Method, device, equipment and medium for determining item attributes

Abstract

This invention discloses a method, apparatus, device, and medium for determining item attributes. The method includes: retrieving a set of statistical transaction data containing statistical transaction data elements of items from a storage unit via a data acquisition module calling a data acquisition interface; for each statistical transaction data element, determining a statistical transaction data fluctuation parameter based on the statistical transaction data contained in the statistical transaction data element via a peak transaction day determination module, and determining the original peak transaction day corresponding to the statistical transaction data element based on the statistical transaction data fluctuation parameter; and determining the item attribute of the item based on the original peak transaction day corresponding to each statistical transaction data element via an item attribute determination module. The item attribute determination method provided by this invention determines the original peak transaction day by using statistical transaction data fluctuation parameters obtained from the statistical transaction data, thereby determining the item attribute of the item, improving the identification accuracy of seasonal items.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data processing, and in particular to a method, device, equipment and medium for determining an attribute of an item. Background Art

[0002] Supply chain management is the core of enterprise management, especially for the e-commerce industry. Replenishment is a very important link in supply chain management, and the amount of replenishment depends on the forecast of future sales of goods. Therefore, sales forecasting is a vital part of supply chain management. There are many sales patterns of goods, among which seasonal products are a special type of goods. Among them, seasonal products are goods whose sales time series show periodic changes every year. Being able to accurately identify seasonal products and the change cycle of seasonal products based on sales patterns, and adjust the forecast amount according to the change cycle, can effectively improve the accuracy of the forecast.

[0003] The existing seasonal product identification methods are mainly as follows: cluster the monthly sales, set a threshold h, calculate the ratio of the sales mean of the class with higher monthly sales to the sales mean of the class with lower monthly sales, take all months of the class with a ratio greater than the threshold h as the peak months, take the peak months of two years as the intersection, if the intersection is not empty, then the product is considered to be a seasonal product, and the sales peak occurs in the intersection month of the peak months.

[0004] In the process of realizing the present invention, the inventors found that there are at least the following technical problems in the prior art: the classification result of clustering based on monthly sales volume to determine whether a product is a seasonal product is inaccurate, which affects the recognition accuracy of seasonal products. Summary of the invention

[0005] The embodiments of the present invention provide a method, device, equipment and medium for determining the attributes of an item, so as to improve the recognition accuracy of seasonal products.

[0006] In a first aspect, an embodiment of the present invention provides a method for determining an attribute of an item, comprising:

[0007] The data acquisition module calls the data acquisition interface to acquire a statistical transaction data set of the item from the storage unit, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each statistical transaction data element includes statistical transaction data within a set time period;

[0008] For each statistical transaction data element, a statistical transaction data floating parameter is determined according to the statistical transaction data contained in the statistical transaction data element by a trading peak day determination module, and an original trading peak day corresponding to the statistical transaction data element is determined according to the statistical transaction data floating parameter;

[0009] The item attribute determination module determines the item attributes of the item according to the original peak trading day corresponding to each statistical transaction data element.

[0010] In a second aspect, an embodiment of the present invention further provides an object attribute determination device, comprising:

[0011] A data acquisition module, used to call a data acquisition interface to acquire a statistical transaction data set of an item from a storage unit, wherein the statistical transaction data set includes at least two statistical transaction data elements, each of which includes statistical transaction data within a set time period;

[0012] A busy trading day determination module, configured to determine, for each of the statistical trading data elements, a statistical trading data floating parameter according to the statistical trading data contained in the statistical trading data element, and determine, according to the statistical trading data floating parameter, an original busy trading day corresponding to the statistical trading data element;

[0013] The item attribute determination module is used to determine the item attribute of the item according to the original peak transaction day corresponding to each statistical transaction data element.

[0014] In a third aspect, an embodiment of the present invention further provides a computer device, the device comprising:

[0015] one or more processors;

[0016] A storage device for storing one or more programs;

[0017] When one or more programs are executed by one or more processors, the one or more processors implement the method for determining the attributes of an item as provided in any embodiment of the present invention.

[0018] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for determining attributes of an item as provided in any embodiment of the present invention.

[0019] The embodiment of the present invention obtains a statistical transaction data set of an item from a storage unit by calling a data acquisition interface through a data acquisition module, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each statistical transaction data element includes statistical transaction data within a set time period; for each statistical transaction data element, a statistical transaction data floating parameter is determined according to the statistical transaction data contained in the statistical transaction data element through a peak transaction day determination module, and the original peak transaction day corresponding to the statistical transaction data element is determined according to the statistical transaction data floating parameter; the item attribute determination module determines the item attribute of the item according to the original peak transaction day corresponding to each statistical transaction data element, and the original peak transaction day is determined by the statistical transaction data floating parameter obtained according to the statistical transaction data, and then the item attribute of the item is determined, thereby improving the recognition accuracy of seasonal items. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flow chart of a method for determining item attributes provided by Embodiment 1 of the present invention;

[0021] Figure 2 is a flow chart of a method for determining item attributes provided by Embodiment 2 of the present invention;

[0022] Figure 3a is a flow chart of a method for determining item attributes provided by Embodiment 3 of the present invention;

[0023] Figure 3b This is a schematic diagram of an application scenario of a method for determining an attribute of an item provided in Embodiment 3 of the present invention;

[0024] Figure 3c This is a schematic diagram of the timing of sales data before smoothing in a method for determining item attributes provided in Embodiment 3 of the present invention;

[0025] Figure 3d This is a schematic diagram of the time series of sales data after smoothing in a method for determining item attributes provided in the third embodiment of the present invention;

[0026] Figure 3e This is a schematic diagram of a sales peak mark in a method for determining item attributes provided in Embodiment 3 of the present invention;

[0027] Figure 3f This is a schematic diagram of the time series after the first year's sales volume is smoothed in a method for determining item attributes provided in the third embodiment of the present invention;

[0028] Figure 3g This is a schematic diagram of the time series after the second year sales volume is smoothed in a method for determining item attributes provided in the third embodiment of the present invention;

[0029] Figure 4 It is a structural schematic diagram of an object attribute determination device provided by a fourth embodiment of the present invention;

[0030] Figure 5 It is a structural diagram of a computer device provided in Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0031] The present invention 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 present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0032] Embodiment 1

[0033] Figure 11 is a flow chart of a method for determining item attributes provided by the first embodiment of the present invention. This embodiment is applicable to situations when determining item attributes, and is particularly applicable to situations when determining whether an item is a seasonal item. The method can be executed by an item attribute determination device, which can be implemented in software and / or hardware. For example, the item attribute determination device can be configured in a computer device. Figure 1 As shown, the method includes:

[0034] S110. Calling a data acquisition interface through a data acquisition module to acquire a statistical transaction data set of an item from a storage unit, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each statistical transaction data element includes statistical transaction data within a set time period.

[0035] In this embodiment, the statistical transaction data set of an item may be a set including at least two statistical transaction data elements, and each statistical transaction data element may include statistical transaction data within a set time period. Optionally, the transaction data within the set time period may be acquired from the storage unit by calling a pre-set data acquisition interface in the data acquisition module, and a statistical transaction data set may be formed based on the acquired transaction data. The transaction data in the storage unit may be log data automatically generated based on the transaction behavior of the user, or may be manually stored transaction data. When the attribute of the item to be determined is whether it is a seasonal item, the set time period may be one year, that is, each statistical transaction data element may be statistical transaction data within one year, and the month corresponding to the statistical transaction data range of each statistical transaction data is the same.

[0036] Exemplarily, assuming that the statistical transaction data set of an item includes statistical transaction data element 1 and statistical transaction data element 2, statistical transaction data element 1 may be statistical transaction data from January 1, 2018 to January 1, 2019, and statistical transaction data element 2 may be statistical transaction data from January 1, 2019 to January 1, 2020. Optionally, statistical transaction data element 2 may also be statistical transaction data from July 1, 2018 to July 1, 2019, and correspondingly, statistical transaction data element 2 may also be statistical transaction data from July 7, 2019 to February 1, 2020.

[0037] In addition, in this embodiment, the statistical time unit of the statistical transaction data elements is not limited here. Optionally, the statistical time unit of the statistical transaction data can be a day, a week, or a month, etc. Considering that the prior art identifies the fluctuations in sales volume according to the monthly dimension, if the product starts to have high sales from the middle of a month, it is too early to start predicting the volume from the beginning of the month, and there may be problems such as poor classification results for monthly sales classification, affecting the accuracy of seasonal product identification. Preferably, statistical transaction data with a time unit of day or week can be obtained, and the periodicity of the items can be refined to weeks or days to solve problems such as premature and high predictions caused by identification based on the month as the dimension.

[0038] S120. For each statistical transaction data element, a peak transaction day determination module determines a statistical transaction data floating parameter according to the statistical transaction data included in the statistical transaction data element, and determines the original peak transaction day corresponding to the statistical transaction data element according to the statistical transaction data floating parameter.

[0039] In this embodiment, the original peak trading day of the item can be determined by the peak trading day determination module in the item attribute determination device according to the acquired statistical transaction data set. Specifically, the peak trading day determination module can first determine the statistical transaction data floating parameter corresponding to each statistical transaction data element in the statistical transaction data set, and then determine the original peak trading day of the statistical transaction data element according to the transaction data floating parameter, so as to identify seasonal items according to the original peak trading day of each statistical transaction data element.

[0040] The statistical transaction data floating parameter may be a statistical transaction data floating value corresponding to a statistical time unit in the statistical transaction data contained in the statistical transaction data element. For example, the statistical transaction data floating parameter may be an absolute value of a slope corresponding to a statistical time unit. Assuming that the statistical time unit of the statistical transaction data contained in the statistical transaction data element is a day, the statistical transaction data floating parameter may be a transaction data floating value corresponding to each day (such as an absolute value of a slope).

[0041] In one embodiment of the present invention, determining the original peak trading day corresponding to the trading data element according to the statistical trading data floating parameter includes: using the peak trading day determination module to take the date corresponding to the statistical trading data floating parameter that is higher than the set threshold as the original peak trading day. Optionally, a threshold can be pre-set in the peak trading day determination module for comparison with the statistical trading data floating parameter corresponding to the statistical time unit in the statistical trading data element. When determining the target peak trading day, the peak trading day determination module takes the date corresponding to the statistical trading data floating parameter that is higher than the set threshold as the original peak trading day corresponding to the statistical trading data element, thereby obtaining the original peak trading day corresponding to each statistical trading data element. Exemplarily, assuming that the statistical time unit of the statistical trading data element is day, then for each statistical trading data element, the statistical trading data floating parameter of each day of the statistical trading data is compared with the set threshold, and the date corresponding to the statistical trading data floating parameter that is higher than the set threshold is taken as the original peak trading day, thereby obtaining the sales peak period of the statistical trading data corresponding to each statistical trading data element.

[0042] S130. Determine the item attributes of the item according to the original peak trading day corresponding to each statistical transaction data element through the item attribute determination module.

[0043] In this embodiment, the item attributes of the item can be determined by the original peak trading days corresponding to each statistical transaction data element of the item attribute determination module in the item attribute determination device. Specifically, the item attribute determination module can determine the item attributes of the item according to the original peak trading days corresponding to each statistical transaction data element after obtaining the original peak trading days corresponding to each statistical transaction data element. In one embodiment of the present invention, the item attributes of the item are determined according to the original peak trading days corresponding to each statistical transaction data element by the item attribute determination module, including: calculating the intersection between the original peak trading days corresponding to each statistical transaction data element by the item attribute determination module; if the intersection is empty, determining the item attribute of the item is a non-seasonal product; if the intersection is not empty, determining the item attribute of the item is a seasonal product, and taking the date corresponding to the intersection as the target peak trading day for the item.

[0044] Optionally, the intersection between the original peak trading days corresponding to each statistical transaction data element can be calculated through the item attribute determination module, and the item attribute of the item can be determined based on the calculation result. If the intersection between the original peak trading days corresponding to each statistical transaction data element is empty, it means that the sales volume of the item is not stable and periodic, and the item attribute of the item is determined to be a non-seasonal item; if the intersection between the original peak trading days corresponding to each statistical transaction data element is not empty, it means that the sales volume of the item is relatively stable and periodic, and the item attribute of the item is determined to be a seasonal item, and the date corresponding to the transaction is used as the target peak trading day for the item. It should be noted that when calculating the intersection between the original peak trading days, if the date years corresponding to the original peak trading days are different, but the month and day are the same, the intersection between the original peak trading days is considered to be non-empty. For example, if the original peak trading day 1 is April 10, 2018 - May 9, 2018, and the original peak trading day 2 is April 12, 2019 - May 10, 2019, then the intersection between the original peak trading day 1 and the original peak trading day 2 is non-empty, and the intersection is April 12 - May 9.

[0045] For example, if the statistical transaction data set includes statistical transaction data element 1 and statistical transaction data element 2, the original peak trading days corresponding to statistical transaction data element 1 are August 25, 2017 to September 28, 2017, and February 20, 2018 to March 28, 2018, and the original peak trading days corresponding to statistical transaction data element 2 are August 11, 2018 to October 31, 2018, February 20, 2019 to March 13, 2019, and March 31, 2019 to October 31, 2018. On May 6, 2019, the intersection of the original peak trading days corresponding to statistical transaction data element 1 and the original peak trading days corresponding to statistical transaction data element 2 is taken as the two periods of August 25-September 28, and February 20-March 13. That is, if the intersection of the original peak trading days corresponding to statistical transaction data element 1 and the original peak trading days corresponding to statistical transaction data element 2 is not empty, then the item is determined to be a seasonal item, and the target peak trading days of the item are August 25-September 28, and February 20-March 13.

[0046] The embodiment of the present invention obtains a statistical transaction data set of an item from a storage unit by calling a data acquisition interface through a data acquisition module, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each statistical transaction data element includes statistical transaction data within a set time period; for each statistical transaction data element, a statistical transaction data floating parameter is determined according to the statistical transaction data contained in the statistical transaction data element through a peak transaction day determination module, and the original peak transaction day corresponding to the statistical transaction data element is determined according to the statistical transaction data floating parameter; the item attribute determination module determines the item attribute of the item according to the original peak transaction day corresponding to each statistical transaction data element, and the original peak transaction day is determined by the statistical transaction data floating parameter obtained according to the statistical transaction data, and then the item attribute of the item is determined, thereby improving the recognition accuracy of seasonal items.

[0047] On the basis of the above scheme, it also includes: determining the transaction forecast value of the target peak trading day according to the target peak trading day of the item through the inventory adjustment module, and determining the target inventory quantity of the item according to the transaction forecast value. Optionally, the transaction value of the item during the target peak trading day can be predicted by the inventory adjustment module in the item attribute determination device, so as to adjust the inventory quantity based on the obtained transaction forecast value. Specifically, after identifying that the item is a seasonal item and the target peak trading day of the item (i.e., the peak sales period), the inventory adjustment module can increase the forecast value in the corresponding period when predicting the sales volume of the item, so as to determine the target inventory quantity of the item according to the increased forecast value, so that the replenishment system determines the replenishment quantity according to the target inventory quantity and the current inventory quantity, so as to improve the accuracy of the replenishment quantity.

[0048] In one embodiment of the present invention, the transaction forecast value of the target peak trading day is determined according to the target peak trading day of the item through the inventory adjustment module, including: the target peak trading day is used as the transaction quantity prediction feature of the item through the inventory adjustment module, and is input into the trained transaction quantity prediction model to obtain the transaction forecast value output by the transaction quantity prediction model. Optionally, the sales volume of the item can be predicted by a machine learning algorithm. Exemplarily, a transaction quantity prediction model is constructed in the inventory adjustment module, the target peak trading day of the sample is used as one of the training features of the transaction quantity prediction model, and the constructed transaction quantity prediction model is trained in combination with other transaction quantity prediction features (such as the features of the existing transaction quantity prediction) to obtain a trained transaction quantity prediction model. When predicting the transaction value, the target peak trading day of the predicted item is used as one of the transaction quantity prediction features, and is input into the trained transaction quantity prediction model in combination with other transaction quantity prediction features of the item (such as the features of the existing transaction quantity prediction), so as to improve the prediction accuracy of the transaction quantity and further improve the accuracy of the replenishment quantity.

[0049] Embodiment 2

[0050] Figure 2 1 is a flow chart of a method for determining item attributes provided by the second embodiment of the present invention. Based on the above embodiment, this embodiment specifies the determination of the statistical transaction data floating parameter by the busy transaction day determination module according to the statistical transaction data contained in the statistical transaction data element. Figure 2 As shown, the method includes:

[0051] S210. Calling a data acquisition interface through a data acquisition module to acquire a statistical transaction data set of an item from a storage unit, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each statistical transaction data element includes statistical transaction data within a set time period.

[0052] S220. For each statistical transaction data element, the statistical transaction data contained in the statistical transaction data element is smoothed by the busy trading day determination module to obtain smoothed transaction data, and the slope corresponding to the preset statistical time unit is calculated based on the smoothed transaction data, and the absolute value of the slope corresponding to the preset statistical time unit is used as the floating parameter of the statistical transaction data.

[0053] Considering the interference of short-term sales peaks due to promotions and other reasons on seasonal product identification, in this embodiment, for each statistical transaction data element, the statistical transaction data corresponding to the statistical transaction data element can be firstly smoothed by the transaction peak day determination module, and then the transaction data floating parameter of the statistical time unit in the statistical transaction data element can be calculated.

[0054] In one embodiment of the present invention, the statistical transaction data included in the statistical transaction data element is subjected to data smoothing processing by the busy trading day determination module to obtain smoothed transaction data, including: performing moving average processing on the statistical transaction data by the busy trading day determination module to obtain moving average transaction data, and performing local weighted regression smoothing processing on the moving average transaction data to obtain smoothed transaction data. Optionally, the data smoothing processing on the statistical transaction data in the busy trading day determination module may include two steps: moving average processing and local weighted regression smoothing processing. The moving average method can effectively eliminate random fluctuations in the prediction. Local weighted regression smoothing processing can perform better data smoothing on data with certain trends and seasonality. Specifically, firstly, the statistical transaction data is subjected to moving average processing to obtain moving average transaction data. Optionally, the statistical transaction data can be subjected to 3-step moving average processing to obtain moving average transaction data, and then the moving average transaction data can be subjected to local weighted regression smoothing processing to obtain smoothed transaction data of the statistical transaction data. Smoothing the statistical transaction data can obtain smoothed transaction data that converts the statistical transaction data into a sales peak trend and filters out abnormal sales in a short period of time.

[0055] On the basis of the above scheme, after obtaining the smoothed transaction data corresponding to the statistical transaction data, the busy trading day determination module determines the statistical data floating parameter according to the data trend in the smoothed transaction data. Optionally, the busy trading day determination module can calculate the slope corresponding to the preset statistical time unit in the smoothed transaction data, and use the absolute value of the slope corresponding to the preset statistical time unit as the statistical transaction data floating parameter. Among them, the preset statistical time unit can be a day, a week, a month, etc. In order to make the identification of seasonal products more accurate, the preset statistical time unit can be set to a day, that is, the slope corresponding to each day in the smoothed transaction data is calculated and used as the floating parameter of the transaction data corresponding to each day.

[0056] In one embodiment of the present invention, the slope corresponding to the preset statistical time unit is calculated based on the smoothed transaction data, including: normalizing the smoothed transaction data through the busy trading day determination module to obtain normalized transaction data; performing differential operation on the normalized transaction data to obtain the slope corresponding to the preset statistical time unit. Specifically, the smoothed transaction data can be normalized through the busy trading day determination module, that is, the annual daily sales time series of the item is normalized to obtain normalized transaction data, and then the normalized transaction data is differentially processed, and the differential result is used as the slope value of the time series on that day, that is, the daily slope.

[0057] S230. Determine the original peak trading day corresponding to the statistical trading data element according to the statistical trading data floating parameter through the peak trading day determination module.

[0058] S240: Determine the item attributes of the item according to the original peak trading day corresponding to each statistical transaction data element through the item attribute determination module.

[0059] The embodiment of the present invention concretizes the determination of the floating parameter of statistical transaction data based on the statistical transaction data included in the statistical transaction data elements, and obtains the smoothed transaction data by performing data smoothing processing on the statistical transaction data included in the statistical transaction data elements for each statistical transaction data element. For the smoothed transaction data, the slope corresponding to the preset statistical time unit is calculated, and the absolute value of the slope corresponding to the preset statistical time unit is used as the floating parameter of the statistical transaction data, thereby filtering out the interference of short-term sales peaks caused by sales promotions and the like on the identification of seasonal products, thereby improving the identification accuracy of seasonal products.

[0060] Embodiment 3

[0061] Figure 3a This is a flow chart of a method for determining item attributes provided by Embodiment 3 of the present invention. This embodiment provides a preferred embodiment based on the above embodiments.

[0062] The method provided by the embodiment of the present invention can be executed by a seasonal product identification device, and can be mainly applied to the upstream system that the supply chain replenishment system relies on - the sales forecasting system. By labeling seasonal products, the products (sku) are classified and refined scenario-based forecasting is performed to improve the forecasting accuracy. The forecast output value is input into the downstream replenishment system to improve the accuracy of the replenishment quantity, which has extremely high actual production value. The links of the seasonal product identification module in the actual production system can be referred to Figure 3b . Figure 3b Schematic diagram of an application scenario of a method for determining item attributes provided by Embodiment 3 of the present invention. Figure 3b As shown, the sales forecasting system includes a seasonal product identification module and a sales forecasting module. After the seasonal product identification module obtains the historical sales data, it analyzes the historical sales data and marks the seasonal products. The sales forecasting module predicts the sales volume of the product according to the identification results of the seasonal product identification module, and sends the predicted sales volume to the target inventory calculation module. The replenishment system and the procurement system replenish or purchase according to the target inventory calculated by the target inventory calculation module.

[0063] Specifically, the method for determining the property of an item provided in the third embodiment of the present invention is as follows: Figure 3a As shown, the method includes:

[0064] S310. Smooth the historical sales volume.

[0065] In this embodiment, the sales volume is smoothed before identifying seasonal products to avoid interference with the identification of seasonal products caused by short-term sales peaks due to promotions. Sales volume smoothing includes two steps: 3-step moving average of sales volume, and local weighted regression (lowess) smoothing of the sales volume after moving average. The general idea of ​​local weighted regression (lowess) is: take a point x as the center, intercept a segment of data with a length of frac forward and backward, and use a weighted linear regression with a weight function w for this segment of data, record is the center value of the regression line, where is the corresponding value of the curve after fitting. For all n data points, n weighted regression lines can be made, and the line connecting the central values ​​of each regression line is the Lowess curve of this data segment.

[0066] Figure 3c 1 is a schematic diagram of the timing of sales data before smoothing in a method for determining item attributes provided in the third embodiment of the present invention. Figure 3d 1 is a schematic diagram of the time series of sales data after smoothing in a method for determining item attributes provided in the third embodiment of the present invention. Figure 3d It is a time series chart after the original data is smoothed by combining moving average and lowess regression. Figure 3c as well as Figure 3dIt can be seen that the smoothed time series chart retains the sales peak trend in September 2017 and March 2018. At the same time, the short-term sales anomalies that appeared in May and June 2018 are also reasonably smoothed out.

[0067] S320. Identify seasonal products based on the sales volume smoothing result.

[0068] Specifically, seasonal product identification based on the sales volume smoothing results can be as follows:

[0069] 1) Normalize the annual daily sales time series of sku.

[0070] 2) Perform differential processing on the normalized time series, and the differential result is used as the slope k value of the time series on that day.

[0071] 3) Set a threshold h and mark the time period when the absolute value of the slope k is greater than h. This period is the period when the sales peak occurs.

[0072] Figure 3e FIG. 1 is a schematic diagram of a sales peak mark in a method for determining item attributes provided in Embodiment 3 of the present invention. Figure 3e As shown in FIG. 1 , the figure schematically marks the time period when the absolute value of the slope k is greater than the threshold h, wherein the points marked with an upper triangle represent points where k is greater than h, and the points marked with a lower triangle represent points where k is less than -h. Figure 3e It can be seen that the peak sales periods are 2017 / 08 / 25-2017 / 09 / 28, and 2018 / 02 / 20-2018 / 03 / 28.

[0073] 4) Take the sales time series of the past two years and mark the sales peak period of each year through the above operations. Take the intersection of the sales peak periods of the two years. If the intersection is empty, it is considered that the SKU will not have periodic sales peaks and is marked as a non-seasonal product. Otherwise, it is marked as a seasonal product, and the intersection of the time periods is the time period when the sales peak of the seasonal product occurs.

[0074] Optionally, when the sku is a seasonal product, the forecast value will be increased accordingly for the period of peak sales volume during the forecast.

[0075] Exemplarily, the identification of seasonal products and sales peak periods using a specific SKU according to the above method is used as an example to illustrate. Figure 3f is a schematic diagram of the smoothed sales volume in the first year in a method for determining item attributes provided in the third embodiment of the present invention. Figure 3g 1 is a schematic diagram of the time series of the second year sales volume smoothed in a method for determining item attributes provided by the third embodiment of the present invention. Figure 3fAs shown in Figure 2, after smoothing the sales volume of the first year’s statistical transaction data, the peak sales volume periods occurred during: 2017 / 08 / 25-2017 / 09 / 28, and 2018 / 02 / 20-2018 / 03 / 28; Figure 3g As shown in the figure, after the sales volume of the statistical transaction data of the second year is smoothed, the sales peak periods occur in: 2018 / 08 / 11-2018 / 10 / 31, 2019 / 02 / 20-2019 / 03 / 13, and 2019 / 03 / 31-2019 / 05 / 06; the intersection of the two sales peak periods is 08 / 25-09 / 28 and 02 / 20-03 / 13, then the SKU is considered to be a seasonal product, and the sales peak occurs in the two periods of 08 / 25-09 / 28 and 02 / 20-03 / 13.

[0076] The embodiment of the present invention uses a slope threshold to determine the sales peak and then identify seasonal products, and combines moving average and Lowess regression on the sales time series to capture the trend characteristics of sales, thereby improving the accuracy of sales characteristics and thus improving the recognition accuracy of seasonal products.

[0077] Embodiment 4

[0078] Figure 4 1 is a schematic diagram of the structure of an item attribute determination device provided by the fourth embodiment of the present invention. The item attribute determination device can be implemented in software and / or hardware, for example, the item attribute determination device can be configured in a computer device. Figure 4 As shown, the device includes a data acquisition module 410, a peak trading day determination module 420 and an item attribute determination module 430, wherein:

[0079] The data acquisition module 410 is used to call the data acquisition interface to acquire a statistical transaction data set of an item from a storage unit, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each of the statistical transaction data elements includes statistical transaction data within a set time period;

[0080] A busy trading day determination module 420 is used to determine, for each of the statistical trading data elements, a statistical trading data floating parameter according to the statistical trading data contained in the statistical trading data element, and determine the original busy trading day corresponding to the statistical trading data element according to the statistical trading data floating parameter;

[0081] The item attribute determination module 430 is used to determine the item attribute of the item according to the original peak transaction day corresponding to each statistical transaction data element.

[0082] In an embodiment of the present invention, a data acquisition module calls a data acquisition interface to acquire a statistical transaction data set of an item from a storage unit, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each statistical transaction data element includes statistical transaction data within a set time period; a peak transaction day determination module determines a statistical transaction data floating parameter for each statistical transaction data element based on the statistical transaction data contained in the statistical transaction data element, and determines the original peak transaction day corresponding to the statistical transaction data element based on the statistical transaction data floating parameter; an item attribute determination module determines the item attributes of the item based on the original peak transaction day corresponding to each statistical transaction data element, determines the original peak transaction day by determining the statistical transaction data floating parameter obtained based on the statistical transaction data, and then determines the item attributes of the item, thereby improving the recognition accuracy of seasonal items.

[0083] Optionally, based on the above solution, the busy trading day determination module 420 is specifically used for:

[0084] The statistical transaction data contained in the statistical transaction data element is subjected to data smoothing to obtain smoothed transaction data, and a slope corresponding to a preset statistical time unit is calculated based on the smoothed transaction data, and the absolute value of the slope corresponding to the preset statistical time unit is used as the floating parameter of the statistical transaction data.

[0085] Optionally, based on the above solution, the busy trading day determination module 420 is specifically used for:

[0086] Normalizing the smoothed transaction data to obtain normalized transaction data;

[0087] The normalized transaction data is subjected to a differential operation to obtain a slope corresponding to the preset statistical time unit.

[0088] Optionally, based on the above solution, the busy trading day determination module 420 is specifically used for:

[0089] The statistical transaction data is subjected to moving average processing to obtain moving average transaction data, and the moving average transaction data is subjected to local weighted regression smoothing processing to obtain the smoothed transaction data.

[0090] Optionally, based on the above solution, the busy trading day determination module 420 is specifically used for:

[0091] The date corresponding to the floating parameter of the statistical transaction data that is higher than the set threshold is used as the original peak transaction day.

[0092] Optionally, based on the above solution, the item attribute determination module 430 is specifically used to:

[0093] Calculating the intersection between the original peak trading days corresponding to the statistical trading data elements;

[0094] If the intersection is empty, the item attribute of the item is determined to be a non-seasonal item;

[0095] If the intersection is not empty, the item attribute of the item is determined to be a seasonal item, and the date corresponding to the intersection is used as the target peak trading day for the item.

[0096] Optionally, based on the above solution, the device further includes a transaction quantity prediction module, which is used to:

[0097] According to the target peak trading day of the item, a transaction forecast value for the target peak trading day is determined, and according to the transaction forecast value, a target inventory quantity of the item is determined.

[0098] Optionally, based on the above solution, the transaction quantity prediction module is specifically used to:

[0099] The target peak trading day is used as a transaction quantity prediction feature of the item and is input into a trained transaction quantity prediction model to obtain the transaction prediction value output by the transaction quantity prediction model.

[0100] The device for determining item attributes provided in the embodiment of the present invention can execute the method for determining item attributes provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0101] Embodiment 5

[0102] Figure 5 It is a structural diagram of a computer device provided in Embodiment 5 of the present invention. Figure 5 A block diagram of an exemplary computer device 512 suitable for use in implementing embodiments of the present invention is shown. Figure 5 The computer device 512 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0103] like Figure 5 As shown, the computer device 512 is in the form of a general-purpose computing device. The components of the computer device 512 may include, but are not limited to: one or more processors 516, a system memory 528, and a bus 518 that connects various system components (including the system memory 528 and the processor 516).

[0104] Bus 518 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a local bus to processor 516, or a bus structure using any of a variety of bus architectures. These architectures include, by way of example, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0105] The computer device 512 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 512, including volatile and non-volatile media, removable and non-removable media.

[0106] System memory 528 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 530 and / or cache memory 532. Computer device 512 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage device 534 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown, usually called a "hard drive"). Although Figure 5 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 518 via one or more data medium interfaces. The memory 528 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0107] A program / utility 540 having a set (at least one) of program modules 542 may be stored, for example, in the memory 528, such program modules 542 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 542 generally perform the functions and / or methods of the embodiments described herein.

[0108] The computer device 512 may also communicate with one or more external devices 514 (e.g., keyboard, pointing device, display 524, etc.), may also communicate with one or more devices that enable a user to interact with the computer device 512, and / or communicate with any device that enables the computer device 512 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed via an input / output (I / O) interface 522. In addition, the computer device 512 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) via a network adapter 520. As shown, the network adapter 520 communicates with other modules of the computer device 512 via a bus 518. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the computer device 512, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0109] The processor 516 executes various functional applications and data processing by running the program stored in the system memory 528, for example, implementing the method for determining the attribute of an item provided in an embodiment of the present invention, which includes:

[0110] Acquire a statistical transaction data set of an item, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each of the statistical transaction data elements includes statistical transaction data within a set time period;

[0111] For each of the statistical transaction data elements, determining a statistical transaction data floating parameter according to the statistical transaction data contained in the statistical transaction data element, and determining the original peak transaction day corresponding to the statistical transaction data element according to the statistical transaction data floating parameter;

[0112] The item attribute of the item is determined according to the original peak transaction day corresponding to each of the statistical transaction data elements.

[0113] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the method for determining the attribute of an item provided in any embodiment of the present invention.

[0114] Embodiment 6

[0115] Embodiment 6 of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the method for determining the attributes of an item provided in the embodiment of the present invention is implemented. The method includes:

[0116] Acquire a statistical transaction data set of an item, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each of the statistical transaction data elements includes statistical transaction data within a set time period;

[0117] For each of the statistical transaction data elements, determining a statistical transaction data floating parameter according to the statistical transaction data contained in the statistical transaction data element, and determining the original peak transaction day corresponding to the statistical transaction data element according to the statistical transaction data floating parameter;

[0118] The item attribute of the item is determined according to the original peak transaction day corresponding to each of the statistical transaction data elements.

[0119] Of course, the computer-readable storage medium provided by an embodiment of the present invention stores a computer program which is not limited to the method operations described above, but can also execute related operations of the method for determining the attributes of an item provided by any embodiment of the present invention.

[0120] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can 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 (non-exhaustive list) of computer-readable storage media include: 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 this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0121] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals 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 a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0122] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0123] Computer program code for performing the operations of the present invention 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).

[0124] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for determining an item attribute, characterized in that: include: Acquire a statistical transaction data set of an item from a storage unit by calling a data acquisition interface through a data acquisition module, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each of the statistical transaction data elements includes statistical transaction data within a set time period; For each of the statistical transaction data elements, a statistical transaction data floating parameter is determined according to the statistical transaction data contained in the statistical transaction data element by a trading peak day determination module, and the original trading peak day corresponding to the statistical transaction data element is determined according to the statistical transaction data floating parameter; wherein the statistical transaction data floating parameter is the absolute value of the slope corresponding to the preset statistical time unit; The item attribute determination module determines the item attribute of the item according to the original peak transaction day corresponding to each statistical transaction data element.

2. The method according to claim 1, characterized in that The method of determining the statistical transaction data floating parameter according to the statistical transaction data included in the statistical transaction data element by the busy transaction day determination module includes: The statistical transaction data contained in the statistical transaction data element is smoothed by the busy trading day determination module to obtain smoothed transaction data, and the slope corresponding to the preset statistical time unit is calculated based on the smoothed transaction data, and the absolute value of the slope corresponding to the preset statistical time unit is used as the floating parameter of the statistical transaction data.

3. The method according to claim 2, characterized in that The calculating the slope corresponding to the preset statistical time unit according to the smoothed transaction data includes: The smoothed transaction data is normalized by a busy trading day determination module to obtain normalized transaction data; The normalized transaction data is subjected to a differential operation to obtain a slope corresponding to the preset statistical time unit.

4. The method according to claim 2, characterized in that: The method of performing data smoothing processing on the statistical transaction data included in the statistical transaction data element by the busy transaction day determination module to obtain smoothed transaction data includes: The busy trading day determination module performs moving average processing on the statistical trading data to obtain moving average trading data, and performs local weighted regression smoothing processing on the moving average trading data to obtain the smoothed trading data.

5. The method according to claim 1, characterized in that The determining, according to the statistical transaction data floating parameter, the original peak trading day corresponding to the transaction data element comprises: The trading peak day determination module uses the date corresponding to the statistical trading data floating parameter that is higher than the set threshold as the original trading peak day.

6. The method according to claim 1, characterized in that The item attribute determination module determines the item attribute of the item according to the original peak trading day corresponding to each statistical transaction data element, including: Calculate the intersection of the original peak trading days corresponding to the statistical transaction data elements through the item attribute determination module; If the intersection is empty, the item attribute of the item is determined to be a non-seasonal item; If the intersection is not empty, the item attribute of the item is determined to be a seasonal item, and the date corresponding to the intersection is used as the target peak trading day for the item.

7. The method according to claim 6, characterized in that Also includes: The inventory adjustment module determines the transaction forecast value of the target peak trading day according to the target peak trading day of the item, and determines the target inventory quantity of the item according to the transaction forecast value.

8. The method according to claim 7, characterized in that The determining the transaction forecast value of the target peak trading day according to the target peak trading day of the item by the inventory adjustment module includes: The target peak trading day is used as a transaction quantity prediction feature of the item through an inventory adjustment module and is input into a trained transaction quantity prediction model to obtain the transaction prediction value output by the transaction quantity prediction model.

9. A device for determining an attribute of an item, characterized in that: include: A data acquisition module, used to call a data acquisition interface to acquire a statistical transaction data set of an item from a storage unit, wherein the statistical transaction data set includes at least two statistical transaction data elements, and each of the statistical transaction data elements includes statistical transaction data within a set time period; A busy trading day determination module is used to determine, for each of the statistical trading data elements, a statistical trading data floating parameter according to the statistical trading data contained in the statistical trading data element, and determine the original busy trading day corresponding to the statistical trading data element according to the statistical trading data floating parameter; wherein the statistical trading data floating parameter is the absolute value of the slope corresponding to the preset statistical time unit; The item attribute determination module is used to determine the item attribute of the item according to the original peak transaction day corresponding to each statistical transaction data element.

10. A computer device, characterized in that: The device comprises: 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 for determining the attributes of an item as described in any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for determining the attributes of an item as described in any one of claims 1 to 8 is implemented.

Images