quartile and / or a standard deviation of the value extracted for the attribute as the attribute value of the attribute; an elasticity determination module configured to input the attribute value of the at least one attribute into a pre-trained random forest model to obtain the price elasticity information of the target product, wherein the random forest model is used to characterize the corresponding relationship between the attribute value and the price elasticity information; The elasticity generating module is configured to generate the price elasticity level information of the target product based on the size relationship between the price elasticity information of the target product and the price elasticity information threshold.

21. The device according to claim 20, wherein: The random forest model is trained by the following steps: Acquire multiple training samples, each training sample including attribute values ​​of a product having the same category identifier as the target product and price elasticity information predetermined based on the attribute values ​​of the product; The attribute value of each training sample is taken as input, and the price elasticity information of each training sample is taken as output, and the random forest model is obtained through training.

22. The device according to claim 12, wherein: The traffic generation unit comprises: A product determination module, configured to determine the number of products in the target store whose sales volume is greater than a preset sales volume and / or whose browsing times are greater than a preset number; A threshold determination module, configured to determine a browsing count threshold based on the number of times the target store is browsed and the determined number of products; The traffic generation module is configured to generate traffic level information of the target product in the target store based on the relationship between the number of times the target product is browsed and the browsing number threshold.

23. A server, comprising: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 11.

24. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 11 is implemented.

Citation Information

Patent Citations

  • Discount billing terminal, discount billing system and discount billing method

    CN104376457A

  • Order price adjustment method and device

    CN106875221A

  • Method and device for calculating daily transaction prices of plastic raw materials

    CN107103500A

  • Automatic price adjustment method based on internet market and automatic price adjustment system thereof

    CN107292643A

  • Auxiliary pricing method and device which are implemented through computer and based on data analysis

    CN107292661A