CN111144985A

CN111144985A - Unit transfer value adjusting method, unit transfer value adjusting device, computer equipment and storage medium

Info

Publication number: CN111144985A
Application number: CN201911350785.4A
Authority: CN
Inventors: 阎相达; 黄平; 李治
Original assignee: Beijing Missfresh Ecommerce Co Ltd
Current assignee: Beijing Missfresh Ecommerce Co Ltd
Priority date: 2019-12-24
Filing date: 2019-12-24
Publication date: 2020-05-12

Abstract

The embodiment of the application discloses a unit transfer value adjusting method, a unit transfer value adjusting device, computer equipment and a storage medium, and belongs to the technical field of computers. The method comprises the following steps: the method comprises the steps of obtaining information sets corresponding to a plurality of candidate article identifications, wherein the information sets at least comprise area identifications of geographic areas to which corresponding articles belong, the area identifications in the information sets are the same, obtaining a model based on abnormal probability of area identification matching, respectively determining the probability of the candidate article identifications, selecting at least one article identification from the candidate article identifications according to the probability of the candidate article identifications, adjusting a unit transfer value corresponding to the at least one article identification, and avoiding manually determining the unit transfer value of the articles. And the articles in the abnormal state can be adjusted in time, so that the loss caused by the articles in the abnormal state is avoided.

Description

Unit transfer value adjusting method, unit transfer value adjusting device, computer equipment and storage medium

Technical Field

The embodiment of the application relates to the technical field of computers, in particular to a unit transfer value adjusting method, a unit transfer value adjusting device, computer equipment and a storage medium.

Background

With the development of computer technology and the wide application of electronic commerce, the online transfer or offline transfer of articles is more and more, and great convenience is brought to the life of people. Generally, when an item is transferred, an item provider sets unit price of the item, a demander determines the number of items to be received, so that a total price can be determined according to the unit price of the item and the number of the items, money corresponding to the total price is transferred to the provider, and the provider transfers the corresponding number of items to the demander.

The item provider, when determining the unit price of an item, is generally determined based on information such as the number of transfers before the item, the remaining number, and the number of transfers of other items. This method requires manual determination of the unit price of the article, which is easily affected by subjective factors, resulting in inaccurate unit price of the article.

Disclosure of Invention

The embodiment of the application provides a unit transfer value adjusting method, a unit transfer value adjusting device, computer equipment and a storage medium, and the unit transfer value adjusting method and the computer equipment can effectively improve the unit price accuracy of an article. The technical scheme is as follows:

in one aspect, a unit transfer value adjustment method is provided, and the method includes:

acquiring information sets corresponding to a plurality of candidate item identifications, wherein the information sets at least comprise area identifications of geographic areas to which corresponding items belong, and the area identifications in the information sets are the same;

processing the acquired information set based on the abnormal probability acquisition model matched with the region identifier, and respectively determining the probability of the multiple candidate article identifiers, wherein the probability is used for representing the probability that the corresponding article is in an abnormal state under the condition of transferring out;

and selecting at least one article identifier from the plurality of candidate article identifiers according to the probability of the plurality of candidate article identifiers, and adjusting a unit transfer numerical value corresponding to the at least one article identifier.

In one possible implementation manner, the obtaining information sets corresponding to a plurality of candidate item identifiers includes:

acquiring an information set corresponding to a plurality of article identifications, wherein the information set at least comprises area identifications, current feature values and historical feature values of geographic areas to which the corresponding articles belong, the area identifications in the information sets are the same, and the feature values of the articles comprise at least one of unit transfer values, transfer-out quantities or residual quantities of the articles;

for each article identification, counting historical characteristic values corresponding to the article identification to obtain an abnormal value range of the historical characteristic values; and if the current characteristic numerical value belongs to the abnormal numerical value range, determining the article identifier as an alternative article identifier.

In another possible implementation manner, the geographic area corresponding to the area identifier includes a plurality of warehouses, and the warehouse identifiers of different warehouses are different; the acquiring of the information sets corresponding to the multiple candidate item identifiers includes:

for any optional article identifier, acquiring a plurality of initial information sets of the optional article identifier, and fusing the initial information sets to serve as information sets corresponding to the optional article identifier;

the initial information set at least includes a warehouse identifier of a warehouse to which the corresponding article belongs and an area identifier of a geographic area to which the article belongs, and the area identifiers included in the plurality of information sets are the same and different.

In another possible implementation manner, the adjusting the unit transfer value corresponding to the at least one item identifier includes:

and processing the information set corresponding to the at least one article identifier based on a unit transfer value determination model, and determining the unit transfer value of the at least one article identifier after adjustment.

In another possible implementation, the method includes:

acquiring a sample information set corresponding to a sample article identifier, wherein the sample information set comprises a unit transfer value of the sample article identifier in a first history period, and unit transfer values and transfer-out quantities in other history periods before the first history period;

and training the unit transfer numerical value determination model according to the sample information set.

for each article identifier in the at least one article identifier, acquiring a corresponding relation between a history unit transfer numerical value and a history transfer-out quantity included in history data of the article identifier;

inquiring the corresponding relation according to the target transfer-out quantity of the article identifier, and determining a historical unit transfer numerical value corresponding to the historical transfer-out quantity matched with the target transfer-out quantity;

and taking the historical unit transfer value as the unit transfer value after the article identification is adjusted.

In another possible implementation, the set of information further includes a remaining number of items;

before querying the corresponding relationship according to the target transfer-out quantity of the item identifier and determining a historical unit transfer value corresponding to the historical transfer-out quantity matched with the target transfer-out quantity, the method further includes:

and determining the difference between the residual quantity of the articles identified by the articles and the target residual quantity at the end moment of the current period as the target roll-out quantity.

In another possible implementation, the at least one item identifier includes a plurality of item identifiers; after the adjusting the unit transfer value corresponding to the at least one article identifier, the method further includes:

dividing the plurality of item identifications into a plurality of item identification combinations so that each item identification combination comprises a first preset number of item identifications;

counting the total transfer value corresponding to each article identification combination according to the unit transfer value and the target transfer-out quantity of each article identification after adjustment;

and taking the article identifier in the article identifier combination with the maximum total transfer value as a target article identifier, and issuing the unit transfer value adjusted by the target article identifier.

dividing the plurality of item identifications into a plurality of item identification combinations so that each item identification combination comprises a second preset number of item identifications;

according to the unit transfer value, the unit cost value and the target transfer-out quantity which are adjusted by each article identifier, counting a total income value corresponding to each article identifier combination;

and taking the article identifier in the article identifier combination with the maximum total profit value as a target article identifier, and issuing the unit transfer value adjusted by the target article identifier.

In another possible implementation manner, the selecting, according to the probabilities of the multiple candidate item identifiers, a target item identifier from the multiple candidate item identifiers includes:

and selecting at least one item identifier with the probability belonging to a preset probability range from the plurality of candidate item identifiers.

In another possible implementation manner, the selecting, from the multiple candidate item identifiers, at least one item identifier with a probability that belongs to a preset probability range includes:

and selecting at least one item identifier with the probability greater than a preset threshold value from the plurality of candidate item identifiers.

In another possible implementation manner, the selecting at least one item identifier from the multiple candidate item identifiers according to the probabilities of the multiple candidate item identifiers includes:

and selecting a third preset number of article identifications from the plurality of candidate article identifications according to the probability of the plurality of candidate article identifications, wherein the probability of the preset number of article identifications is greater than the probability of other candidate article identifications in the plurality of candidate article identifications.

In another possible implementation manner, the abnormal probability obtaining model includes at least one of a conversion probability obtaining model, an order rejection probability obtaining model, a sold-out probability obtaining model, or a loss reporting probability obtaining model.

In another aspect, there is provided a unit shift value adjusting apparatus, the apparatus including:

an information set acquisition module, configured to acquire an information set corresponding to multiple candidate item identifiers, where the information set at least includes area identifiers of geographic areas to which corresponding items belong, and the area identifiers included in the multiple information sets are the same;

a probability determination module, configured to process the acquired information set based on the abnormal probability acquisition model matched with the region identifier, and respectively determine probabilities of the multiple candidate item identifiers, where the probabilities are used to indicate probabilities that corresponding item roll-out situations are in abnormal states;

and the transfer value adjusting module is used for selecting at least one article identifier from the multiple candidate article identifiers according to the probability of the multiple candidate article identifiers and adjusting the unit transfer value corresponding to the at least one article identifier.

In one possible implementation manner, the information set obtaining module includes:

an information set obtaining unit, configured to obtain an information set corresponding to a plurality of article identifiers, where the information set includes at least an area identifier of a geographic area to which a corresponding article belongs, a current feature value, and a historical feature value, and the area identifiers included in the plurality of information sets are the same, and the feature value of the article includes at least one of a unit transfer value, a transfer-out quantity, or a remaining quantity of the article;

the first article identification determining unit is used for counting the historical characteristic numerical values corresponding to the article identifications for each article identification to obtain the abnormal numerical value range of the historical characteristic numerical values; and if the current characteristic numerical value belongs to the abnormal numerical value range, determining the article identifier as an alternative article identifier.

In another possible implementation manner, the geographic area corresponding to the area identifier includes a plurality of warehouses, and the warehouse identifiers of different warehouses are different; the information set acquisition module comprises:

a second article identifier determining unit, configured to acquire, for any one of the candidate article identifiers, multiple initial information sets of the candidate article identifier, and fuse the multiple initial information sets to serve as an information set corresponding to the candidate article identifier;

In another possible implementation manner, the transfer value adjusting module includes:

and the first transfer value adjusting unit is used for processing the information set corresponding to the at least one article identifier based on the unit transfer value determining model and determining the unit transfer value after the at least one article identifier is adjusted.

In another possible implementation, the apparatus includes:

the system comprises a sample information set acquisition module, a sample information set processing module and a sample information processing module, wherein the sample information set acquisition module is used for acquiring a sample information set corresponding to a sample article identifier, and the sample information set comprises a unit transfer value of the sample article identifier in a first history period, and unit transfer values and transfer-out quantities in other history periods before the first history period;

and the model training module is used for training the unit transfer numerical value determination model according to the sample information set.

a corresponding relation obtaining unit, configured to obtain, for each of the at least one item identifier, a corresponding relation between a history unit transfer value and a history transfer-out quantity included in history data of the item identifier;

a numerical value determining unit, configured to query the correspondence according to the target roll-out quantity of the item identifier, and determine a historical unit transfer numerical value corresponding to a historical roll-out quantity that matches the target roll-out quantity;

and the second transfer value adjusting unit is used for taking the historical unit transfer value as the unit transfer value after the article identification is adjusted.

the transfer value adjusting module comprises:

and the target roll-out quantity determining unit is used for determining the difference value between the residual quantity of the articles identified by the articles and the target residual quantity at the end moment of the current period as the target roll-out quantity.

In another possible implementation, the at least one item identifier includes a plurality of item identifiers; the device further comprises:

the first combination dividing module is used for dividing the plurality of item identifications into a plurality of item identification combinations so that each item identification combination comprises a first preset number of item identifications;

the transfer value counting module is used for counting the total transfer value corresponding to each article identification combination according to the unit transfer value and the target transfer-out quantity adjusted by each article identification;

and the first numerical value issuing module is used for issuing the unit transfer numerical value adjusted by the target article identifier by taking the article identifier in the article identifier combination with the maximum total transfer numerical value as the target article identifier.

the second combination dividing module is used for dividing the plurality of item identifications into a plurality of item identification combinations so that each item identification combination comprises a second preset number of item identifications;

the income value counting module is used for counting the total income value corresponding to each article identification combination according to the unit transfer value, the unit cost value and the target transfer-out quantity which are adjusted by each article identification;

and the second numerical value issuing module is used for issuing the unit transfer numerical value adjusted by the target article identifier by taking the article identifier in the article identifier combination with the maximum total profit numerical value as the target article identifier.

and the first article identification selecting unit is used for selecting at least one article identification with the probability belonging to a preset probability range from the plurality of candidate article identifications.

In another possible implementation manner, the first item identifier selecting unit is further configured to select at least one item identifier, of which the probability is greater than a preset threshold, from the multiple candidate item identifiers.

and the second article identifier selecting unit is used for selecting a third preset number of article identifiers from the plurality of candidate article identifiers according to the probabilities of the plurality of candidate article identifiers, wherein the probability of the preset number of article identifiers is greater than the probabilities of other candidate article identifiers in the plurality of candidate article identifiers.

In another aspect, a computer device is provided, which includes a processor and a memory, wherein at least one program code is stored in the memory, and the at least one program code is loaded and executed by the processor to implement the unit transfer value adjustment method according to the above aspect.

In another aspect, a computer-readable storage medium is provided, in which at least one program code is stored, the at least one program code being loaded and executed by a processor to implement the unit transfer value adjustment method according to the above aspect.

The beneficial effects brought by the technical scheme provided by the embodiment of the application at least comprise:

the method, the device, the computer equipment and the storage medium provided by the embodiment of the application obtain information sets corresponding to a plurality of candidate article identifications, the information sets at least comprise area identifications of geographic areas to which corresponding articles belong, the area identifications in the information sets are the same, based on an abnormal probability obtaining model matched with the area identifications, the probabilities of the candidate article identifications are respectively determined, the probabilities are used for representing the probability that the corresponding articles are in abnormal states in a transferring-out condition, at least one article identification is selected from the candidate article identifications according to the probabilities of the candidate article identifications, unit transfer values corresponding to the at least one article identification are adjusted, the unit transfer values of the articles do not need to be manually determined, the unit transfer values after article adjustment are prevented from being influenced by subjective factors, and the accuracy is improved. And the article needing to adjust the unit transfer value is determined according to the probability that the article is in the abnormal state under the condition of transferring out, so that the article in the abnormal state can be adjusted in time, and the loss caused by the article in the abnormal state is avoided.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic illustration of an implementation environment provided by an embodiment of the present application;

FIG. 2 is a schematic illustration of an implementation environment provided by an embodiment of the present application;

FIG. 3 is a flow chart of a method for adjusting a unit transition value according to an embodiment of the present disclosure;

FIG. 4 is a flow chart of a method for adjusting a unit transition value according to an embodiment of the present disclosure;

FIG. 5 is a flow chart of a method for adjusting a unit transition value according to an embodiment of the present disclosure;

fig. 6 is a schematic structural diagram of a unit transfer value adjusting apparatus according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of a unit transfer value adjusting apparatus according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of a terminal according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of a server according to an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the embodiments of the present application more clear, the embodiments of the present application will be further described in detail with reference to the accompanying drawings.

Before explaining embodiments of the present application in detail, some terms that may be referred to in the embodiments of the present application will be introduced.

Front-end bin mode: the front-mounted bins are also called micro-bins, the front-mounted bin mode is a bin configuration mode, each front-mounted bin can be a small and medium-sized storage distribution center, so that a central main bin (also called a main bin) only needs to supply materials to each front-mounted bin, and different front-mounted bins do not need to be mutually allocated. After a customer (also referred to as a consumer) places an order, the items purchased by the customer may be shipped from a nearby front warehouse, rather than from a general warehouse such as one located in a rural area.

In short, the front warehouse is arranged at a place close to the user, for example, a city main warehouse can be established in a city, then a plurality of front warehouses are established in a plurality of communities or business circles of the city according to the order density and other factors, and the articles purchased by the user can come from one front warehouse arranged in the nearby communities or business circles, so that the user can be guaranteed to be delivered to the home in a short time after placing an order, and the quality and the delivery speed of the articles are guaranteed.

The number of roll-outs: in the embodiment of the present application, the roll-out number refers to the sales volume of the items sold in the front bin. The roll-out number may be divided into a daily roll-out number, a monthly roll-out number, an annual roll-out number, and the like, which is not specifically limited in this embodiment of the present application.

The remaining amount: in the embodiment of the present application, the remaining amount refers to the stock amount of the articles in the front warehouse, and the stock amount refers to the goods actually stored in the warehouse. In a broad sense, inventory generally refers to the amount of cargo actually stored in a warehouse.

Unit transfer value: in the embodiment of the present application, the unit transfer value refers to the unit price of the item, that is, the value that the user should transfer to the item provider when transferring out an item.

Fig. 1 is a schematic structural diagram of an implementation environment provided in an embodiment of the present application, and as shown in fig. 1, the implementation environment includes a terminal 101 and a server 102. The terminal 101 establishes a communication connection with the server 102, and performs interaction through the established communication connection.

The terminal 101 may be various types of terminals 101 such as a mobile phone, a computer, a tablet computer, and the like. The server 102 may be a server, a server cluster composed of several servers, or a cloud computing server center. The server 102 may be a server of a digital e-commerce platform or a server of another e-commerce platform.

The server 102 selects at least one article identifier from the plurality of candidate article identifiers by obtaining an information set corresponding to the plurality of candidate article identifiers, adjusts a unit transfer value corresponding to the at least one article identifier, sends the unit transfer value adjusted by the at least one article identifier to the terminal 101, and after receiving the unit transfer value adjusted by the at least one article identifier, the terminal 101 displays the unit transfer value corresponding to the at least one article identifier so that a user can view the unit transfer value.

After checking the unit transfer value corresponding to each article identifier, the user can select any article identifier to perform order placing operation, the terminal 101 sends an order carrying the article identifier to the server 102, and the server 102 receives the order sent by the terminal 101 and informs an operator to deliver the article corresponding to the article identifier to the user.

Alternatively, referring to fig. 2, the fresh electric commercial platform adopts a two-stage distributed warehousing system of a total bin + a front bin. A total bin is established within a region and leading bins are established at a plurality of locations within the region. For example, a city may be built with a total city warehouse, and then multiple front warehouses may be built in multiple communities or business circles of the city according to the order density and other factors.

The implementation environment may also include a total bin 201 and a plurality of leading bins 202. When the server 102 receives the order sent by the terminal 101, the order is distributed to a front bin 202 near the user, the article corresponding to the article identifier is selected from the front bin 201 and distributed to the user, and therefore the user can be guaranteed that the article can be distributed to the home in a short time after placing the order.

As shown in fig. 3, the server may include a data service system 301, an intelligent pricing control system 302, an intelligent pricing algorithm system 303, and a marketing management system 304.

The data service system 301 may obtain a plurality of item data, such as inventory, sales, prices, promotional indicia, and the like, for a plurality of geographic regions. The data service system 301 further has an anomaly detection function and a real-time visualization function, and for the anomaly detection function, according to historical data distribution, determines whether current item data is in an anomaly range, and sends the item data in the anomaly range to the intelligent pricing control system 302, or marks the item data in the anomaly range and sends a plurality of item data to the intelligent pricing control system 302. The platform for providing data observation for the real-time visualization function has the characteristics of intuition, quickness and centralized information quantity, and is favorable for promoting business understanding, strategy verification and algorithm optimization.

The intelligent pricing control system 302 receives the item data sent by the data service system 301, and may also obtain item configuration information input by a manager, integrate the item configuration information and the item data, determine the state of an item corresponding to the item data through the risk control system, and send the item data in an abnormal state to the intelligent pricing algorithm system 303 to adjust the price.

The intelligent pricing algorithm system 303 receives the data of the items to be priced, which is transmitted from the intelligent pricing control system 302, calculates the current optimal price based on the inventory, historical characteristics, volume-price relationship and the like of the front bin according to different strategy objectives, and returns the price result data to the intelligent pricing control system 302.

The intelligent pricing control system 302 may further receive price result data returned by the intelligent pricing algorithm system 303, calculate the pricing space range and the upgradability of the item in the abnormal state through a meta & reinformance learning model by the profit control system, maximize the overall profit margin or actual income, determine an item pricing combination, and send the determined item pricing combination to the marketing management system 304.

The marketing management system 304 receives the item pricing combination transmitted by the intelligent pricing control system 302, and can directly issue the pricing combination, or send a price adjusting application to the management terminal and issue the pricing combination after approval.

Fig. 4 is a flowchart of a unit transfer value adjustment method provided in an embodiment of the present application, and as shown in fig. 4, the method includes:

401. acquiring information sets corresponding to a plurality of candidate article identifications, wherein the information sets at least comprise area identifications of geographic areas to which the corresponding articles belong, and the area identifications in the information sets are the same;

402. respectively determining the probability of a plurality of candidate article identifications based on the abnormal probability acquisition model matched with the region identifications, wherein the probability is used for expressing the probability that the corresponding article transfer-out condition is in an abnormal state;

403. and selecting at least one article identifier from the multiple candidate article identifiers according to the probability of the multiple candidate article identifiers, and adjusting the unit transfer value corresponding to the at least one article identifier.

The method provided by the embodiment of the application comprises the steps of obtaining an information set corresponding to a plurality of candidate article identifications, wherein the information set at least comprises area identifications of geographic areas to which the corresponding articles belong, the area identifications in the information sets are the same, respectively determining the probabilities of the candidate article identifications based on an abnormal probability obtaining model matched with the area identifications, the probabilities are used for representing the probability that the corresponding article roll-out conditions are in abnormal states, selecting at least one article identification from the candidate article identifications according to the probabilities of the candidate article identifications, adjusting a unit transfer value corresponding to the at least one article identification, and avoiding manually determining the unit transfer value of the articles, so that the unit transfer value after article adjustment is prevented from being influenced by subjective factors, and the accuracy is improved. And the article needing to adjust the unit transfer value is determined according to the probability that the article is in the abnormal state under the condition of transferring out, so that the article in the abnormal state can be adjusted in time, and the loss caused by the article in the abnormal state is avoided.

In one possible implementation manner, obtaining an information set corresponding to a plurality of candidate item identifications includes:

acquiring an information set corresponding to a plurality of article identifications, wherein the information set at least comprises area identifications, current characteristic numerical values and historical characteristic numerical values of geographical areas to which the corresponding articles belong, the area identifications in the information sets are the same, and the characteristic numerical values of the articles comprise at least one of unit transfer numerical values, transfer-out quantities or residual quantities of the articles;

for each article identification, counting historical characteristic values corresponding to the article identification to obtain an abnormal value range of the historical characteristic values; and if the current characteristic numerical value belongs to the abnormal numerical value range, determining the article identification as the alternative article identification.

In another possible implementation manner, the geographic area corresponding to the area identifier includes a plurality of warehouses, and the warehouse identifiers of different warehouses are different; acquiring an information set corresponding to a plurality of candidate item identifications, wherein the information set comprises:

the initial information set at least comprises a warehouse identifier of a warehouse to which the corresponding article belongs and an area identifier of a geographic area to which the article belongs, and the area identifiers in the information sets are the same and different.

In another possible implementation manner, adjusting the unit transfer value corresponding to at least one article identifier includes:

and determining the unit transfer value adjusted by at least one article identifier based on the unit transfer value determination model.

In another possible implementation, a method includes:

acquiring a sample information set corresponding to the sample article identification, wherein the sample information set comprises unit transfer values of the sample article identification in a first history period, unit transfer values and transfer-out quantities in other history periods before the first history period;

and training the unit transfer numerical determination model according to the sample information set.

for each article identifier in at least one article identifier, acquiring the corresponding relation between a history unit transfer numerical value and a history transfer-out quantity in the history data of the article identifier;

inquiring the corresponding relation according to the target transfer-out quantity of the article identification, and determining a historical unit transfer numerical value corresponding to the historical transfer-out quantity matched with the target transfer-out quantity;

In another possible implementation, the information set further includes a remaining number of items;

according to the target transfer-out quantity of the article identification, the corresponding relation is inquired, and before the historical unit transfer value corresponding to the historical transfer-out quantity matched with the target transfer-out quantity is determined, the method comprises the following steps:

and determining the difference value between the article residual quantity of the article identification and the target residual quantity at the end moment of the current period as the target roll-out quantity.

In another possible implementation, the at least one item identifier includes a plurality of item identifiers; after the unit transfer value corresponding to at least one article identifier is adjusted, the method further comprises the following steps:

counting a total transfer value corresponding to each article identification combination according to the unit transfer value and the target transfer-out quantity of each article identification after adjustment;

dividing the plurality of article identifications into a plurality of article identification combinations so that each article identification combination comprises a second preset number of article identifications;

according to the adjusted unit transfer value, unit cost value and target transfer-out quantity of each article identifier, counting a total income value corresponding to each article identifier combination;

In another possible implementation manner, selecting a target item identifier from the multiple candidate item identifiers according to the probabilities of the multiple candidate item identifiers includes:

and selecting at least one article identifier with the probability belonging to a preset probability range from the plurality of candidate article identifiers.

In another possible implementation manner, selecting at least one item identifier with a probability that belongs to a preset probability range from a plurality of candidate item identifiers includes:

In another possible implementation manner, selecting at least one item identifier from the multiple candidate item identifiers according to the probabilities of the multiple candidate item identifiers includes:

and selecting a third preset number of article identifications from the multiple candidate article identifications according to the probability of the multiple candidate article identifications, wherein the probability of the preset number of article identifications is greater than the probability of other candidate article identifications in the multiple candidate article identifications.

In another possible implementation manner, the abnormal probability obtaining model includes at least one of a transition probability obtaining model, a return probability obtaining model, a sold-out probability obtaining model or a loss reporting probability obtaining model.

Fig. 5 is a flowchart of a unit transfer value adjustment method provided in an embodiment of the present application, and as shown in fig. 5, the method is applied to a server, and includes:

501. the server acquires information sets corresponding to the multiple candidate item identifications.

The article identifier may be an identifier that can identify a unique article type, and may be a type number, a type name, or the like of the article. The embodiment of the present application does not limit the specific form of the article identifier. The information set corresponding to the item identifier is a set of related information of the item corresponding to the item identifier. The information set at least comprises area identifications of the geographic areas to which the corresponding items belong.

The area identifier is an identifier capable of determining a unique geographic area, the geographic area may be a city, a city area, or the like, and the area identifier may be a city name, an area number, or the like. Because the article transfer conditions in different areas may be different, the area identifiers included in the acquired information sets are the same, which means that the information sets of multiple articles belonging to the same geographic area are acquired, the acquired information sets can reflect the article transfer conditions in the same geographic area, and subsequently, the unit transfer value of the article can be accurately adjusted according to the acquired information sets.

In addition, the information set may further include item information such as item identification, unit transfer value, current remaining quantity, and historical data of the item, and the historical data may include historical unit transfer value, and one or more roll-out quantities for historical time period.

For the obtaining manner of the information sets corresponding to the multiple candidate item identifiers, in a possible implementation manner, the server obtains multiple information sets from the database, and according to the item identifier included in each information set and the area identifier of the geographic area to which the corresponding item belongs, the multiple information sets are screened to determine the information sets corresponding to the multiple candidate item identifiers, so that the area identifiers included in the information sets corresponding to different candidate item identifiers are the same.

In the storage method of the database, article information of a plurality of articles, article identifiers corresponding to each article information, and area identifiers corresponding to each article information may be stored in association with each other to form an information set, as shown in table 1.

TABLE 1

Region identification	Article identification	Article information
			Region identifier 1	Article identification A	Unit price, sales volume, stock quantity
Region identifier 1	Article identification B	Unit price, sales volume, stock quantity
			Region identification 2	Article identification 3	Unit price, sales volume, stock quantity

In another possible implementation, the step 501 includes: the server acquires information sets corresponding to the article identifications; for each article identification, counting historical characteristic values corresponding to the article identification to obtain an abnormal value range of the historical characteristic values; and if the current characteristic numerical value belongs to the abnormal numerical value range, determining the article identification as the alternative article identification.

The information sets at least comprise area identifications, current characteristic numerical values and historical characteristic numerical values of geographic areas to which corresponding articles belong, and the area identifications included in the information sets are the same, so that the acquired article identifications belong to the same geographic area.

The characteristic value of the article includes at least one of a unit transfer value, a roll-out quantity, or a remaining quantity of the article. The abnormal numerical range may be an abnormal numerical range preset by a manager, and if the characteristic numerical value belongs to the abnormal numerical range, the characteristic numerical value is abnormal, and if the characteristic numerical value does not belong to the abnormal numerical range, the characteristic numerical value is normal.

The abnormal value range can be determined according to the occurrence frequency corresponding to various historical characteristic values or the occurrence probability corresponding to various historical characteristic values. For example, the number of occurrences of the past history feature value is comprehensively counted, and the history feature value with the too small number of occurrences is used as the abnormal feature value, thereby determining the abnormal value range.

The server counts the historical characteristic values corresponding to the article identification, and can respectively count the occurrence times of different historical characteristic values and the total occurrence times of a plurality of historical characteristic values in the historical characteristic values in a normal distribution mode to determine a normal distribution diagram of the historical characteristic values, wherein the confidence interval is used as a normal range, and the value intervals except the confidence interval are used as an abnormal value range. For example, if the value range of the historical feature value is (0, 100), and the confidence interval with the confidence level of 95% is (30, 70), the value ranges (0, 30) and (70, 100) are taken as the abnormal value range of the feature value.

And screening the plurality of article identifications according to the abnormal value range of the historical characteristic value of each article identification, and screening the article identifications of which the current characteristic values belong to the abnormal value range for subsequent processing, so that the screened alternative article identifications can be subsequently processed, the number of the subsequent article identifications needing to be processed is reduced, and the efficiency of subsequently adjusting the unit transfer value of the article identification is improved.

In one possible implementation, this step 501 includes: for any optional article identifier, acquiring a plurality of initial information sets of the optional article identifier, fusing the initial information sets to serve as information sets corresponding to the optional article identifier, wherein the initial information sets at least comprise warehouse identifiers of warehouses to which the corresponding articles belong and area identifiers of geographical areas to which the corresponding articles belong, and the area identifiers in the information sets are the same and the warehouse identifiers are different.

The embodiment of the application processes a plurality of candidate item identifiers in a geographic area, and because each geographic area may include a plurality of warehouses, different warehouses may include the same item, and information sets corresponding to different warehouses are different for the same item, when an information set corresponding to any item identifier in any geographic area is obtained, an information set corresponding to the item identifier in the plurality of warehouses in the geographic area needs to be obtained first. The geographic area corresponding to the area identification comprises a plurality of warehouses, and the warehouse identifications of different warehouses are different.

For example, the geographical area includes 3 warehouses, the inventory level of the article a in the warehouse 1 is 20, the inventory level of the article a in the warehouse 2 is 30, and the inventory level of the article a in the warehouse 3 is 25, then the inventory level of the alternative article identifier a is 75.

502. And the server processes the acquired information set based on the abnormal probability acquisition model matched with the area identifier, and respectively determines the probability of the multiple candidate article identifiers.

And respectively inputting the information set corresponding to each candidate article identifier into an abnormal probability acquisition model, and outputting the probability of the corresponding candidate article identifier by the abnormal probability acquisition model. The probability is used for representing the probability that the corresponding article is in an abnormal state under the condition of transferring out, and whether the corresponding article is in the abnormal state or not can be determined according to the probability of each candidate article identifier, so that the article in the abnormal state can be processed in the following process. The abnormal probability obtaining model can be a random forest model, a linear programming model, a deep learning model, a small sample learning model and the like.

Because the information sets corresponding to the multiple candidate item identifiers included in different geographic areas are different, different abnormal probability acquisition models can be set for different geographic areas, and therefore, the server needs to determine the abnormal probability acquisition model matched with the current area identifier first, and then process the acquired information sets based on the abnormal probability acquisition model. The abnormal probability obtaining models corresponding to different geographic areas can be obtained through training of historical data of the corresponding geographic areas.

For the process of training the abnormal probability obtaining model, in a possible implementation manner, a sample information set corresponding to a plurality of sample article identifiers and an indication identifier corresponding to each sample article identifier are obtained, and the indication identifier is used for indicating whether the corresponding article is in an abnormal state or not; performing iterative training on the initial abnormal probability acquisition model through a sample information set corresponding to a plurality of sample article identifications and an indication identification corresponding to each sample article identification, and stopping the iterative training when the iteration times reach preset times; or stopping the iterative training when the accuracy of the probability output by the abnormal probability acquisition model reaches a preset threshold. The plurality of sample article identifications can comprise positive sample article identifications and negative sample article identifications, and the indication identifications corresponding to the positive sample article identifications indicate that the roll-out conditions of the corresponding articles are in an abnormal state; the indication mark corresponding to the negative sample article mark indicates that the roll-out condition of the corresponding article is in a non-abnormal state.

In addition, since the information sets of the same article may be different at different times, each of the acquired plurality of sample information sets includes a corresponding time, and the probability that the roll-out condition of the article is in an abnormal state at different times can be determined through the abnormal probability acquisition model trained by the plurality of sample information sets.

As for the type of the abnormal probability obtaining model, in a possible implementation manner, the abnormal probability obtaining model includes at least one of a transfer-out probability obtaining model, a drop-out probability obtaining model, a sold-out probability obtaining model or a loss reporting probability obtaining model.

The roll-out probability obtaining model is used for obtaining the ratio of the user attention frequency of any article to the roll-out frequency, namely the roll-out probability; the order refund probability obtaining model is used for obtaining the ratio of the number of times of transferring out any article to the number of times of refunding the order, namely the order refund probability; a sold-out probability acquisition model is used for acquiring the probability that any article is in a preset hot sale state, namely the probability that the residual quantity of any article is 0 before the current period is finished; the damage reporting probability obtaining model is used for obtaining the probability that any article is in a preset hysteresis pin state, namely the ratio of the damage number of any article to the total number of the article before the current period is ended, namely the damage probability.

During the process of transferring the articles, each article can have various types of abnormal states, for example, when the user of the article is high in the flow rate of interest, but the transferring quantity of the article is small, the conversion rate of the article is determined to be low; the ratio of the order-returning quantity of the article to the order quantity of the article is the order-returning rate, and the order-returning rate of the article is high, which indicates that the article may have quality problems; when the quantity of the articles transferred out is large and the current residual quantity is small, the articles are indicated to be sold out in advance before the current period is ended; when the current remaining quantity of the article is large, but the output quantity is small, the article is indicated to have the problem of high failure rate. Therefore, multiple types of abnormal probability acquisition models are set for different types of abnormal states, and multiple types of articles in abnormal states can be detected, so that the articles in abnormal states can be processed subsequently, and the influence of multiple types of abnormal states on benefits can be avoided.

In addition, when determining the probability of the candidate item identifier, the anomaly probability obtaining model may predict the probability of the candidate item identifier at a plurality of times between the current time and the end time of the current period. For example, each period is 8 to 20 points, and at the current time 10 points, the probability that any integer point between 10 to 20 points is in an abnormal state can be obtained by the abnormal probability obtaining model.

503. And the server selects at least one item identifier from the multiple candidate item identifiers according to the probability of the multiple candidate item identifiers.

And the server selects at least one article identifier in an abnormal state from the multiple candidate article identifiers according to the probability of the multiple candidate article identifiers, so that the unit transfer value corresponding to the selected article identifier can be adjusted subsequently.

For the selected manner of item identification, in one possible implementation, the step 503 includes: and selecting at least one article identifier with the probability belonging to a preset probability range from the plurality of candidate article identifiers. The preset probability range may be set by a manager or calculated by historical data. For example, the preset probability range is (0.7, 1), and when the probability of the candidate item identifier is 0.8 and belongs to the preset probability range, it is determined that the roll-out condition of the item corresponding to the candidate item identifier is in an abnormal state.

In another possible implementation, the step 503 includes: and selecting at least one item identifier with the probability greater than a preset threshold value from the plurality of candidate item identifiers.

In another possible implementation, the step 503 includes: and selecting a third preset number of article identifications from the multiple candidate article identifications according to the probability of the multiple candidate article identifications, wherein the probability of the preset number of article identifications is greater than the probability of other candidate article identifications in the multiple candidate article identifications.

Due to the limitation of the trading platform, the unit transfer value of the articles corresponding to the third preset number of article identifiers can be adjusted each time, the higher the probability of each article identifier is, the more serious the transfer-out condition of the corresponding article is in an abnormal state, and therefore, the third preset number of article identifiers with higher probability is selected from the multiple alternative article identifiers.

In addition, the manager may set a flag state for each article id, where the flag state includes a first flag state indicating that the unit transfer value of the article id is not allowed to be adjusted and a second flag state indicating that the unit transfer value of the article id needs to be adjusted. Based on the selection modes of the three article identifiers, when the selected at least one article identifier meets any one of the three modes, it is also required that the at least one article identifier is not in the first marking state. Alternatively, the step of selecting at least one item identifier in the second marked state is performed without performing the steps 501-503.

504. For each item identifier in the at least one item identifier, the server obtains a corresponding relation between a history unit transfer numerical value and a history transfer-out quantity included in history data of the item identifier.

The information set corresponding to each article identifier may include historical data, and the historical data may include a correspondence between a transfer value of a history unit in a plurality of cycles and a transfer amount of the history, or may include a correspondence between a transfer value of a history unit in a plurality of periods in a plurality of cycles and a transfer amount of the history.

The historical unit transfer value can be the historical unit price of the article, and the historical transfer-out quantity can be the historical sales quantity of the article. That is, the server may obtain the unit price and sales volume of the item over the previous time period.

505. The server inquires the corresponding relation according to the target transfer-out quantity of the article identifier, determines a historical unit transfer numerical value corresponding to the historical transfer-out quantity matched with the target transfer-out quantity, and takes the historical unit transfer numerical value as the unit transfer numerical value after the article identifier is adjusted.

For any article, different unit transfer values can be adopted to lead to different transfer-out quantities, so that when the unit transfer value of any article is determined, the target transfer quantity of the article needs to be determined firstly. The target roll-out quantity may be an expected sales quantity of the article. The target transfer-out quantity may be expressed as a transfer-out quantity of an article to which the article identifier is expected to correspond from a current time to a time before an end time of the current cycle, may also be a transfer-out quantity of an article to which the article identifier is expected to correspond in any future time period of the current cycle, and may also be equal to a current remaining quantity of the article to which the article identifier corresponds.

In the article transferring process, in the same time period in different history periods, the transferring-out quantities corresponding to the same unit transferring numerical value have similarity, so that the corresponding relation in the history data is inquired, the history unit transferring numerical value corresponding to the history transferring-out quantity matched with the target transferring-out quantity is determined, and the history unit transferring numerical value is used as the unit transferring numerical value after the article identification is adjusted, so that the article transferring-out quantity after the unit transferring numerical value is adjusted can reach the target transferring-out quantity. In addition, the history data may further include a corresponding relationship between the history unit transfer value of the plurality of history time periods and the history transfer-out number, and when the server queries the corresponding relationship, the server may determine the history unit transfer value corresponding to the history transfer-out number that is matched with the current time period and the target transfer-out number. For example, if the current day of the week is sunday and the target transfer number of the apple is 80, the corresponding relationship is inquired, and the historical unit transfer numerical value corresponding to the same sunday and the historical transfer number of the apple is 80 is selected.

For this determination of the target destage amount, in one possible implementation, the information set may also include a remaining number of items. And determining the difference between the residual quantity of the articles corresponding to the article identification and the target residual quantity at the end moment of the current period as the target roll-out quantity. The target remaining amount may be set for each item identification by the manager and may be included in the information set. For example, the manager sets a target remaining quantity for each item identifier, and stores the target remaining quantity in the corresponding information set in correspondence with the corresponding item identifier.

It should be noted that, in the embodiment of the present application, the adjusted unit transfer value is determined for the selected at least one item identifier according to the historical data, and in another embodiment, the step 504 and the step 506 may be replaced by the following steps: and processing the information set corresponding to the at least one article identifier based on the unit transfer value determination model, and determining the unit transfer value adjusted by the at least one article identifier. The unit transfer value determination model may be a linear regression fitting model, a GBDT (Gradient Boosting Decision Tree), a GNN (Graph Neural Networks), a probability map model, a demand model, or the like. And respectively inputting the information set corresponding to each article identifier into the unit transfer numerical value determination model, and outputting the unit transfer numerical value corresponding to each article identifier by the unit transfer numerical value determination model.

For this training process of the unit transition value determination model, in one possible implementation, the method further includes: and acquiring a sample information set corresponding to the sample article identification, wherein the sample information set comprises unit transfer numerical values of the sample article identification in the first historical period, unit transfer numerical values and transfer-out quantities in other historical periods before the first historical period, and training the unit transfer numerical value determination model according to the sample information set.

And training the unit transfer numerical model by taking the unit transfer numerical values and the transfer-out quantity in other historical periods before the first historical period as the input of the unit transfer numerical value determination model and taking the unit transfer numerical values in the first historical period as the output, so that the trained unit transfer determination model can determine the unit transfer numerical values for the articles corresponding to the article identifications through the information sets corresponding to the article identifications. In addition, the training stop condition for the unit transition value determination model may be: respectively carrying out iterative training on the unit transfer value determination model by a plurality of sample information sets, and stopping the iterative training when the iteration times reach preset times; or stopping the iterative training when the unit transfer value determines that the accuracy of the unit transfer value output by the model reaches a preset threshold.

When the unit transfer value of the article identifier is determined, the unit transfer value determination model can predict the unit transfer value of the article identifier according to different strategy purposes by considering various factors influencing the unit transfer value of the article, such as historical sales, historical unit transfer values, weather, holiday information and the like.

In order to obtain the unit transfer value of the item identifier and maximize the profit value, when the unit transfer value is determined for the item, the unit transfer value determination model can predict the transfer-out quantity and the profit value which can be realized by the item in the current period or the future period according to the corresponding relation between the historical unit transfer value and the historical transfer-out quantity in the historical data, so as to determine the unit transfer value of the item.

When the unit transfer numerical value determining model determines the unit transfer numerical value for the article, information of different dimensions of the article, such as the unit transfer numerical value, the transfer quantity, the article type, the income numerical value and the like, can be used as different nodes, the relationship among the nodes can be determined according to historical data, and a probability graph model is constructed; when a unit transfer value is determined for an article, a node characteristic value of each node corresponding to the article in a current period or a future period is respectively predicted, a plurality of unit transfer values and the probability of each unit transfer value are determined according to the node characteristic value of each node and the relation between the nodes based on a probability graph model, and the unit transfer value with the maximum probability is determined as the unit transfer value of the article.

It should be noted that, when at least one article identifier includes a plurality of article identifiers and the adjusted unit transfer values of the article identifiers are obtained, after step 506, the method further includes the following two manners:

the first mode is as follows: dividing the plurality of item identifications into a plurality of item identification combinations so that each item identification combination comprises a first preset number of item identifications; counting a total transfer value corresponding to each article identification combination according to the unit transfer value and the target transfer-out quantity of each article identification after adjustment; and taking the article identifier in the article identifier combination with the maximum total transfer value as a target article identifier, and issuing the unit transfer value adjusted by the target article identifier.

Due to the flow limitation of the trading platform, the unit transfer value of the first preset number of article identifications can be adjusted each time. Therefore, in order to ensure the adjustment effect of the unit transfer value and maximize the profit value, the screening is performed on the plurality of article identifiers, and the unit transfer value of the first preset number of article identifiers after the screening is issued.

For example, 10 items are combined respectively, each combination includes 3 items, the unit price of each item is multiplied by the target sales amount of the item to serve as the income amount of the item, the income amounts of the 3 items included in each combination are added to obtain the total income amount corresponding to each combination, and the adjusted unit price of the item in the item combination with the maximum total income amount is issued.

The second mode is as follows: dividing the plurality of article identifications into a plurality of article identification combinations so that each article identification combination comprises a second preset number of article identifications; according to the adjusted unit transfer value, unit cost value and target transfer-out quantity of each article identifier, counting a total income value corresponding to each article identifier combination; and taking the article identifier in the article identifier combination with the maximum total profit value as a target article identifier, and issuing the unit transfer value adjusted by the target article identifier.

For example, 10 items are respectively combined, each combination comprises 3 items, the difference between the unit price of each item and the corresponding cost is determined, the difference is multiplied by the corresponding target sales amount to be used as the income amount of the item, the income amounts of the 3 items in each combination are added to obtain the total income amount corresponding to each combination, and the adjusted unit price of the item in the item combination with the maximum total income amount is released.

In addition, issuing the adjusted unit transfer value of the target item identifier may include: after the target object identification is determined, an adjustment request is sent to a management terminal, the adjustment request carries the target object identification and the corresponding adjusted unit transfer numerical value, and after a confirmation notice sent by the management terminal is received, the unit transfer numerical value adjusted by the target object identification is issued, so that the user terminal receives the unit transfer numerical value adjusted by the target object identification for the user to check.

The method provided by the embodiment of the application comprises the steps of obtaining an information set corresponding to a plurality of candidate item identifications, wherein the information set at least comprises area identifications of geographic areas to which the corresponding items belong, the area identifications in the information sets are the same, respectively determining the probabilities of the candidate item identifications based on an abnormal probability obtaining model matched with the area identifications, the probabilities are used for representing the probability that the corresponding item transfer-out conditions are in abnormal states, selecting at least one item identification from the candidate item identifications according to the probabilities of the candidate item identifications, and adjusting unit transfer values corresponding to the at least one item identification. The unit transfer value of the article does not need to be determined manually, the unit transfer value after the article is adjusted is prevented from being influenced by subjective factors, and the accuracy is improved. And the article needing to adjust the unit transfer value is determined according to the probability that the article is in the abnormal state under the condition of transferring out, so that the article in the abnormal state can be adjusted in time, and the loss caused by the article in the abnormal state is avoided.

It should be noted that, in the embodiment of the present application, after the unit transfer value adjusted by the target article identifier is determined, the unit transfer value adjusted by the target article identifier is issued for explanation, and in another embodiment, after the unit transfer value adjusted by the target article identifier is issued, the remaining quantity and the transferring quantity of the target article identifier are monitored in real time, and the transfer value or the profit value corresponding to the target article identifier can be determined, so that the adjustment frequency of the unit transfer value of the target article identifier and the adjustment amplitude of the unit transfer value can be dynamically adjusted. For example, the remaining quantity and the roll-out quantity of the target article identifier are monitored in real time, whether the target article identifier can reach an expected target or not is predicted, for example, an expected roll-out value or an expected profit value is obtained, when it is determined that the currently adjusted unit transfer value cannot reach the expected target, the information set of the target article identifier is obtained again, the unit transfer value of the target article identifier is adjusted again, and the adjustment range of the unit transfer value can be increased.

Fig. 6 is a schematic structural diagram of a unit transition value adjusting apparatus according to an embodiment of the present application, as shown in fig. 6, the apparatus includes:

an information set obtaining module 601, configured to obtain information sets corresponding to multiple candidate item identifiers, where the information sets at least include area identifiers of geographic areas to which corresponding items belong, and the area identifiers included in the multiple information sets are the same;

a probability determining module 602, configured to process the acquired information set based on the abnormal probability acquisition model matched with the region identifier, and respectively determine probabilities of multiple candidate item identifiers, where the probabilities are used to indicate probabilities that corresponding item roll-out situations are in abnormal states;

a transfer value adjusting module 603, configured to select at least one item identifier from the multiple candidate item identifiers according to the probabilities of the multiple candidate item identifiers, and adjust a unit transfer value corresponding to the at least one item identifier.

The device provided by the embodiment of the application obtains the information sets corresponding to the multiple candidate article identifications, the information sets at least comprise the area identifications of the geographic areas to which the corresponding articles belong, the area identifications in the multiple information sets are the same, the probability of the multiple candidate article identifications is respectively determined based on the abnormal probability obtaining model matched with the area identifications, the probability is used for representing the probability that the corresponding article roll-out condition is in an abnormal state, at least one article identification is selected from the multiple candidate article identifications according to the probability of the multiple candidate article identifications, and the unit transfer value corresponding to the at least one article identification is adjusted. The unit transfer value of the article does not need to be determined manually, the unit transfer value after the article is adjusted is prevented from being influenced by subjective factors, and the accuracy is improved. And the article needing to adjust the unit transfer value is determined according to the probability that the article is in the abnormal state under the condition of transferring out, so that the article in the abnormal state can be adjusted in time, and the loss caused by the article in the abnormal state is avoided.

In one possible implementation, as shown in fig. 7, the information set obtaining module 601 includes:

an information set obtaining unit 6101, configured to obtain an information set corresponding to a plurality of item identifiers, where the information set includes at least an area identifier of a geographic area to which a corresponding item belongs, a current feature value, and a historical feature value, and the area identifiers included in the plurality of information sets are the same, and the feature value of the item includes at least one of a unit transfer value, a transfer-out quantity, or a remaining quantity of the item;

a first article identifier determining unit 6102, configured to count, for each article identifier, a historical feature value corresponding to the article identifier, so as to obtain an abnormal value range of the historical feature value; and if the current characteristic numerical value belongs to the abnormal numerical value range, determining the article identification as the alternative article identification.

In another possible implementation manner, as shown in fig. 7, a geographic area corresponding to the area identifier includes a plurality of warehouses, and warehouse identifiers of different warehouses are different; the information collection obtaining module 601 includes:

a second item identifier determining unit 6103, configured to, for any one of the candidate item identifiers, obtain multiple initial information sets of the candidate item identifier, and fuse the multiple initial information sets to serve as an information set corresponding to the candidate item identifier;

In another possible implementation manner, as shown in fig. 7, the transfer value adjusting module 603 includes:

a first transfer value adjusting unit 6301, configured to process the information set corresponding to the at least one article identifier based on the unit transfer value determination model, and determine a unit transfer value after the adjustment of the at least one article identifier.

In another possible implementation, as shown in fig. 7, an apparatus includes:

a sample information set obtaining module 604, configured to obtain a sample information set corresponding to the sample article identifier, where the sample information set includes a unit transfer value of the sample article identifier in a first history period, and unit transfer values and transfer-out quantities in other history periods before the first history period;

and the model training module 605 is configured to train the unit transfer value determination model according to the sample information set.

a correspondence obtaining unit 6302 configured to obtain, for each of the at least one item identifier, a correspondence between a history unit transfer value and a history transfer-out amount included in history data of the item identifier;

a value determining unit 6303, configured to query the corresponding relationship according to the target roll-out quantity of the item identifier, and determine a historical unit transfer value corresponding to the historical roll-out quantity matched with the target roll-out quantity;

a second transfer value adjusting unit 6304, configured to use the history unit transfer value as the unit transfer value after the item identifier is adjusted.

In another possible implementation, as shown in FIG. 7, the information set also includes a remaining number of items; the transition value adjusting module 603 includes:

a target roll-out number determining unit 6305, configured to determine, as the target roll-out number, a difference between the remaining number of items identified by the item and the target remaining number at the end time of the current cycle.

In another possible implementation, as shown in fig. 7, the at least one item identifier includes a plurality of item identifiers; the device still includes:

a first combination dividing module 605, configured to divide the plurality of item identifiers into a plurality of item identifier combinations, so that each item identifier combination includes a first preset number of item identifiers;

a transfer value counting module 606, configured to count a total transfer value corresponding to each article identifier combination according to the unit transfer value and the target transfer-out quantity adjusted by each article identifier;

the first value issuing module 607 is configured to issue the unit transfer value adjusted by the target item identifier, using the item identifier in the item identifier combination with the largest total transfer value as the target item identifier.

a second combination dividing module 608, configured to divide the plurality of item identifiers into a plurality of item identifier combinations, so that each item identifier combination includes a second preset number of item identifiers;

a profit value counting module 609, configured to count a total profit value corresponding to each article identifier combination according to the adjusted unit transfer value, unit cost value, and target transfer-out quantity of each article identifier;

and a second value issuing module 610, configured to issue the unit transfer value after the adjustment of the target item identifier, with the item identifier in the item identifier combination with the largest total profit value as the target item identifier.

the first item identifier selecting unit 6306 is configured to select, from the multiple candidate item identifiers, at least one item identifier whose probability belongs to a preset probability range.

In another possible implementation manner, as shown in fig. 7, the first item identifier selecting unit 6306 is further configured to select, from the multiple candidate item identifiers, at least one item identifier with a probability greater than a preset threshold.

the second item identifier selecting unit 6307 is configured to select a third preset number of item identifiers from the multiple candidate item identifiers according to the probabilities of the multiple candidate item identifiers, where the probability of the preset number of item identifiers is greater than the probabilities of other candidate item identifiers in the multiple candidate item identifiers.

In another possible implementation manner, the abnormal probability obtaining model includes at least one of a conversion probability obtaining model, a drop probability obtaining model, a sold-out probability obtaining model or a loss reporting probability obtaining model.

Fig. 8 is a schematic structural diagram of a terminal according to an embodiment of the present application, which can implement operations executed by the first terminal, the second terminal, and the third terminal in the foregoing embodiments. The terminal 800 may be a portable mobile terminal such as: the mobile terminal comprises a smart phone, a tablet computer, an MP3 player (Moving Picture Experts Group Audio Layer III, Moving Picture Experts compress standard Audio Layer 3), an MP4 player (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts compress standard Audio Layer 4), a notebook computer, a desktop computer, a head-mounted device, a smart television, a smart sound box, a smart remote controller, a smart microphone, or any other smart terminal. The terminal 800 may also be referred to by other names such as user equipment, portable terminal, laptop terminal, desktop terminal, etc.

In general, the terminal 800 includes: a processor 801 and a memory 802.

The processor 801 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and so forth. The memory 802 may include one or more computer-readable storage media, which may be non-transitory, for storing at least one instruction for the processor 801 to have in implementing the unit transfer value adjustment methods provided by method embodiments herein.

In some embodiments, the terminal 800 may further include: a peripheral interface 803 and at least one peripheral. The processor 801, memory 802 and peripheral interface 803 may be connected by bus or signal lines. Various peripheral devices may be connected to peripheral interface 803 by a bus, signal line, or circuit board. Specifically, the peripheral device includes: at least one of radio frequency circuitry 804, display 805, and audio circuitry 806.

The Radio Frequency circuit 804 is used for receiving and transmitting RF (Radio Frequency) signals, also called electromagnetic signals. The radio frequency circuit 804 communicates with communication networks and other communication devices via electromagnetic signals.

The display screen 805 is used to display a UI (user interface). The UI may include graphics, text, icons, video, and any combination thereof. The display 805 may be a touch display and may also be used to provide virtual buttons and/or a virtual keyboard.

The audio circuitry 806 may include a microphone and a speaker. The microphone is used for collecting audio signals of a user and the environment, converting the audio signals into electric signals, and inputting the electric signals to the processor 801 for processing, or inputting the electric signals to the radio frequency circuit 804 to realize voice communication. For the purpose of stereo sound collection or noise reduction, a plurality of microphones may be provided at different portions of the terminal 800. The microphone may also be an array microphone or an omni-directional pick-up microphone. The speaker is used to convert electrical signals from the processor 801 or the radio frequency circuit 804 into audio signals.

Those skilled in the art will appreciate that the configuration shown in fig. 8 is not intended to be limiting of terminal 800 and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components may be used.

Fig. 9 is a schematic structural diagram of a server according to an embodiment of the present application, where the server 900 may generate a relatively large difference due to a difference in configuration or performance, and may include one or more processors (CPUs) 901 and one or more memories 902, where the memory 902 stores at least one instruction, and the at least one instruction is loaded and executed by the processors 901 to implement the methods provided by the foregoing method embodiments. Of course, the server may also have components such as a wired or wireless network interface, a keyboard, and an input/output interface, so as to perform input/output, and the server may also include other components for implementing the functions of the device, which are not described herein again.

The server 900 may be used to perform the unit shift value adjustment method described above.

The embodiment of the present application further provides a computer device, where the computer device includes a processor and a memory, where the memory stores at least one program code, and the at least one program code is loaded and executed by the processor, so as to implement the unit transfer value adjustment method of the foregoing embodiment.

The embodiment of the present application further provides a computer-readable storage medium, where at least one program code is stored in the computer-readable storage medium, and the at least one program code is loaded and executed by a processor, so as to implement the unit transfer value adjustment method of the foregoing embodiment.

The embodiment of the present application further provides a computer program, where at least one program code is stored in the computer program, and the at least one program code is loaded and executed by a processor, so as to implement the unit transfer value adjustment method of the foregoing embodiment.

It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

The above description is only an alternative embodiment of the present application and should not be construed as limiting the present application, and any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for adjusting a value of a unit transition, the method comprising:

2. The method according to claim 1, wherein the obtaining of the information sets corresponding to the plurality of candidate item identifiers comprises:

3. The method according to claim 1, wherein the geographic area corresponding to the area identifier comprises a plurality of warehouses, and the warehouse identifiers of different warehouses are different; the acquiring of the information sets corresponding to the multiple candidate item identifiers includes:

4. The method of claim 1, wherein said adjusting the unit transfer value corresponding to said at least one item identifier comprises:

5. The method of claim 4, wherein the set of information further comprises a remaining number of items;

6. The method of claim 1, wherein the at least one item identifier comprises a plurality of item identifiers; after the adjusting the unit transfer value corresponding to the at least one article identifier, the method further includes:

7. The method of claim 1, wherein the at least one item identifier comprises a plurality of item identifiers; after the adjusting the unit transfer value corresponding to the at least one article identifier, the method further includes:

8. A unit shift value adjusting apparatus, comprising:

9. A computer device comprising a processor and a memory, the memory having stored therein at least one program code, the at least one program code loaded into and executed by the processor, to implement a unit transfer value adjustment method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium having at least one program code stored therein, the at least one program code being loaded and executed by a processor to implement the unit transfer value adjustment method according to any one of claims 1 to 7.