CN113538055B - Sales volume estimation method, device and processor for Amazon platform commodity - Google Patents
Sales volume estimation method, device and processor for Amazon platform commodity Download PDFInfo
- Publication number
- CN113538055B CN113538055B CN202110822098.9A CN202110822098A CN113538055B CN 113538055 B CN113538055 B CN 113538055B CN 202110822098 A CN202110822098 A CN 202110822098A CN 113538055 B CN113538055 B CN 113538055B
- Authority
- CN
- China
- Prior art keywords
- sales
- commodity
- commodities
- ranking
- predicted
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000004364 calculation method Methods 0.000 claims abstract description 44
- 238000003860 storage Methods 0.000 claims abstract description 18
- 230000008859 change Effects 0.000 claims abstract description 14
- 230000001419 dependent effect Effects 0.000 claims description 17
- 238000000611 regression analysis Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 description 26
- 238000004590 computer program Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 9
- 238000012552 review Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000012417 linear regression Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000003796 beauty Effects 0.000 description 1
- 230000001364 causal effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Abstract
The embodiment of the application provides a sales volume estimation method, a sales volume estimation device, a processor and a storage medium for Amazon platform commodities. The method comprises the following steps: aiming at each commodity of the Amazon platform, acquiring the current commodity inventory and sales ranking of the commodity at intervals of preset time; determining sales of the commodity according to the change of the commodity inventory; inputting sales of a plurality of commodities and corresponding sales ranking into a regression prediction model, and determining parameter values of a polynomial regression equation included in the regression prediction model; acquiring sales ranking of commodities to be predicted; inputting sales ranking of commodities to be predicted into a regression prediction model to determine an exponential weighted average of the commodities to be predicted according to a polynomial regression equation; determining a sales ranking calculation formula of the Amazon platform; and carrying out reverse engineering treatment on the sales ranking calculation formula so as to determine the sales of the commodity to be predicted in a preset time period according to the index weighted average value of the commodity to be predicted.
Description
Technical Field
The application relates to the technical field of computers, in particular to a sales volume estimation method, a sales volume estimation device, a storage medium and a processor for Amazon platform commodities.
Background
The sales of the commodity change, which represents the change of the intensity of the demand. The evaluation of the monthly sales (the last 30 days sales) of the goods on the amazon platform is very important for the market analysis and selection of sellers. In the prior art, sales calculation methods generally include Review estimation, feed back estimation, daily inventory difference analysis, and the like.
However, the estimation methods of Review and feed back have great errors, can only be used as references, and basically have no great application value in the age of fine operation. In addition to the aforementioned drawbacks, the current method for comparing common daily inventory difference is that the model is too simple (the model considers that the BSR rank is only related to daily sales), the factors are too few, the granularity of data is too coarse (sales are estimated daily), and the like, so that the deviation of sales prediction is larger.
Disclosure of Invention
The embodiment of the application aims to provide a sales predicting method, a sales predicting device, a storage medium and a processor for Amazon platform commodities.
In order to achieve the above object, a first aspect of the present application provides a sales predicting method for an amazon platform commodity, including:
aiming at each commodity of the Amazon platform, acquiring the current commodity inventory and sales ranking of the commodity at intervals of preset time;
determining sales of the commodity according to the variation of the commodity inventory;
inputting sales of a plurality of commodities and corresponding sales ranking into a regression prediction model, and determining parameter values of a polynomial regression equation included in the regression prediction model;
acquiring sales ranking of commodities to be predicted;
inputting sales ranking of commodities to be predicted into a regression prediction model to determine an exponential weighted average of the commodities to be predicted according to a polynomial regression equation;
determining a sales ranking calculation formula of the Amazon platform, wherein independent variables of the sales ranking calculation formula comprise sales of commodities in a preset time period, an exponential weighted average value, an exponential function and a weighting coefficient of the exponential function, and the dependent variables are sales ranks;
and carrying out reverse engineering treatment on the sales ranking calculation formula so as to determine the sales of the commodity to be predicted in a preset time period according to the index weighted average value of the commodity to be predicted.
Optionally, the weighting coefficients are determined by regression analysis by inputting sales and sales ranks of the commodity into a linear model.
Optionally, the sales ranking calculation formula is:wherein y is an exponentially weighted average, x n The sales of commodity per hour in n hours are represented, a is an exponential function, and t is a weighting coefficient of the exponential function.
Optionally, the polynomial regression equation is: y=a+b x 2 +c*x 3 +d*x 4 +e*x 5 ；
Where y is an exponentially weighted average, x is the sales ranking of the good, and a, b, c, d, e are parameters of a polynomial regression equation.
Optionally, the preset time and the preset time period are per hour; the step of obtaining the current commodity inventory and sales ranking of the commodities at intervals of preset time comprises the following steps: and acquiring the current commodity inventory of the commodity and the BSR ranking of the commodity every other hour.
Optionally, in the sales ranking calculation formula, the more recently sales data, the more weight and the exponential variation are presented.
The second aspect of the application provides a sales predicting device for an amazon platform commodity, comprising:
the data statistics module is used for acquiring the current commodity inventory and sales ranking of the commodities at intervals of preset time for each commodity of the Amazon platform, and determining the sales of the commodities according to the variation of the commodity inventory;
the parameter determining module is used for inputting sales of a plurality of commodities and corresponding sales ranking into the regression prediction model and determining parameter values of a polynomial regression equation included in the regression prediction model;
the sales predicting module is used for obtaining sales ranking of the commodities to be predicted; inputting sales ranking of commodities to be predicted into a regression prediction model to determine an exponential weighted average of the commodities to be predicted according to a polynomial regression equation; determining a sales ranking calculation formula of the Amazon platform, wherein independent variables of the sales ranking calculation formula comprise sales of commodities in a preset time period, an exponential weighted average value, an exponential function and a weighting coefficient of the exponential function, and the dependent variables are sales ranks; and carrying out reverse engineering treatment on the sales ranking calculation formula so as to determine the sales of the commodity to be predicted in a preset time period according to the index weighted average value of the commodity to be predicted.
A third aspect of the application provides a machine-readable storage medium having instructions stored thereon that, when executed by a processor, cause the processor to be configured to perform the sales prediction method for amazon platform commodity described above.
In a fourth aspect, the present application provides a processor configured to perform the sales prediction method for amazon platform commodity described above.
According to the sales volume estimation method for the Amazon platform commodity, the acquired data are the BSR and inventory information acquired per hour, so that the established model and parameters can be updated once per hour. Whereas in the prior art at most only daily updates can be achieved. Therefore, the model established in the application is more accurate, the iteration and optimization frequency is faster, and sales volume prediction of commodities to be predicted is more accurate. In addition, by updating every hour, not only the influence of the current sales on the BSR but also the influence of the sales speed (the rate of change of sales) over the past 14 days on the BSR was analyzed for correlation, and this influence was not negligible in practice. In addition, the sales volume estimation model established in the application has the advantages that the completeness and the updating frequency of data are obviously improved compared with the prior art, meanwhile, the influence of ASIN sales speed is taken as an important factor for influencing BSR ranking, and the Amazon BSR algorithm is highly restored, so that the obtained estimated sales volume is closer to the actual sales volume, and the prediction accuracy is obviously improved.
Additional features and advantages of embodiments of the application will be set forth in the detailed description which follows.
Drawings
The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain, without limitation, the embodiments of the application. In the drawings:
FIG. 1 schematically illustrates a flow diagram of a sales prediction method for Amazon platform commodity according to an embodiment of the present application;
FIG. 2 schematically illustrates a block diagram of a sales volume estimation apparatus for Amazon platform commodity according to an embodiment of the present application;
fig. 3 schematically shows an internal structural view of a computer device according to an embodiment of the present application.
Detailed Description
The following describes the detailed implementation of the embodiments of the present application with reference to the drawings. It should be understood that the detailed description and specific examples, while indicating and illustrating the application, are not intended to limit the application.
In the prior art, for BSR ranking calculation of commodities on amazon, there are the following calculation methods:
1. review estimation method
The rating of the consumer in the general united states on amazon platforms is between 3% and 5%, and thus the sales are deduced therefrom. The statistics and display of commodity reviews on Amazon are started from the beginning of the Listing, and the data before three years is not much referenceable to the current operation, so that when the Review data is counted, the sales estimation is performed only by taking the last three months or half a year as a time scale, and the sales estimation is closer to the real data. For example, by counting the number of remarks last three months (calculated as 90 days), it is inferred at a ratio of 1:300 or 1:500. If the number of the last three months is 10, then the total sales for the last 90 days is approximately 3000, then 3000/90.apprxeq.33, i.e. the daily sales is 33, according to a ratio of 1:300; if the ratio is 1:500, the daily sales of the commodity can be estimated to be 55.
2. Feed back estimation method
Directly find the number of feeds back of the store corresponding to the commodity for the last 30 days 5 is approximately equal to the average daily sales. If it is said that there are many vendors that need to brush reviews for operations, then only few vendors will brush feeds. Since the number of feeds back is very small and helps the operation, it can be said that there is no effect at all, so there are few sellers to brush feeds back, which means that the number of feeds back is closer to the real data. Empirically, on a U.S. station, if the sales of a store are substantially stable, then 4-5 times the number of feeds back in its store for 30 days corresponds to the number of orders in the store for one day. Knowing the total order quantity of the store, entering the store to check, and checking the positions of the Listing in the store, which are being evaluated by the user, because the defaults of the commodities in the store are ordered according to the sales quantity, and roughly evaluating the sales quantity of the Listing by matching with the evaluation according to the Review quantity.
3. Daily inventory difference analysis method
The item is added to the shopping cart and then the order quantity is set to the maximum quantity 999. If the other party's stock quantity is less than 999, amazon will indicate how much is available for sale, so that the current stock quantity of the item is known. The following day repeats the above operation for one or two consecutive weeks, and the average daily sales of the Listing can be estimated by averaging the recorded data. This approach has several drawbacks. First, if the inventory is greater than 999, or if the seller may set a maximum order quantity (Max Order Quantity) to limit the purchase of the merchandise, this method of estimating sales by inventory differences fails. Secondly, one premise of this approach is that there are only 1 seller of the Listing, which is not applicable to Listing where there are many people to follow. Third, this method usually finds ASIN within 50 ten thousand of the BSR rank of the large class in Amazon station in the beginning of the month, and then estimates the month sales of the ASIN in the natural month based on various models, and this recording principle brings many problems, such as:
1) Many ASIN's have not been included for many of the clothing, outdoor, and many of these SKUs.
2) New products that the early BSR was not the first 50 thousands, and later surge, are likely not collected.
3) For those variants that do not share a parent BSR, sales may be small and no BSR is available nor is it collected.
4) Some minor categories, such as tablets standards (ipad rack, amazon in-station search: ipad holder), only a small BSR, and therefore no listings.
It can be seen that the above prior art has various disadvantages. For example, the estimation methods of Review and feed back have great errors, can only be used as references, and basically have no great application value in the age of fine operation. In addition to the aforementioned drawbacks, the current method for comparing common daily inventory difference is that the model is too simple (the model considers that the BSR rank is only related to daily sales), the factors are too few, the granularity of data is too coarse (sales are estimated daily), and the like, so that the deviation of sales prediction is larger. For example:
1) Because BSR is based on order volume, rather than sales volume, an order containing 10 items and an order containing 1 item, the same is true for BSR promotion. So when a certain type of product users often buy more than 1 piece, sales prediction bias is large.
2) If the BSR changes drastically due to a promotion such as killing for seconds on the same day, for example, from 10 tens of thousands to 2000, the average value on the same day is about 2 tens of thousands (the average value is calculated in four time periods on the same day), it is not very accurate.
3) If the good is out of stock, the BSR steadily decreases but has a value, the system will predict sales on the same day, with the actual sales being 0.
4) Because data is not monitored timely, such as if some ASIN does not monitor one or more times a day, there is no BSR on the same day, and the average on the same day can only be predicted by the BSR on the last days.
5) Some minor categories, such as tablets standards (ipad rack, amazon in-station search: ipad holder), basically without the large BSR, cannot be predicted based on the large BSR. Sales may be quite discrete if carelessly recorded and predicted.
6) An overall increase in sales for a category, such as amazon Prime Day and black five, results in ASIN for the same BSR rank, which is less well predicted when sales are increased over the month of the month.
7) For head goods, especially BSR's within 500, the prediction error may be relatively large. Because the BSR is based on a ranking, such as the top 10 product of a category, the ranking is substantially stable, but the BSR ranking for a certain ASIN increases from 8 th to 5 th, because there are days when sales increase from 300 to 1000, and the actual daily sales increase from 300 to 500, as well as the BSR ranking.
8) The ASIN category suddenly changes from Beauty & Personal Care to Health & Household and sales may be inaccurate for several days.
Based on the above consideration, in an embodiment of the present application, a sales predicting method for an amazon platform commodity is provided, as shown in fig. 1, the method includes the following steps:
step 101, acquiring current commodity inventory and sales ranking of commodities at intervals of preset time for each commodity of the Amazon platform.
Step 102, determining sales of the commodity according to the change of the commodity inventory.
And step 103, inputting sales of the commodities and corresponding sales ranking into the regression prediction model, and determining parameter values of a polynomial regression equation included in the regression prediction model.
And 104, acquiring sales ranking of the commodities to be predicted.
And 105, inputting sales ranking of the commodities to be predicted into a regression prediction model to determine an exponentially weighted average of the commodities to be predicted according to a polynomial regression equation.
And 106, determining a sales ranking calculation formula of the Amazon platform, wherein independent variables of the sales ranking calculation formula comprise sales of commodities in a preset time period, an exponential weighted average value, an exponential function and a weighting coefficient of the exponential function, and the dependent variables are sales ranks.
And 107, performing reverse engineering processing on the sales ranking calculation formula to determine the sales of the commodity to be predicted in a preset time period according to the index weighted average value of the commodity to be predicted.
Firstly, for each commodity of the Amazon platform, the current commodity inventory and sales ranking of the commodity can be obtained every other preset time. In particular, the preset time may refer to an hourly sales ranking, referring to an amazon BSR ranking. BSR ranking refers to the marketable ranking of items on the amazon platform, which is updated continuously, typically once an hour. Wherein, the lower the number of BSR ranks of the items, the higher the rank. Because the amazon platform has a large number of commodities and needs to acquire data once per hour, a crawler server group with mass data processing capability can be established, and commodity basic information of each amazon site can be crawled per hour, including but not limited to inventory of the commodities and BSR ranking. After the inventory of the commodity is obtained, the sales of the commodity can be calculated according to the change of the inventory of the commodity. For example, if the stock amount of commodity a is 2000 at 19:00 and the stock amount of commodity a is 1000 again at 20:00, it can be determined that the sales amount of commodity a in this hour is 1000. Then, based on these data, a regression prediction model may be built with the sales and BSR ranks for each period over the past period of time as independent and dependent variables, respectively, such as the sales and BSR ranks for each period over the past 14 days as independent and dependent variables, respectively.
Specifically, amazon has twenty billions of ASINs (amazon commodity is a special coded identifier, each commodity is different), inventory and BSR ranking of all ASINs is obtained every hour through a large number of crawler servers, and sales and BSR ranking of each ASIN can be calculated every hour. The acquired ASINs can be sampled, and the sampling probability can be determined according to historical sales or ranking of the ASINs. Sample ASINs representing high sales and sample ASINs representing low sales can be extracted from all ASINs, and the higher the sales, the higher the sampling probability, and finally the sample ASINs can be obtained. For example, product A would sell 8 products in one hour, product B would sell 4 products in one hour, and product C would sell 1 product in one hour. At the time of sampling, the probability that product A was drawn was 20%, the probability that product B was drawn was 10%, and the probability that product C was drawn was 2.5%. Based on such a sampling probability model, a library of sample ASIN's may be obtained, and crawlers acquire an inventory and BSR of these sample ASIN's per hour. Subtracting the last hour of inventory from the hour of inventory yields the difference in inventory, which is the sales of the ASIN at that hour.
Regression analysis is a method of predictive modeling technology that studies the relationship between dependent variables (targets) and independent variables (predictors). This technique is used to causal relationships between predictions, time series models, and finding variables. For example, studying the relationship between driver reckless driving and the frequency of occurrence of traffic accidents can be solved by regression analysis. Regression analysis can estimate the relationship between two or more variables and is an important tool for data modeling and analysis. There are many benefits to using regression models. It reveals a significant relationship between dependent and independent variables, effectively measuring the extent of influence of multiple independent variables on a single dependent variable, and allows us to compare the influence of variables measured on different scales, such as the influence of price changes and the influence of the number of promotional campaigns, which can help market researchers, data analysts and data scientists evaluate the best variable set to choose for building a predictive model.
Regression analysis may also be used in the present application to determine the functional relationship between BSR hourly ranking and hourly sales. In the regression prediction model, the dependent variables are continuous, and the independent variables may be continuous or discrete. The nature of regression is linear. Linear regression establishes a relationship between a dependent variable (Y) and one or more independent variables (X) by using a best fit straight line (also referred to as a regression line). The basic expression is as follows: y=a+b x+e, where a is the linear intercept, b is the linear slope, and e is the error term. If the independent variable X is given, the predicted value, i.e., the dependent variable Y, can be calculated from this linear regression expression. The distinction between the unified linear regression and the multiple linear regression is that the multiple linear regression has more than 1 independent variable, whereas the unified linear regression has only 1 independent variable.
Further, sales and BSR rankings of a plurality of sample ASINs corresponding to a plurality of time periods can be input into a regression prediction model, and a functional relationship between the hourly ranking of the BSR and the hourly sales can be determined through the regression prediction model. In the regression model used in the present embodiment, a best-fit straight line is obtained using a least square method (Least Square Method). The least squares method is a common algorithm for fitting regression lines. It calculates the best fit line by minimizing the sum of squares of the vertical errors of each data point and the predicted line. Since the above calculation is the sum of squares of errors, the positive and negative values of errors do not cancel each other. Specifically, the regression prediction model in the present embodiment uses a polynomial regression equation. Polynomial regression may be performed using a polynomial regression fit BSR and an exponentially weighted average y (Polynomial Regression). Corresponding to a regression equation, if the exponent of the argument is greater than 1, it is a polynomial regression equation.
In one embodiment, the polynomial regression equation is: y=a+b x 2 +c*x 3 +d*x 4 +e*x 5 The method comprises the steps of carrying out a first treatment on the surface of the Where y is an exponentially weighted average, x is the sales ranking of the good, and a, b, c, d, e are parameters of a polynomial regression equation. The parameter values of the polynomial regression equation, i.e. a, b, c, d, e, are calculated by continuously substituting into the known sales ranking x.
Further, for the commodity to be predicted, the sales ranking (BSR ranking) of the commodity to be predicted can be directly obtained on the amazon platform, and then the sales ranking is substituted into the polynomial regression equation, so that the exponential weighted average y can be calculated according to the calculated parameter and the substituted sales ranking. Then, according to analysis of the sales and the BSR ranks of the amazon platform commodity in each hour, it can be seen that the ranking of the amazon BSR algorithm is calculated according to an exponential weighted average y of sales x in each hour for n hours, and the smaller the time t, the larger the weight and the exponential relation. Therefore, the sales ranking calculation formula of the Amazon platform can be determined as follows:wherein y is an exponentially weighted average, x n The sales of commodity per hour in n hours are represented, a is an exponential function, and t is a weighting coefficient of the exponential function. It can be seen that the independent variables of the sales ranking calculation formula include sales of the commodity in a preset time period, an exponential weighted average, an exponential function and a weighting coefficient of the exponential function, and the dependent variables are sales ranks. Wherein the weighting coefficient t is used for inputting sales and sales ranking of the commodities into linearityRegression analysis was performed in the model.
Through extensive data analysis, it can be determined that the amazon BSR ranking updated once an hour reflects not only the sales of the current ASIN, but also relates to the rate of change of sales or to historical data of sales. An ASIN has a sales of 10 units in one hour, but drops to a sales of only one unit in the next hour, which does not mean that it will automatically drop to a BSR that only sells one product per hour. Numerous examples indicate that the sales speed, or rate of change of sales, of an ASIN affects its BSR ranking. For example, commodity A ranks 500 in the Amazon Patio, down & Garden category, which sells 8 products in the past hour. At the same time, commodity B ranks 1000 in the same category, with only 4 products sold in the past hour. In the next hour, a sells only 4 products and B sells 8 products. If the sales speed factor is not considered, it appears that both would interchange BSR ranks, i.e., a would rank 1000, and B would rank 500. In practice we found that the BSR for commodity a became around 600, while the BSR for commodity B programmed around 900. This shows that although the hourly sales of the two commodities a and B are "swap locations," their amazon BSR does not fluctuate drastically, which is inherent in the amazon BSR algorithm system taking into account the historical sales of ASIN.
In the actual process, when the amazon seller promotes at a discount price through promotion or coupon, even if sales volume is very high, the BSR of the ASIN does not change very rapidly, which can be considered that the amazon platform prevents the seller from operating the BSR ranking system by promoting the sales speed by using short-term means such as promotion gifts and the like. By trial and error and analysis of large amounts of data, it can be determined that:
1) Although the BSR recalculates every hour, the update time lags the sales time by 2-3 hours;
2) The calculation of the BSR takes more consideration of recent sales, but the historical sales still account for a considerable proportion in the calculation;
3) New goods without sales do not have BSR;
4) Some variants have no own BSR, sales will count on the parent ASIN, but others have own BSR;
5) The BSR depends on the order volume, independent of sales. For example, an order with 90 items has the same effect as an order for one item.
Based on the above, it can be determined that the Amazon BSR algorithm ranking basis is calculated by an exponential weighted average y of sales x per hour for n hours, and the smaller the time t, the larger the weight and the exponential relationship. The sales ranking calculation formula may then be reverse engineered. In this case, the unknown amount y becomes a known amount, and the known amount x becomes an unknown amount, so x can be obtained in the reverse direction from y. Therefore, the calculated index weighted average value can be substituted into a sales ranking calculation formula, namely the sales of the commodity to be predicted in the preset time period can be reversely calculated, namely the sales of the commodity to be predicted in the time period can be determined according to the BSR ranking of the commodity to be predicted.
According to the sales volume estimation method for the Amazon platform commodity, the acquired data are the BSR and inventory information acquired per hour, so that the established model and parameters can be updated once per hour. Whereas in the prior art at most only daily updates can be achieved. Therefore, the model established in the application is more accurate, the iteration and optimization frequency is faster, and sales volume prediction of commodities to be predicted is more accurate. In addition, by updating every hour, not only the influence of the current sales on the BSR but also the influence of the sales speed (the rate of change of sales) over the past 14 days on the BSR was analyzed for correlation, and this influence was not negligible in practice. In addition, the sales volume estimation model established in the application has the advantages that the completeness and the updating frequency of data are obviously improved compared with the prior art, meanwhile, the influence of ASIN sales speed is taken as an important factor for influencing BSR ranking, and the Amazon BSR algorithm is highly restored, so that the obtained estimated sales volume is closer to the actual sales volume, and the prediction accuracy is obviously improved.
In one embodiment, as shown in fig. 2, there is provided a sales volume estimation device for amazon platform commodity, including:
the data statistics module 201 is configured to obtain, for each commodity of the amazon platform, a current commodity sales and sales ranks of the commodity at intervals of a preset time, and determine sales of the commodity according to a variation of the commodity inventory;
the parameter determining module 202 is configured to input sales of a plurality of commodities and corresponding sales ranks into the regression prediction model, and determine parameter values of a polynomial regression equation included in the regression prediction model;
the sales predicting module 203 is configured to obtain sales ranking of the commodity to be predicted; inputting sales ranking of commodities to be predicted into a regression prediction model to determine an exponential weighted average of the commodities to be predicted according to a polynomial regression equation; determining a sales ranking calculation formula of the Amazon platform, wherein independent variables of the sales ranking calculation formula comprise sales of commodities in a preset time period, an exponential weighted average value, an exponential function and a weighting coefficient of the exponential function, and the dependent variables are sales ranks; and carrying out reverse engineering treatment on the sales ranking calculation formula so as to determine the sales of the commodity to be predicted in a preset time period according to the index weighted average value of the commodity to be predicted.
In one embodiment, the weighting coefficients are determined by regression analysis by inputting sales and sales ranking of the good into a linear model.
The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel can be provided with one or more than one kernel, and the sales volume estimation method for the Amazon platform commodity is realized by adjusting kernel parameters.
The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip.
The embodiment of the application provides a storage medium, wherein a program is stored on the storage medium, and the program is executed by a processor to realize the sales predicting method for the Amazon platform commodity.
The embodiment of the application provides a processor for running a program, wherein the program runs to execute the sales prediction method for Amazon platform commodities.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 3. The computer device includes a processor a01, a network interface a02, a memory (not shown) and a database (not shown) connected by a system bus. Wherein the processor a01 of the computer device is adapted to provide computing and control capabilities. The memory of the computer device includes internal memory a03 and nonvolatile storage medium a04. The nonvolatile storage medium a04 stores an operating system B01, a computer program B02, and a database (not shown in the figure). The internal memory a03 provides an environment for the operation of the operating system B01 and the computer program B02 in the nonvolatile storage medium a04. The database of the computer device is used to store inventory and sales ranking of the merchandise. The network interface a02 of the computer device is used for communication with an external terminal through a network connection. The computer program B02, when executed by the processor a01, implements a sales prediction method for amazon platform commodity.
It will be appreciated by those skilled in the art that the structure shown in FIG. 3 is merely a block diagram of some of the structures associated with the present inventive arrangements and is not limiting of the computer device to which the present inventive arrangements may be applied, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
The embodiment of the application provides equipment, which comprises a processor, a memory and a program stored in the memory and capable of running on the processor, wherein the processor realizes the following steps when executing the program: aiming at each commodity of the Amazon platform, acquiring the current commodity inventory and sales ranking of the commodity at intervals of preset time; determining sales of the commodity according to the change of the commodity inventory; inputting sales of a plurality of commodities and corresponding sales ranking into a regression prediction model, and determining parameter values of a polynomial regression equation included in the regression prediction model; acquiring sales ranking of commodities to be predicted; inputting sales ranking of commodities to be predicted into a regression prediction model to determine an exponential weighted average of the commodities to be predicted according to a polynomial regression equation; determining a sales ranking calculation formula of the Amazon platform, wherein independent variables of the sales ranking calculation formula comprise sales of commodities in a preset time period, an exponential weighted average value, an exponential function and a weighting coefficient of the exponential function, and the dependent variables are sales ranks; and carrying out reverse engineering treatment on the sales ranking calculation formula so as to determine the sales of the commodity to be predicted in a preset time period according to the index weighted average value of the commodity to be predicted.
In one embodiment, the weighting coefficients are determined by regression analysis by inputting sales and sales ranking of the good into a linear model.
In one embodiment, the sales ranking calculation formula is:wherein y is an exponentially weighted average, x n The sales of commodity per hour in n hours are represented, a is an exponential function, and t is a weighting coefficient of the exponential function.
In one embodiment, the polynomial regression equation is: y=a+b x 2 +c*x 3 +d*x 4 +e*x 5 The method comprises the steps of carrying out a first treatment on the surface of the Where y is an exponentially weighted average, x is the sales ranking of the good, and a, b, c, d, e are parameters of a polynomial regression equation.
In one embodiment, the preset time and preset period of time are each hour; the step of obtaining the current commodity inventory and sales ranking of the commodities at intervals of preset time comprises the following steps: and acquiring the current commodity inventory of the commodity and the BSR ranking of the commodity every other hour.
In one embodiment, in the sales ranking calculation formula, the more recent sales data, the more weight and exponential changes are presented.
The application also provides a computer program product adapted to perform, when executed on a data processing device, a program initialized with the method steps of: aiming at each commodity of the Amazon platform, acquiring the current commodity inventory and sales ranking of the commodity at intervals of preset time; determining sales of the commodity according to the change of the commodity inventory; inputting sales of a plurality of commodities and corresponding sales ranking into a regression prediction model, and determining parameter values of a polynomial regression equation included in the regression prediction model; acquiring sales ranking of commodities to be predicted; inputting sales ranking of commodities to be predicted into a regression prediction model to determine an exponential weighted average of the commodities to be predicted according to a polynomial regression equation; determining a sales ranking calculation formula of the Amazon platform, wherein independent variables of the sales ranking calculation formula comprise sales of commodities in a preset time period, an exponential weighted average value, an exponential function and a weighting coefficient of the exponential function, and the dependent variables are sales ranks; and carrying out reverse engineering treatment on the sales ranking calculation formula so as to determine the sales of the commodity to be predicted in a preset time period according to the index weighted average value of the commodity to be predicted.
In one embodiment, the weighting coefficients are determined by regression analysis by inputting sales and sales ranking of the good into a linear model.
In one embodiment, the sales ranking calculation formula is:wherein y is an exponentially weighted average, x n The sales of commodity per hour in n hours are represented, a is an exponential function, and t is a weighting coefficient of the exponential function.
In one embodiment, the polynomial regression equation is: y=a+b x 2 +c*x 3 +d*x 4 +e*x 5 The method comprises the steps of carrying out a first treatment on the surface of the Wherein Y is an exponentially weighted average, x is the sales ranking of the good, and a, b, c, d, e are parameters of a polynomial regression equation.
In one embodiment, the preset time and preset period of time are each hour; the step of obtaining the current commodity inventory and sales ranking of the commodities at intervals of preset time comprises the following steps: and acquiring the current commodity inventory of the commodity and the BSR ranking of the commodity every other hour.
In one embodiment, in the sales ranking calculation formula, the more recent sales data, the more weight and exponential changes are presented.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.
Claims (7)
1. A sales volume estimation method for amazon platform commodity, the method comprising:
aiming at each commodity of the Amazon platform, acquiring the current commodity inventory and sales ranking of the commodity at intervals of preset time;
determining sales of the commodity according to the variation of the commodity inventory;
inputting sales of a plurality of commodities and corresponding sales ranking into a regression prediction model, and determining parameter values of a polynomial regression equation included in the regression prediction model;
acquiring sales ranking of commodities to be predicted;
inputting sales ranking of the commodities to be predicted into the regression prediction model to determine an exponentially weighted average of the commodities to be predicted according to the polynomial regression equation;
determining a sales ranking calculation formula of the Amazon platform, wherein independent variables of the sales ranking calculation formula comprise sales of commodities in a preset time period, an exponential weighted average value, an exponential function and a weighting coefficient of the exponential function, and the dependent variables are sales ranks;
performing reverse engineering treatment on the sales ranking calculation formula to determine sales of the commodity to be predicted in a preset time period according to an exponential weighted average value of the commodity to be predicted;
the sales ranking calculation formula is as follows:
wherein y is an exponentially weighted average, x n Representing sales of goods per hour over n hours, a being an exponential function, t being an exponential functionWeighting coefficients of (2);
the polynomial regression equation is:
y＝a+b*x 2 +c*x 3 +d*x 4 +e*x 5 ；
wherein y is an exponentially weighted average, x is sales ranking of commodities, and a, b, c, d and e are parameters of the polynomial regression equation;
the preset time and the preset time period are each hour; the step of obtaining the current commodity inventory and sales ranking of the commodities at intervals of preset time comprises the following steps:
and acquiring the current commodity inventory of the commodity and the BSR ranking of the commodity every other hour.
2. The method of claim 1, wherein the weighting coefficients are determined by regression analysis by inputting sales and sales ranking of the commodity into a linear model.
3. The method of claim 1, wherein the sales data in the sales ranking calculation formula, the more recent the time, the more heavily weighted, and exhibits an exponential change.
4. A sales predicting apparatus using the sales predicting method for amazon platform commodity according to claim 1, comprising:
the data statistics module is used for acquiring the current commodity inventory and sales ranking of the commodities at intervals of preset time for each commodity of the Amazon platform, and determining the sales of the commodities according to the variation of the commodity inventory;
the parameter determining module is used for inputting sales of a plurality of commodities and corresponding sales ranking into the regression prediction model and determining parameter values of a polynomial regression equation included in the regression prediction model;
the sales predicting module is used for obtaining sales ranking of the commodities to be predicted; inputting sales ranking of the commodities to be predicted into the regression prediction model to determine an exponentially weighted average of the commodities to be predicted according to the polynomial regression equation; determining a sales ranking calculation formula of the Amazon platform, wherein independent variables of the sales ranking calculation formula comprise sales of commodities in a preset time period, an exponential weighted average value, an exponential function and a weighting coefficient of the exponential function, and the dependent variables are sales ranks; and carrying out reverse engineering treatment on the sales ranking calculation formula so as to determine the sales of the commodity to be predicted in a preset time period according to the index weighted average value of the commodity to be predicted.
5. The sales prediction apparatus according to claim 4, wherein the weighting coefficient is determined by inputting sales and sales ranks of the commodity into a linear model for regression analysis.
6. A machine-readable storage medium having instructions stored thereon, which when executed by a processor, cause the processor to be configured to perform the sales prediction method for amazon platform commodity according to any one of claims 1 to 3.
7. A processor configured to perform the sales volume estimation method for amazon platform commodity according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110822098.9A CN113538055B (en) | 2021-07-20 | 2021-07-20 | Sales volume estimation method, device and processor for Amazon platform commodity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110822098.9A CN113538055B (en) | 2021-07-20 | 2021-07-20 | Sales volume estimation method, device and processor for Amazon platform commodity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113538055A CN113538055A (en) | 2021-10-22 |
| CN113538055B true CN113538055B (en) | 2023-11-07 |
Family
ID=78100557
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110822098.9A Active CN113538055B (en) | 2021-07-20 | 2021-07-20 | Sales volume estimation method, device and processor for Amazon platform commodity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113538055B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0581301A (en) * | 1991-09-19 | 1993-04-02 | Nec Software Ltd | Management reference value determining system for stock management system |
| JP2001216372A (en) * | 2000-01-31 | 2001-08-10 | Toshiba Corp | Device and method for predicting sales, and storage medium |
| JP2007018216A (en) * | 2005-07-07 | 2007-01-25 | Nec Corp | Reasonable inventory quantity calculation system and reasonable inventory quantity calculation method |
| GB201505033D0 (en) * | 2014-11-24 | 2015-05-06 | Inst Information Industry | Product sales forcasting system, method and non-transitory computer readable storage medium thereof |
| CN104700152A (en) * | 2014-10-22 | 2015-06-10 | 浙江中烟工业有限责任公司 | Method for predicting tobacco sales volumes by means of fusing seasonal sales information with search behavior information |
| CN105184618A (en) * | 2015-10-20 | 2015-12-23 | 广州唯品会信息科技有限公司 | Commodity individual recommendation method for new users and system |
| CN109509030A (en) * | 2018-11-15 | 2019-03-22 | 北京旷视科技有限公司 | Method for Sales Forecast method and its training method of model, device and electronic system |
| CN110782266A (en) * | 2018-07-31 | 2020-02-11 | 阿里巴巴集团控股有限公司 | Commodity sales prediction and model training method, device and equipment |
| CN111178624A (en) * | 2019-12-26 | 2020-05-19 | 浙江大学 | Method for predicting new product demand |
| CN113553540A (en) * | 2020-04-24 | 2021-10-26 | 株式会社日立制作所 | Commodity sales prediction method |
| CN115358651A (en) * | 2021-05-17 | 2022-11-18 | 星巴克公司 | Method, system and computer readable storage medium for automatic replenishment of retail store |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3992890A1 (en) * | 2020-10-28 | 2022-05-04 | Tata Consultancy Services Limited | Method and system for forecasting sales based on n-gram model |
-
2021
- 2021-07-20 CN CN202110822098.9A patent/CN113538055B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0581301A (en) * | 1991-09-19 | 1993-04-02 | Nec Software Ltd | Management reference value determining system for stock management system |
| JP2001216372A (en) * | 2000-01-31 | 2001-08-10 | Toshiba Corp | Device and method for predicting sales, and storage medium |
| JP2007018216A (en) * | 2005-07-07 | 2007-01-25 | Nec Corp | Reasonable inventory quantity calculation system and reasonable inventory quantity calculation method |
| CN104700152A (en) * | 2014-10-22 | 2015-06-10 | 浙江中烟工业有限责任公司 | Method for predicting tobacco sales volumes by means of fusing seasonal sales information with search behavior information |
| GB201505033D0 (en) * | 2014-11-24 | 2015-05-06 | Inst Information Industry | Product sales forcasting system, method and non-transitory computer readable storage medium thereof |
| CN105184618A (en) * | 2015-10-20 | 2015-12-23 | 广州唯品会信息科技有限公司 | Commodity individual recommendation method for new users and system |
| CN110782266A (en) * | 2018-07-31 | 2020-02-11 | 阿里巴巴集团控股有限公司 | Commodity sales prediction and model training method, device and equipment |
| CN109509030A (en) * | 2018-11-15 | 2019-03-22 | 北京旷视科技有限公司 | Method for Sales Forecast method and its training method of model, device and electronic system |
| CN111178624A (en) * | 2019-12-26 | 2020-05-19 | 浙江大学 | Method for predicting new product demand |
| CN113553540A (en) * | 2020-04-24 | 2021-10-26 | 株式会社日立制作所 | Commodity sales prediction method |
| CN115358651A (en) * | 2021-05-17 | 2022-11-18 | 星巴克公司 | Method, system and computer readable storage medium for automatic replenishment of retail store |
Non-Patent Citations (15)
| Title |
|---|
| prediction of merchandise sales on e-commerce platforms based on data mining and deep learning;xiaoting yin;《scientific programming》(11);全文 * |
| 吴琳洁 ; 郑钰洁 ; 范云飞 ; .新零售目标产品的精准需求分析与预测.科技与创新.2020,(第18期),全文. * |
| 基于R语言的煤炭产品销售预测实证研究;黄宇达;王换换;王迤冉;;计算机与数字工程(第01期);全文 * |
| 基于WaveNet-LSTM网络的商品销量预测研究;蒋文武;《中国优秀硕士学位论文全文数据库信息科技辑》(第2期);第I138-1037页 * |
| 基于综合排序视角的在线商品销量的影响因素;李清;《天津职业技术师范大学学报》(第3期);全文 * |
| 基于网络搜索数据的品牌汽车销量预测研究;谢天保;崔田;;信息技术与网络安全(第08期);全文 * |
| 张蔚虹 ; 刘立 ; .指数平滑法在全面预算管理销售预算中的应用.财会通讯(理财版).2007,(第05期),全文. * |
| 指数平滑法在全面预算管理销售预算中的应用;张蔚虹;刘立;;财会通讯(理财版)(第05期);全文 * |
| 新零售目标产品的精准需求分析与预测;吴琳洁;郑钰洁;范云飞;;科技与创新(第18期);全文 * |
| 王伟 ; 王洪伟 ; .面向竞争力的特征比较网络:情感分析方法.管理科学学报.2016,(第09期),全文. * |
| 谢天保 ; 崔田 ; .基于网络搜索数据的品牌汽车销量预测研究.信息技术与网络安全.2018,(第08期),全文. * |
| 车企利用机动车销售发票统计车辆销量的方法;陈涛;;时代汽车(第07期);全文 * |
| 陈涛 ; .车企利用机动车销售发票统计车辆销量的方法.时代汽车.2019,(第07期),全文. * |
| 面向竞争力的特征比较网络:情感分析方法;王伟;王洪伟;;管理科学学报(第09期);全文 * |
| 黄宇达 ; 王换换 ; 王迤冉 ; .基于R语言的煤炭产品销售预测实证研究.计算机与数字工程.2017,(第01期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113538055A (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8631040B2 (en) | Computer-implemented systems and methods for flexible definition of time intervals | |
| US11195193B2 (en) | System and method for price testing and optimization | |
| Wu | Quantity flexibility contracts under Bayesian updating | |
| US11790383B2 (en) | System and method for selecting promotional products for retail | |
| US20170116624A1 (en) | Systems and methods for pricing optimization with competitive influence effects | |
| CA2429189A1 (en) | Promotion pricing system and method | |
| WO2002082211A9 (en) | Assortment decisions | |
| US20150235239A1 (en) | Predicting demand of a newly introduced short lifecycle product within an assortment | |
| US11373199B2 (en) | Method and system for generating ensemble demand forecasts | |
| CN110189164B (en) | Commodity-store recommendation scheme based on information entropy measurement and random feature sampling | |
| US7996254B2 (en) | Methods and systems for forecasting product demand during promotional events using a causal methodology | |
| US20150227951A1 (en) | Forecasting holiday weekly sales profiles | |
| Brander et al. | Lot sizes in a capacity constrained facility—a simulation study of stationary stochastic demand | |
| US20090327027A1 (en) | Methods and systems for transforming logistic variables into numerical values for use in demand chain forecasting | |
| CN111861605A (en) | Business object recommendation method | |
| CA2881812A1 (en) | System, method and computer program for forecasting residual values of a durable good over time | |
| CN111127074B (en) | Data recommendation method | |
| US10467654B2 (en) | Forecasting customer channel choice using cross-channel loyalty | |
| JP2009251779A (en) | Sales estimation system, method and program | |
| GÜR ALI | Driver moderator method for retail sales prediction | |
| CN114004680A (en) | Order quantity information processing method, device and storage medium | |
| JP6193817B2 (en) | Demand prediction apparatus and program | |
| Anggrainingsih et al. | Forecasting The Number of Patients at RSUD Sukoharjo Using Double Exponential Smoothing Holt | |
| CN113538055B (en) | Sales volume estimation method, device and processor for Amazon platform commodity | |
| WO2023120126A1 (en) | Unit-sales prediction system and unit-sales prediction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |