CN113095893A

Movatterモバイル変換

Info

Publication number: CN113095893A
Application number: CN202110507191.0A
Authority: CN
Inventors: 陈力; 陈松蹊
Original assignee: Beijing Jingdong Zhenshi Information Technology Co Ltd
Current assignee: Beijing Jingdong Zhenshi Information Technology Co Ltd
Priority date: 2021-05-10
Filing date: 2021-05-10
Publication date: 2021-07-09

Abstract

Translated fromChinese

本发明公开了一种确定物品销量的方法和装置，涉及仓储物流技术领域。其中，该方法包括：获取目标物品对应的多个相似物品的历史销量数据；根据目标物品和所述多个相似物品的历史销量数据计算目标物品和所述多个相似物品的相似度；根据目标物品和所述多个相似物品的相似度计算所述多个相似物品的权重；根据所述多个相似物品的权重和所述多个相似物品的历史销量数据，确定目标物品的销量预测值。通过以上步骤，能够提高新品销量预测结果的精确度和可靠性，解决了现有技术中由于新品的历史销售信息较少导致销量预测结果的精确度低、可靠性差、甚至无法预测等问题，进而引导人们基于销量预测结果及时、准确地进行诸如补货等业务操作。

The invention discloses a method and a device for determining the sales volume of items, and relates to the technical field of warehousing and logistics. Wherein, the method includes: obtaining historical sales data of a plurality of similar items corresponding to the target item; calculating the similarity between the target item and the plurality of similar items according to the historical sales data of the target item and the plurality of similar items; The similarity between the item and the multiple similar items calculates the weight of the multiple similar items; according to the weight of the multiple similar items and the historical sales data of the multiple similar items, the sales forecast value of the target item is determined. Through the above steps, the accuracy and reliability of the sales forecast results of new products can be improved, and the problems of low accuracy, poor reliability, and even unpredictability of sales forecast results in the prior art due to the lack of historical sales information of new products are solved. Guide people to conduct timely and accurate business operations such as replenishment based on sales forecast results.

Description

Method and device for determining sales of articles

Technical Field

The invention relates to the technical field of warehouse logistics, in particular to a method and a device for determining the sales volume of articles.

Background

The method has important significance in predicting the goods sales volume in the actual business scenes of warehouse management of an e-commerce platform, intelligent retail of stores and the like. For example, when the e-commerce platform performs warehousing management, the sales volume of the articles in a future period of time needs to be predicted, and the replenishment volume is determined according to the sales volume prediction result and the inventory condition of the articles, so as to guide relevant personnel to perform replenishment in time.

When an article initially enters the market, the article sales prediction at this time is always difficult because there is no historical sales data or only a small amount of historical sales data. In the prior art, the following item sales prediction methods are mainly available: firstly, the sales of the goods are predicted by directly using a conventional time series prediction model, such as ARIMA, Exponental smoothening and other models; secondly, an article prediction method based on Logistic, Gompertz and other growth models; third, a user behavior feature based item prediction method, such as a Bass model based item sales prediction method.

In the process of implementing the invention, the inventor of the invention finds that the prior article sales prediction method has the following problems: firstly, a conventional time series prediction model has certain requirements on the historical sales data volume of an article, and the article has less sales information in the initial marketing stage, so that the prediction effect is poor or the time series prediction model cannot be fitted at all; secondly, the growth models such as Logistic and Gompertz are only suitable for predicting new products with gradually increased sales after being listed, but in practice, the sales of many new products are gradually reduced because the new products are not accepted by the market after being listed, and the accuracy of the sales prediction result obtained by directly adopting the historical sales information of the products to fit the growth models such as Logistic and Gompertz is poor; thirdly, the item prediction method based on the user behavior characteristics is more suitable for predicting sales of items which are already on the market for a period of time, and if the item prediction method is used for predicting new sales, the reliability and the accuracy of the sales prediction result are poor.

Disclosure of Invention

In view of the above, the present invention provides a method and an apparatus for determining an article sales amount, which can improve the accuracy and reliability of a new article sales amount prediction result, and solve the problems in the prior art that the accuracy of the sales amount prediction result is low, the reliability is poor, and even the prediction cannot be performed due to less historical sales information of the new article, so as to guide people to perform timely and accurately business operations such as replenishment based on the sales amount prediction result.

To achieve the above object, according to a first aspect of the present invention, there is provided a method of determining sales of an article.

The method for determining the sales volume of the article is executed by the electronic equipment and comprises the following steps: acquiring historical sales data of a plurality of similar articles corresponding to the target article; wherein the plurality of similar items are sold earlier in time than the target item; calculating the similarity of the target item and the similar items according to the historical sales data of the target item and the similar items; calculating the weights of the similar items according to the similarity of the target item and the similar items; and determining a sales predicted value of the target item according to the weights of the similar items and historical sales data of the similar items.

Optionally, the calculating the similarity between the target item and the plurality of similar items according to the historical sales data of the target item and the plurality of similar items comprises: constructing a corresponding feature vector according to the sales volume of the target object in at least one historical stage; for each of the plurality of similar items, constructing a corresponding feature vector according to the sales volume of the similar item in the at least one historical stage; and calculating the similarity of the target object and the similar object according to the feature vector of the target object and the feature vector of the similar object.

Optionally, the calculating the similarity between the target item and the plurality of similar items according to the historical sales data of the target item and the plurality of similar items comprises: calculating the growth rate of the target item in at least one historical stage according to the historical sales data of the target item; constructing a corresponding feature vector according to the growth rate of the target object in at least one historical stage; for each of the plurality of similar items, constructing a corresponding feature vector according to the growth rate of the similar item in the at least one historical stage; and calculating the similarity of the target object and the similar object according to the feature vector of the target object and the feature vector of the similar object.

Optionally, the calculating the similarity between the target item and the similar item according to the feature vector of the target item and the feature vector of the similar item includes: and calculating the distance between the feature vector of the target object and the feature vector of the similar object based on an Euclidean distance or cosine similarity formula, and taking the distance as the similarity of the target object and the similar object.

Optionally, the calculating the weights of the plurality of similar items according to the similarity between the target item and the plurality of similar items includes: calculating the sum of the similarity of the target article and the similar articles; and for each of the similar items, taking the ratio of the similarity of the target item and the similar item to the sum of the similarities as the weight of the similar item.

Optionally, the determining a sales forecast value of the target item according to the weights of the similar items and historical sales data of the similar items comprises: for the historical stage of the target item, carrying out weighted summation on the maximum sales volume ratio of the plurality of similar items in the corresponding stage according to the weights of the plurality of similar items to obtain the maximum sales volume ratio predicted value of the target item in the historical stage; determining a maximum sales volume predicted value of the target article according to the sales volume of the target article in the historical stage and the maximum sales volume ratio predicted value; for the stage to be predicted of the target article, carrying out weighted summation on the maximum sales volume ratio of the plurality of similar articles in the corresponding stage according to the weights of the plurality of similar articles to obtain the maximum sales volume ratio predicted value of the target article in the stage to be predicted; and determining the sales volume predicted value of the target article in the stage to be predicted according to the maximum sales volume predicted value of the target article and the maximum sales volume ratio predicted value of the target article in the stage to be predicted.

Optionally, the method further comprises: and before calculating the similarity of the target item and the similar items according to the historical sales data of the target item and the similar items, aligning the historical sales data of the similar items according to the historical stage of sales.

To achieve the above object, according to a second aspect of the present invention, there is provided an apparatus for determining the sales volume of an article.

The device for determining the sales volume of an object is arranged in an electronic device and comprises: the acquisition module is used for acquiring historical sales data of a plurality of similar articles corresponding to the target article; wherein the plurality of similar items are sold earlier in time than the target item; the weight determining module is used for calculating the similarity between the target object and the similar objects according to the historical sales data of the target object and the similar objects; calculating the weights of the similar items according to the similarity of the target item and the similar items; and the sales amount determining module is used for determining the sales amount predicted value of the target item according to the weights of the similar items and the historical sales amount data of the similar items.

To achieve the above object, according to a third aspect of the present invention, there is provided an electronic apparatus.

The electronic device of the present invention includes: one or more processors; and storage means for storing one or more programs; when executed by the one or more processors, cause the one or more processors to implement the method of determining an item sales volume of the present invention.

To achieve the above object, according to a fourth aspect of the present invention, there is provided a computer-readable medium.

The computer-readable medium of the invention, on which a computer program is stored which, when being executed by a processor, carries out the method of the invention for determining an amount of sales of an item.

One embodiment of the above invention has the following advantages or benefits: the method comprises the steps of obtaining historical sales data of a plurality of similar articles corresponding to a target article by electronic equipment, calculating the similarity between the target article and the similar articles according to the historical sales data of the target article and the similar articles, calculating the weights of the similar articles according to the similarity between the target article and the similar articles, and determining a sales predicted value of the target article according to the weights of the similar articles and the historical sales data of the similar articles, so that the accuracy and reliability of a new sales predicted result can be improved, the problems that in the prior art, due to the fact that the historical sales information of the new article is less, the accuracy of the sales predicted result is low, the reliability is poor, even the sales predicted result cannot be obtained are solved, and further people are guided to timely and accurately carry out business operations such as replenishment based on the sales predicted result.

Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.

Drawings

The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:

FIG. 1 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;

FIG. 2 is a schematic main flow chart of a method of determining an amount of sales of an item according to a first embodiment of the present invention;

FIG. 3 is a schematic main flow chart of a method of determining an article sales volume according to a second embodiment of the present invention;

FIG. 4 is a schematic diagram of the main modules of an apparatus for determining the sales volume of an article according to a third embodiment of the present invention;

FIG. 5 is a schematic diagram of the main modules of an apparatus for determining the sales volume of an article according to a fourth embodiment of the present invention;

FIG. 6 is a schematic block diagram of a computer system suitable for use with the electronic device to implement an embodiment of the invention.

Detailed Description

Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

It should be noted that the embodiments and technical features of the embodiments of the present invention may be combined with each other without affecting the implementation of the present invention.

Fig. 1 illustrates anexemplary system architecture 100 of a method of determining an item sales volume or an apparatus for determining an item sales volume to which embodiments of the present invention may be applied.

As shown in fig. 1, thesystem architecture 100 may include

terminal devices

101, 102, 103, anetwork 104, and aserver 105. Thenetwork 104 serves as a medium for providing communication links between the

terminal devices

101, 102, 103 and theserver 105.Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The user may use the

terminal devices

101, 102, 103 to interact with theserver 105 via thenetwork 104 to receive or send messages or the like. Various communication client applications, such as a warehouse management application, a shopping application, a web browser application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like, may be installed on the

terminal devices

101, 102, 103.

The

terminal devices

101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

Theserver 105 may be a server providing various services, such as a background management server providing support for a warehouse management type application browsed by a user using the

terminal device

101, 102, 103. For example, the backend management server may process a data processing request and the like sent by the terminal device through the network, and feed back a processing result to the terminal device.

It should be noted that the method for determining the sales volume of the item provided by the embodiment of the present invention is generally executed by theserver 105, and accordingly, the apparatus for determining the sales volume of the item is generally disposed in theserver 105.

It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Fig. 2 is a main flow chart illustrating a method of determining an article sales amount according to a first embodiment of the present invention. The method for determining the sales volume of the article is executed by the electronic equipment. As shown in fig. 2, the method for determining the sales volume of an article according to an embodiment of the present invention includes:

step S201: and acquiring historical sales data of a plurality of similar articles corresponding to the target article.

Wherein the similar item is sold earlier than the target item. Preferably, the length of the historical sales data of the similar items is greater than 12 periods and is greater than or equal to the sum of the predicted period number of the target item and the historical sales period number of the target item. For example, assuming that the target item has historical sales data for 6 days and needs to predict sales for the next 9 days, the historical sales data for similar items should be greater than or equal to 15 days. In specific implementation, the selling stages may be divided by month, week, etc. after the goods are listed, for example, the first month after the goods are listed is the first selling stage (or referred to as the first stage), the second month after the goods are listed is the second selling stage (or referred to as the second stage), and the third month after the goods are listed is the third selling stage (or referred to as the third stage).

In an alternative embodiment, similar items corresponding to the target item are selected as items that belong to the same classification category as the target item and that are sold earlier in time than the target item. In this alternative embodiment, the item information storage system may be queried according to the classification category information of the target item to query for items belonging to the same classification category as the target item, and then a plurality of items having an initial sale time earlier than that of the target item may be screened out therefrom, and the historical sales data of these screened items may be used as the historical sales data of a plurality of similar items corresponding to the target item.

In another alternative embodiment, the feature vector of the target item and the feature vector of the item which has been listed for a long time can be constructed according to the item attribute data, a similar item corresponding to the target item is screened from the items which have been listed for a long time by combining a model such as KNN (K nearest neighbor), and then historical sales data of the similar item corresponding to the target item is obtained.

Step S202: calculating the similarity of the target item and the similar items according to the historical sales data of the target item and the similar items; and calculating the weights of the similar items according to the similarity of the target item and the similar items.

In an alternative embodiment, the calculating the similarity between the target item and the plurality of similar items according to the historical sales data of the target item and the plurality of similar items comprises: constructing a corresponding feature vector according to the sales volume of the target object in at least one historical stage; for each of the plurality of similar items, constructing a corresponding feature vector according to the sales volume of the similar item in the at least one historical stage; and calculating the similarity of the target object and the similar object according to the feature vector of the target object and the feature vector of the similar object.

For example, assuming that the target item has historical sales data of a total of 4 periods (or 4 historical stages) from the first period to the fourth period, a feature vector corresponding to the target item may be constructed according to the sales data of the 4 periods, for example, a feature vector with a dimension of 4 is constructed by using the sales data of each period or data obtained by normalizing the sales data of each period as one dimension of the vector; assuming that similar articles share historical sales data of 16 periods in total from the first period to the 16 th period, a feature vector corresponding to the similar articles can be constructed according to the sales data of the previous 4 periods of the similar articles, then the distance between the feature vector of the target article and the feature vector of the similar articles is calculated, and the distance is used as the similarity between the target article and the similar articles.

In another alternative embodiment, said calculating the similarity of the target item and the plurality of similar items according to the historical sales data of the target item and the plurality of similar items comprises: calculating the growth rate of the target item in at least one historical stage according to the historical sales data of the target item; constructing a corresponding feature vector according to the growth rate of the target object in at least one historical stage; for each of the plurality of similar items, constructing a corresponding feature vector according to the growth rate of the similar item in the at least one historical stage; and calculating the similarity of the target object and the similar object according to the feature vector of the target object and the feature vector of the similar object.

For example, assuming that the target item has historical sales data of a total of 4 stages (or 4 historical stages) from the first stage to the fourth stage, the sales increase rate of the target item in the second stage compared to the first stage, the sales increase rate of the target item in the third stage compared to the second stage, and the sales increase rate of the target item in the fourth stage compared to the third stage may be calculated according to the sales data of the 4 stages, and then a feature vector corresponding to the target item is constructed according to the three growth rates, for example, a feature vector with a dimension of 3 is constructed by taking each growth rate as one dimension of the vector; assuming that similar articles share historical sales data of 16 periods in total from the first period to the 16 th period, the sales increase rate of the similar articles in the second period compared with the first period, the sales increase rate of the similar articles in the third period compared with the second period, and the sales increase rate of the similar articles in the fourth period compared with the third period can be calculated according to the sales data of the previous 4 periods of the similar articles, then a feature vector corresponding to the similar articles is constructed according to the three increase rates, then the distance between the feature vector of the target article and the feature vector of the similar articles is calculated, and the distance is used as the similarity between the target article and the similar articles.

Further, in the two above-mentioned optional embodiments, the calculating the similarity between the target item and the similar item according to the feature vector of the target item and the feature vector of the similar item includes: and calculating the distance between the feature vector of the target object and the feature vector of the similar object based on formulas such as Euclidean distance, cosine similarity or Hamming distance, and taking the distance as the similarity between the target object and the similar object.

For example, after obtaining the similarity between the target item and each similar item, the weight of each similar item may be calculated in an Inverse Distance Weighting (IDW) manner, specifically including: calculating the sum of the similarity of the target article and the similar articles; and for each of the similar items, taking the ratio of the similarity of the target item and the similar item to the sum of the similarities as the weight of the similar item. For example, if the target article includes only two similar articles, that is, article B and article C, and the similarity between article B and the target article is 0.73 and the similarity between article C and the target article is 0.41, the weight calculation formula for article B is 0.73/(0.73+0.41) and the weight calculation formula for article C is 0.41/(0.73+ 0.41).

Step S203: and determining a sales predicted value of the target item according to the weights of the similar items and historical sales data of the similar items.

In an alternative embodiment, step S203 comprises: and for the stage to be predicted of the target item, carrying out weighted summation on the historical sales of the similar items in the corresponding stage according to the weights of the similar items so as to obtain the sales prediction value of the target item in the stage to be predicted. For example, assuming that the target items share the historical sales data of the first to fourth periods, when predicting the sales of the target items in the fifth period, the historical sales of the similar items in the 5 th period are weighted and summed according to the weights of the similar items corresponding to the target items, and the weighted and summed result is used as the sales predicted value of the target items in the 5 th period.

In another alternative embodiment, step S203 comprises: for the historical stage of the target item, carrying out weighted summation on the maximum sales volume ratio of the plurality of similar items in the corresponding stage according to the weights of the plurality of similar items to obtain the maximum sales volume ratio predicted value of the target item in the historical stage; determining a maximum sales volume predicted value of the target article according to the sales volume of the target article in the historical stage and the maximum sales volume ratio predicted value; for the stage to be predicted of the target article, carrying out weighted summation on the maximum sales volume ratio of the plurality of similar articles in the corresponding stage according to the weights of the plurality of similar articles to obtain the maximum sales volume ratio predicted value of the target article in the stage to be predicted; and determining the sales volume predicted value of the target article in the stage to be predicted according to the maximum sales volume predicted value of the target article and the maximum sales volume ratio predicted value of the target article in the stage to be predicted. The maximum sales volume ratio is the ratio of the sales volume of the article in the current stage to the maximum sales volume, and the maximum sales volume ratio reflects whether the product is in a high sales volume stage or a low sales volume stage within a certain period of time.

In the above optional embodiment, the maximum sales volume ratio of the target item is obtained by performing weighted summation on the maximum sales volume ratio of the similar items according to the weights of the similar items, and the sales volume predicted value of the target item in each prediction stage is determined based on the change of the maximum sales volume ratio predicted value, so that the adverse effect on the sales volume prediction result due to the large sales volume difference among the similar items can be eliminated, and the accuracy and reliability of the sales volume prediction result are improved.

In the embodiment of the invention, the accurate and reliable prediction of the target article sales volume is realized through the steps, and the method is particularly suitable for predicting the new article sales volume. Compared with the prior art, the method and the device can improve the accuracy and reliability of the new product sales prediction result through the steps, solve the problems that the sales prediction result is low in accuracy, poor in reliability and even impossible to predict due to the fact that the historical sales information of the new product is less in the prior art, and further guide people to timely and accurately conduct business operations such as replenishment and the like based on the sales prediction result.

Fig. 3 is a main flow chart illustrating a method of determining an article sales amount according to a second embodiment of the present invention. The method of the embodiment of the invention is executed by the electronic equipment. As shown in fig. 3, the method for determining the sales volume of an article according to an embodiment of the present invention includes:

step S301: and determining a plurality of similar items corresponding to the target item.

Wherein the similar item is sold earlier than the target item. Preferably, the length of the historical sales data of the similar items is greater than 12 periods and is greater than or equal to the sum of the predicted period number of the target item and the historical sales period number of the target item. For example, assuming that the target item has 4-stage historical sales data and the sales volume of the subsequent 12-stage needs to be predicted, the length of the historical sales volume data of the similar items is greater than or equal to 16 stages. In specific implementation, the selling stages may be divided by month, week, etc. after the goods are listed, for example, the first month after the goods are listed is the first selling stage (or referred to as the first stage), the second month after the goods are listed is the second selling stage (or referred to as the second stage), and the third month after the goods are listed is the third selling stage (or referred to as the third stage).

In an alternative embodiment, similar items corresponding to the target item are selected as items that belong to the same classification category as the target item and that are sold earlier in time than the target item. For example, assuming that the historical sales period number of the target item has 4 total periods and the sales volume of the subsequent 12 periods needs to be predicted, a plurality of similar items corresponding to items of which the target item belongs to the same classification category and the historical sales period number is equal to or greater than 16 may be used as the target item.

In another alternative embodiment, the feature vectors of the target item and the feature vectors of the items which have been listed for a longer time can be constructed according to the item attribute data, and similar items corresponding to the target item can be screened from the items which have been listed for a longer time by combining a KNN (K nearest neighbor) model and the like.

Step S302: obtaining historical sales data of the plurality of similar items.

After a plurality of similar items corresponding to the target item are determined according to step S301, historical sales data of the similar items are acquired. For example, assuming that the historical sales of the target item total 4 stages, and the sales of the subsequent 12 stages need to be predicted, the historical sales data of these similar items from the first stage to the sixteenth stage can be obtained.

Step S303: the historical sales data for the plurality of similar items is aligned according to the historical stage of sales.

Considering that the time of entering the market of each similar item is likely to be different, in order to facilitate the subsequent sales prediction based on the historical sales data of the target item and the similar items, the historical sales data of each similar item needs to be aligned according to the historical stage of sales. For example, when the sales stages are divided by month, the number of the historical sales data in the first month of listing each similar item may be set to 1, the number of the historical sales data in the second month of listing each similar item may be set to 2, and so on, the number of the historical sales data in the 16 th month of listing each similar item may be set to 16, so as to achieve the purpose of data alignment.

Step S304: and calculating the maximum sales volume ratio of the similar items in a plurality of historical stages according to the historical sales volume data of the similar items.

The maximum sales volume ratio is the ratio of the sales volume of the article in the current stage to the maximum sales volume, and the maximum sales volume ratio reflects whether the product is in a high sales volume stage or a low sales volume stage within a certain period of time. For example, assuming that the target item is item a, which shares the historical sales data of the first to fourth periods, the sales of the subsequent 12 periods need to be predicted, and 90 similar items are found through step S301 and step S302. Taking the article B and the article C among the 90 similar articles as an example, table 1 shows the maximum sales volume ratio of the article B and the article C in each period, i.e., each historical stage. Wherein, the sale amount of the article B in the first month on the market is the highest in 16 days, so that the ratio of the sale amount of the article B to the maximum sale amount of the article B, namely the maximum sale amount of the first day, is 100 percent, and the ratio of the sale amount of the article B in the second day to the maximum sale amount of the article B, namely the maximum sale amount of the second day, is 50.4 percent. Similarly, for item C, its sales at stage eight was the highest at stage 16, so its maximum sales at stage eight accounted for 100%.

TABLE 1

Step S305: calculating the similarity of the target article and the similar articles according to the historical sales data of the target article and the similar articles; and calculating the weight of the similar articles according to the similarity.

Illustratively, in step S305, the calculating the similarity between the target item and the similar item according to the historical sales data of the target item and the similar item includes: calculating the growth rate of the target item in at least one historical stage according to the historical sales data of the target item; constructing a corresponding feature vector according to the growth rate of the target object in at least one historical stage; for each of the plurality of similar items, constructing a corresponding feature vector according to the growth rate of the similar item in the at least one historical stage; and calculating the similarity of the target object and the similar object according to the feature vector of the target object and the feature vector of the similar object. In specific implementation, the distance between the feature vector of the target object and the feature vector of the similar object can be calculated based on the formula such as the euclidean distance, the cosine similarity or the hamming distance, and the distance is used as the similarity between the target object and the similar object.

Further, after obtaining the similarity between the target item and each similar item, the weight of each similar item may be calculated in an Inverse Distance Weighting (IDW) manner, specifically including: calculating the sum of the similarity of the target article and the similar articles; and for each of the similar items, taking the ratio of the similarity of the target item and the similar item to the sum of the similarities as the weight of the similar item.

For example, assuming that the target item is item a, which shares the historical sales data of the first to fourth periods (as shown in table 2), the sales of the subsequent 12 periods needs to be predicted, and 90 similar items are found through step S301 and step S302. Taking similar items shown in table 2 (i.e., items B and C) as an example, an approximate calculation process of the weights of item B and item C is given below. Specifically, for the article a, the increase rate of the sales of the second phase compared with the first phase is (3958-; similarly, for item B, the sales increase rate of the second stage compared with the first stage, the sales increase rate of the third stage compared with the second stage, and the sales increase rate of the fourth stage compared with the third stage can be calculated, and the eigenvector of item B constructed according to the three increase rates is (-0.49, -0.45, 0); similarly, the feature vector of the constructed article C is (0.006, 0.17, 0.55). Then, the cosine similarity of the feature vector of the article a and the feature vector of the article B is calculated, the obtained similarity of the article a and the article B is 0.41, the cosine similarity of the feature vector of the article a and the feature vector of the article C is calculated, and the obtained similarity of the article a and the article C is 0.73. Then, the weight of item B, C among the 90 similar items can be found to be 0.025519, 0.014282 in an Inverse Distance Weighted (IDW) manner.

TABLE 2

Sales history phase	Sales of article A	Sales of article B	Sales of article C
				1	8485	13677	999
2	3958	6895	1005
				3	3606	3770	1182
4	4691	3768	1832

Step S306: and for a historical stage of the target item, carrying out weighted summation on the maximum sales volume occupation ratios of the similar items in the corresponding stage according to the weights of the similar items to obtain the maximum sales volume occupation ratio predicted value of the target item in the historical stage.

For example, for the 4 th sales history stage, i.e., stage 4, of item a shown in table 2, the maximum sales percentage of 90 similar items at stage 4 is weighted and summed according to the weights of 90 similar items, and the predicted value of the maximum sales percentage of item a at stage 4 is 38.53%.

Step S307: and determining the maximum sales volume predicted value of the target item according to the sales volume of the target item in the historical stage and the maximum sales volume ratio predicted value.

For example, in the 4 th sales history stage of article a shown in table 2, that is, the 4 th stage, the predicted maximum sales ratio is 38.53%, and the sales volume is 4691, so that the predicted maximum sales volume is 4691/38.53% — 12175.

Step S308: and determining the predicted value of the sales volume of the target article in the stage to be predicted according to the predicted value of the maximum sales volume of the target article and the predicted value of the maximum sales volume ratio of the target article in the stage to be predicted.

For example, in the 5 th sales stage of article a shown in table 2, i.e., the 5 th stage, the predicted maximum sales ratio was 36.47%, and since the predicted maximum sales ratio was 12175, the predicted sales ratio in the 5 th stage was 4440. For the 6 th sales stage of item a shown in table 2, i.e., 6 th stage, the predicted maximum sales percentage was 40.55%, and since the predicted maximum sales was 12175, the predicted sales at 6 th stage was 4937.

Optionally, after obtaining the sales prediction values of the target item at each prediction stage, the method according to the embodiment of the present invention further includes: the sales forecast values of the target articles in each forecast stage are stored so as to directly call the sales forecast data in various related business processes, and the sales forecast data do not need to be calculated in real time in various related business processes, so that the processing efficiency of the related business processes is improved. Taking the business process of determining the replenishment quantity as an example, after receiving a request of a user, the method can directly obtain a sales predicted value of the article and the stock of the article in a future period of time from the database, determine the replenishment quantity according to the stock of the article and the sales predicted value of the article in the future period of time, and send the replenishment suggestion carrying the replenishment quantity to a terminal corresponding to the user, so that the user can replenish the replenishment in time and accurately according to the replenishment suggestion.

In addition, after the sales volume predicted value of the target article in each prediction stage is obtained, the sales volume predicted value of the target article in each prediction stage can be displayed.

Fig. 4 is a schematic view of main blocks of an apparatus for determining an article sales amount according to a third embodiment of the present invention. The device provided by the embodiment of the invention is arranged in the electronic equipment. As shown in fig. 4, theapparatus 400 for determining the sales volume of an article according to an embodiment of the present invention includes: anacquisition module 401, aweight determination module 402, and asales determination module 403.

The obtainingmodule 401 is configured to obtain historical sales data of a plurality of similar items corresponding to the target item.

In an alternative embodiment, similar items corresponding to the target item are selected as items that belong to the same classification category as the target item and that are sold earlier in time than the target item. In this alternative embodiment, the obtainingmodule 401 may query the item information storage system according to the classification category information of the target item to query for items belonging to the same classification category as the target item, then screen out a plurality of items having a sale starting time earlier than that of the target item, and use the historical sale amount data of the screened items as the historical sale amount data of a plurality of similar items corresponding to the target item.

In another alternative embodiment, the feature vector of the target item and the feature vector of the item which has been listed for a long time may be constructed according to the item attribute data, and a similar item corresponding to the target item is screened from the items which have been listed for a long time by combining with a model such as KNN (K nearest neighbor), and then the obtainingmodule 401 obtains historical sales data of the similar item corresponding to the target item.

Aweight determining module 402, configured to calculate similarities between the target item and the multiple similar items according to historical sales data of the target item and the multiple similar items, and calculate weights of the multiple similar items according to the similarities between the target item and the multiple similar items.

In an alternative embodiment, theweight determination module 402 calculating the similarity between the target item and the plurality of similar items according to the historical sales data of the target item and the plurality of similar items comprises: theweight determination module 402 constructs a corresponding feature vector according to the sales volume of the target object in at least one historical stage; for each of the plurality of similar items, theweight determination module 402 constructs a corresponding feature vector according to the sales volume of the similar item in the at least one historical stage; and calculating the similarity of the target object and the similar object according to the feature vector of the target object and the feature vector of the similar object.

In another alternative embodiment, theweight determination module 402 calculating the similarity of the target item and the plurality of similar items according to the historical sales data of the target item and the plurality of similar items comprises: theweight determination module 402 calculates the growth rate of the target item in at least one historical stage according to the historical sales data of the target item; theweight determination module 402 constructs a corresponding feature vector according to the growth rate of the target object in at least one historical stage; for each of the plurality of similar items, theweight determination module 402 constructs a corresponding feature vector according to the growth rate of the similar item in the at least one historical stage; theweight determination module 402 calculates the similarity between the target item and the similar item according to the feature vector of the target item and the feature vector of the similar item.

For example, assuming that the target item has historical sales data of a total of 4 periods (or 4 historical stages) from the first period to the fourth period, the weight determination module 402 may calculate, according to the sales data of the 4 periods, a sales increase rate of the target item in the second period compared to the first period, a sales increase rate of the third period compared to the second period, and a sales increase rate of the fourth period compared to the third period, and then construct a feature vector corresponding to the target item according to the three growth rates, for example, construct a feature vector with a dimension of 3 with each growth rate as one dimension of the vector; assuming that similar articles share historical sales data of 16 periods in total from the first period to the 16 th period, the sales increase rate of the similar articles in the second period compared with the first period, the sales increase rate of the similar articles in the third period compared with the second period, and the sales increase rate of the similar articles in the fourth period compared with the third period can be calculated according to the sales data of the previous 4 periods of the similar articles, then a feature vector corresponding to the similar articles is constructed according to the three increase rates, then the distance between the feature vector of the target article and the feature vector of the similar articles is calculated, and the distance is used as the similarity between the target article and the similar articles.

Further, in the two alternative embodiments, the calculating, by theweight determining module 402, the similarity between the target item and the similar item according to the feature vector of the target item and the feature vector of the similar item includes: theweight determination module 402 calculates the distance between the feature vector of the target object and the feature vector of the similar object based on the formula such as the euclidean distance, the cosine similarity, or the hamming distance, and takes the distance as the similarity between the target object and the similar object.

For example, after obtaining the similarity between the target item and each similar item, theweight determining module 402 may calculate the weight of each similar item according to an Inverse Distance Weighting (IDW) method, which specifically includes: theweight determination module 402 calculates the sum of the similarities of the target item and the plurality of similar items; for each of the plurality of similar items, theweight determination module 402 takes the ratio of the sum of the similarity and the similarity of the target item and the similar item as the weight of the similar item. For example, if the target article includes only two similar articles, that is, article B and article C, and the similarity between article B and the target article is 0.73 and the similarity between article C and the target article is 0.41, the weight calculation formula for article B is 0.73/(0.73+0.41) and the weight calculation formula for article C is 0.41/(0.73+ 0.41).

Asales determining module 403, configured to determine a sales predicted value of the target item according to the weights of the similar items and historical sales data of the similar items.

In an alternative embodiment, the determiningmodule 403 determines the predicted sales value of the target item according to the weights of the similar items and the historical sales data of the similar items, including: for the stage to be predicted of the target item, the salesamount determining module 403 performs weighted summation on the historical sales amounts of the plurality of similar items at the corresponding stage according to the weights of the plurality of similar items to obtain the sales amount predicted value of the target item at the stage to be predicted. For example, assuming that the target items share the historical sales data of the first to fourth periods, when predicting the sales of the target items in the fifth period, the historical sales of the similar items in the 5 th period are weighted and summed according to the weights of the similar items corresponding to the target items, and the weighted and summed result is used as the sales predicted value of the target items in the 5 th period.

In another alternative embodiment, the determining module 403 determines the predicted sales value of the target item according to the weights of the similar items and the historical sales data of the similar items, including: for the historical stage of the target item, the sales determining module 403 performs weighted summation on the maximum sales proportions of the plurality of similar items in the corresponding stage according to the weights of the plurality of similar items to obtain a predicted value of the maximum sales proportions of the target item in the historical stage; the sales amount determining module 403 determines a predicted maximum sales amount value of the target item according to the sales amount of the target item in the historical stage and the predicted maximum sales amount ratio value; for the stage to be predicted of the target item, the sales determining module 403 performs weighted summation on the maximum sales proportions of the plurality of similar items in the corresponding stage according to the weights of the plurality of similar items to obtain a maximum sales proportion predicted value of the target item in the stage to be predicted; the sales determining module 403 determines the predicted value of the sales of the target item in the stage to be predicted according to the predicted value of the maximum sales of the target item and the predicted value of the maximum sales ratio of the target item in the stage to be predicted. The maximum sales volume ratio is the ratio of the sales volume of the article in the current stage to the maximum sales volume, and the maximum sales volume ratio reflects whether the product is in a high sales volume stage or a low sales volume stage within a certain period of time.

In the above optional embodiment, thesales determining module 403 obtains the predicted value of the maximum sales ratio of the target item by performing weighted summation on the maximum sales ratios of the similar items according to the weights of the similar items, and determines the predicted value of the sales of the target item in each prediction stage based on the change of the predicted value of the maximum sales ratio, so that the adverse effect on the sales prediction result due to the large difference in sales among the similar items can be eliminated, and the accuracy and reliability of the sales prediction result can be improved.

In the embodiment of the invention, the accurate and reliable prediction of the target article sales volume is realized through the device, and the device is particularly suitable for predicting the new article sales volume. Compared with the prior art, the device can improve the accuracy and reliability of the sales prediction result of the new product, solves the problems of low accuracy, poor reliability, even no prediction and the like of the sales prediction result caused by less historical sales information of the new product in the prior art, and further guides people to timely and accurately perform business operations such as replenishment and the like based on the sales prediction result.

Fig. 5 is a schematic view of main blocks of an apparatus for determining an article sales amount according to a fourth embodiment of the present invention. The device provided by the embodiment of the invention is arranged in the electronic equipment. As shown in fig. 5, theapparatus 500 for determining the sales volume of an article according to an embodiment of the present invention includes: anacquisition module 501, analignment module 502, aweight determination module 503, and asales determination module 504.

An obtainingmodule 501, configured to obtain historical sales data of a plurality of similar items corresponding to a target item.

With regard to how to determine a plurality of similar items of the target item in the embodiment of the present invention, reference may be made to the relevant contents of the embodiment shown in fig. 4.

Analignment module 502 for aligning historical sales data of the plurality of similar items according to historical stages of sales.

Theweight determination module 503 is configured to calculate similarity between the target item and the similar item according to historical sales data of the target item and the similar item; and calculating the weight of the similar articles according to the similarity.

Illustratively, theweight determination module 503 calculating the similarity of the target item and the similar items according to the historical sales data of the target item and the similar items includes: theweight determination module 503 calculates the growth rate of the target item in at least one historical stage according to the historical sales data of the target item; theweight determination module 503 constructs a corresponding feature vector according to the growth rate of the target object in at least one history stage; for each of the plurality of similar items, theweight determination module 503 constructs a corresponding feature vector according to the growth rate of the similar item in the at least one history stage; theweight determination module 503 calculates the similarity between the target item and the similar item according to the feature vector of the target item and the feature vector of the similar item. In specific implementation, theweight determining module 503 may calculate the distance between the feature vector of the target object and the feature vector of the similar object based on an expression such as euclidean distance, cosine similarity, or hamming distance, and use the distance as the similarity between the target object and the similar object.

Further, after obtaining the similarity between the target item and each similar item, theweight determining module 503 may calculate the weight of each similar item according to an Inverse Distance Weighting (IDW) method, specifically including: theweight determination module 503 calculates the sum of the similarity of the target item and the plurality of similar items; for each of the plurality of similar items, theweight determination module 503 takes the ratio of the sum of the similarity and the similarity of the target item and the similar item as the weight of the similar item.

Thesales determination module 504 is configured to determine a sales predicted value of the target item according to the weights of the similar items and historical sales data of the similar items, and specifically includes: for the historical stage of the target item, thesales determining module 504 performs weighted summation on the maximum sales proportions of the plurality of similar items in the corresponding stage according to the weights of the plurality of similar items to obtain a predicted value of the maximum sales proportions of the target item in the historical stage; the salesamount determining module 504 determines a maximum sales amount predicted value of the target item according to the sales amount of the target item in the historical stage and the maximum sales amount ratio predicted value; for the stage to be predicted of the target item, thesales determining module 504 performs weighted summation on the maximum sales proportions of the plurality of similar items in the corresponding stage according to the weights of the plurality of similar items to obtain a maximum sales proportion predicted value of the target item in the stage to be predicted; thesales determination module 504 determines the predicted sales value of the target item in the stage to be predicted according to the predicted maximum sales value of the target item and the predicted maximum sales ratio of the target item in the stage to be predicted. The maximum sales volume ratio is the ratio of the sales volume of the article in the current stage to the maximum sales volume, and the maximum sales volume ratio reflects whether the product is in a high sales volume stage or a low sales volume stage within a certain period of time.

In the embodiment of the present invention, thesales determining module 504 obtains the predicted value of the maximum sales ratio of the target item by performing weighted summation on the maximum sales ratios of the similar items according to the weights of the similar items, and determines the predicted value of the sales of the target item in each prediction stage based on the change of the predicted value of the maximum sales ratio, so as to eliminate the adverse effect on the sales prediction result due to the large difference in sales among the similar items, and improve the accuracy and reliability of the sales prediction result.

In the embodiment of the invention, the accuracy and the reliability of the new product sales prediction result can be further improved through the device, the problems of low accuracy, poor reliability, even no prediction and the like of the sales prediction result caused by less historical sales information of the new product in the prior art are solved, and then people are guided to timely and accurately carry out business operations such as replenishment and the like based on the sales prediction result.

Referring now to FIG. 6, shown is a block diagram of acomputer system 600 suitable for use with the electronic device implementing an embodiment of the present invention. The computer system illustrated in FIG. 6 is only one example and should not impose any limitations on the scope of use or functionality of embodiments of the invention.

As shown in fig. 6, thecomputer system 600 includes a Central Processing Unit (CPU)601 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)602 or a program loaded from astorage section 608 into a Random Access Memory (RAM) 603. In theRAM 603, various programs and data necessary for the operation of thesystem 600 are also stored. TheCPU 601,ROM 602, andRAM 603 are connected to each other via abus 604. An input/output (I/O)interface 605 is also connected tobus 604.

The following components are connected to the I/O interface 605: aninput portion 606 including a keyboard, a mouse, and the like; anoutput portion 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; astorage section 608 including a hard disk and the like; and acommunication section 609 including a network interface card such as a LAN card, a modem, or the like. Thecommunication section 609 performs communication processing via a network such as the internet. Thedriver 610 is also connected to the I/O interface 605 as needed. Aremovable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on thedrive 610 as necessary, so that a computer program read out therefrom is mounted in thestorage section 608 as necessary.

In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through thecommunication section 609, and/or installed from theremovable medium 611. The computer program performs the above-described functions defined in the system of the present invention when executed by the Central Processing Unit (CPU) 601.

It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes an acquisition module, a weight determination module, and a sales determination module. The names of the modules do not limit the module itself in some cases, for example, the acquiring module may also be described as a module for acquiring historical sales data of a plurality of similar items corresponding to the target item.

As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to perform the following: acquiring historical sales data of a plurality of similar articles corresponding to the target article; wherein the plurality of similar items are sold earlier in time than the target item; calculating the similarity of the target item and the similar items according to the historical sales data of the target item and the similar items; calculating the weights of the similar items according to the similarity of the target item and the similar items; and determining a sales predicted value of the target item according to the weights of the similar items and historical sales data of the similar items.

According to the technical scheme of the embodiment of the invention, the accuracy and the reliability of the new product sales prediction result can be improved, the problems of low accuracy, poor reliability, even no prediction and the like of the sales prediction result caused by less historical sales information of the new product in the prior art are solved, and then people are guided to timely and accurately carry out business operations such as replenishment and the like based on the sales prediction result.

The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.