CN110070382A - Method and apparatus for generating information - Google Patents
Method and apparatus for generating information Download PDFInfo
- Publication number
- CN110070382A CN110070382A CN201810069211.9A CN201810069211A CN110070382A CN 110070382 A CN110070382 A CN 110070382A CN 201810069211 A CN201810069211 A CN 201810069211A CN 110070382 A CN110070382 A CN 110070382A
- Authority
- CN
- China
- Prior art keywords
- product
- information
- feature vector
- sales volume
- sales
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 108
- 238000012549 training Methods 0.000 claims abstract description 45
- 230000004044 response Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000003066 decision tree Methods 0.000 description 23
- 230000006870 function Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 9
- 230000015572 biosynthetic process Effects 0.000 description 7
- 238000003786 synthesis reaction Methods 0.000 description 7
- 238000004378 air conditioning Methods 0.000 description 6
- 230000006854 communication Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000005057 refrigeration Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000003197 gene knockdown Methods 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 230000005291 magnetic effect Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 230000003442 weekly effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 210000003127 knee Anatomy 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0206—Price or cost determination based on market factors
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the present application discloses the method and apparatus for generating information.One specific embodiment of this method includes: reception predictions request, and predictions request includes the attribute information, price related information and category information of target product；According to the attribute information and price related information of target product, the feature vector of target product is generated；Feature vector is input to training in advance, corresponding with category information Method for Sales Forecast model, generates the sales volume information of target product, wherein Method for Sales Forecast model is used to characterize the corresponding relationship between the feature vector of product and the sales volume information of product.This embodiment improves the accuracys of sales volume information.
Description
Technical field
The invention relates to field of computer technology, and in particular to Internet technical field, it is more particularly, to raw
At the method and apparatus of information.
Background technique
Supply Chain Planner is the pool and planning that enterprise carries out supply chain execution, starts to work it in each node of supply chain
Before, the action plan of the various aspects such as demand, production, logistics, inventory is formulated for supply chain end to end, and in supply chain index
When occurring abnormal, carry out it is unsalable clear up, scrap, the plan that supplier's return of goods etc. are coped with extremely.
The core of Supply Chain Planner is Method for Sales Forecast, no matter the scale of enterprise, personnel how much, Method for Sales Forecast influences
Including various works such as plan, production and inventories.Traditional Method for Sales Forecast method is mainly by nearest a period of time (example
Such as, nearest one week or one month) sales volume be weighted and averaged to predict following a period of time (for example, one week or one following
Month) sales volume.
Summary of the invention
The embodiment of the present application proposes the method and apparatus for generating information.
In a first aspect, the embodiment of the present application provides a kind of method for generating information, this method comprises: receiving prediction
Request, predictions request includes the attribute information, price related information and category information of target product；According to the attribute of target product
Information and price related information generate the feature vector of target product；Feature vector is input to train in advance and category to believe
Corresponding Method for Sales Forecast model is ceased, the sales volume information of target product is generated, wherein Method for Sales Forecast model is used to characterize the spy of product
Levy the corresponding relationship between vector and the sales volume information of product.
In some embodiments, training obtains Method for Sales Forecast model corresponding with category information as follows: obtaining
The sales volume data of at least one product within a preset period of time under the affiliated category of target product, wherein preset time period includes extremely
A few sub- period；For each product at least one product, the product is extracted from the sales volume data of the product and is existed
Attribute information, price related information and history sales volume information in each sub- period, and it is based on attribute information and price phase
It closes information and generates feature vector of the product within each sub- period；By feature of each product within each sub- period to
Amount is as input, and using history sales volume information of each product within the corresponding sub- period as output, training obtains Method for Sales Forecast
Model.
In some embodiments, Method for Sales Forecast model includes more regression trees, and kth regression tree passes through as follows
Step training obtains: for each feature vector of each product in the feature vector in each sub- period, being based on the spy
The output valve and this feature vector that vector is levied in output valve to -1 regression tree of kth of first regression tree are corresponding
History sales volume information, determine this feature vector in the target value of kth regression tree, wherein k is integer greater than 1；It will
Feature vector of each product within each sub- period is as input, by each feature vector in kth regression tree
Target value obtains kth regression tree as output, training.
In some embodiments, the training step of Method for Sales Forecast model corresponding with category information further include: obtain target
The feature vector of different product and history sales volume information under the affiliated category of product；The feature vector of every kind of product is input to k
Regression tree obtains the prediction sales volume information of every kind of product；Based on history sales volume information and prediction sales volume information, every kind is determined
The accuracy rate of the sales volume information of product；It is more than or equal to k-1 in response to the difference between the accuracy rate of the sales volume information of different product
Regression tree determines the difference between the accuracy rate of the sales volume information of different product, and k-1 regression tree is determined as
Method for Sales Forecast model corresponding with category information.
Second aspect, the embodiment of the present application provide a kind of for generating the device of information, request reception unit, configuration use
In receiving predictions request, predictions request includes the attribute information, price related information and category information of target product；Vector generates
Unit is configured to attribute information and price related information according to target product, generates the feature vector of target product；Information
Generation unit is configured to for feature vector being input to Method for Sales Forecast model train in advance, corresponding with category information, generates
The sales volume information of target product, wherein Method for Sales Forecast model is for characterizing between the feature vector of product and the sales volume information of product
Corresponding relationship.
In some embodiments, training obtains Method for Sales Forecast model corresponding with category information as follows: obtaining
The sales volume data of at least one product within a preset period of time under the affiliated category of target product, wherein preset time period includes extremely
A few sub- period；For each product at least one product, the product is extracted from the sales volume data of the product and is existed
Attribute information, price related information and history sales volume information in each sub- period, and it is based on attribute information and price phase
It closes information and generates feature vector of the product within each sub- period；By feature of each product within each sub- period to
Amount is as input, and using history sales volume information of each product within the corresponding sub- period as output, training obtains Method for Sales Forecast
Model.
In some embodiments, Method for Sales Forecast model includes more regression trees, and kth regression tree passes through as follows
Step training obtains: for each feature vector of each product in the feature vector in each sub- period, being based on the spy
The output valve and this feature vector that vector is levied in output valve to -1 regression tree of kth of first regression tree are corresponding
History sales volume information, determine this feature vector in the target value of kth regression tree, wherein k is integer greater than 1；It will
Feature vector of each product within each sub- period is as input, by each feature vector in kth regression tree
Target value obtains kth regression tree as output, training.
In some embodiments, the training step of Method for Sales Forecast model corresponding with category information further include: obtain target
The feature vector of different product and history sales volume information under the affiliated category of product；The feature vector of every kind of product is input to k
Regression tree obtains the prediction sales volume information of every kind of product；Based on history sales volume information and prediction sales volume information, every kind is determined
The accuracy rate of the sales volume information of product；It is more than or equal to k-1 in response to the difference between the accuracy rate of the sales volume information of different product
Regression tree determines the difference between the accuracy rate of the sales volume information of different product, and k-1 regression tree is determined as
Method for Sales Forecast model corresponding with category information.
Method and apparatus provided by the embodiments of the present application for generating information, pass through the attribute information according to target product
The feature vector that target product is generated with price related information, is then input to target product institute for the feature vector of target product
The Method for Sales Forecast model for belonging to category obtains the sales volume information of target product, to improve the accuracy of sales volume information.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is that this application can be applied to exemplary system architecture figures therein；
Fig. 2 is the flow chart according to one embodiment of the method for generating information of the application；
Fig. 3 A and Fig. 3 B are the schematic diagrames according to the Method for Sales Forecast model of the method for generating information of the application；
Fig. 4 is the schematic diagram according to an application scenarios of the method for generating information of the application；
Fig. 5 is the structural schematic diagram according to one embodiment of the device for generating information of the application；
Fig. 6 is adapted for the structural schematic diagram for the computer system for realizing the terminal of the embodiment of the present application.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can be using the method for generating information of the application or the implementation of the device for generating information
The exemplary system architecture 100 of example.
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out
Send message etc..Various telecommunication customer end applications can be installed, such as shopping class is applied, searched on terminal device 101,102,103
The application of rope class, instant messaging tools, mailbox client, social platform software etc..
Terminal device 101,102,103 can be various electronic equipments, including but not limited to smart phone, tablet computer,
E-book reader, MP3 player (Moving Picture Experts Group Audio Layer III, dynamic image
Expert's compression standard audio level 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic shadow
As expert's compression standard audio level 4) player, pocket computer on knee and desktop computer etc..
Server 105 can be to provide the server of various services, and for example, prediction sales volume information provides the backstage supported
Server.Server 105 can be analyzed and processed the data such as the predictions request received, and by processing result (for example, pin
Amount information) feed back to terminal device.
It should be noted that the method provided by the embodiment of the present application for generating information is generally held by server 105
Row, correspondingly, the device for generating information is generally positioned in server 105.
It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
With continued reference to Fig. 2, the process of one embodiment of the method for generating information according to the application is shown
200.The method for being used to generate information, comprising the following steps:
Step 201, predictions request is received, predictions request includes the attribute information, price related information and product of target product
Category information.
In the present embodiment, the method for generating information runs electronic equipment (such as service shown in FIG. 1 thereon
Device) can be received by wired connection mode or radio connection from terminal include target product attribute information, price
The predictions request of relevant information and category information.Wherein, target product can be products in kind (for example, mobile phone) and be also possible to void
Quasi- product (for example, service)；The category information of product is the description information of product category, product category, that is, product category, one
Product category refers to one group of product that is associated and/or being substituted for each other, such as " mobile phone " category, " tablet computer " category,
" air-conditioning " category etc.；The attribute information of product includes each attribute of product, and the attribute information of the product of different categories can not
Together, for example, the attribute information of the product of " air-conditioning " category may include brand, model, color, classification, be applicable in area, refrigeration
Amount, refrigeration work consumption, heating capacity, heats power, efficiency grade etc.；Price related information refers to letter associated with the price of product
Breath, for example, unit price, knock-down price, completely subtracting, landing vertically, completely give, the information such as present, coupon, suit.
It should be pointed out that above-mentioned radio connection can include but is not limited to 3G/4G connection, WiFi connection, bluetooth
Connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection and other currently known or exploitations in the future
Radio connection.
Step 202, according to the attribute information of target product and price related information, the feature vector of target product is generated.
In the present embodiment, it is based on the received predictions request of step 201, above-mentioned electronic equipment (such as service shown in FIG. 1
Device) can be extracted from the attribute information and price related information of target product influence product sales volume key feature (for example, certain
Certain brand, is applicable in area 70m at black2, the attribute informations such as level-one efficiency, and unit price≤3000, often expire 1000 and subtract 50, land vertically
500, give quality guarantee in 2 years, the price related informations such as coupon can be used), the key feature based on extraction generates target and produces later
The feature vector of product.
Step 203, feature vector is input to training in advance, corresponding with category information Method for Sales Forecast model, generated
The sales volume information of target product, wherein Method for Sales Forecast model is for characterizing between the feature vector of product and the sales volume information of product
Corresponding relationship.
In the present embodiment, the method for generating information runs electronic equipment (such as service shown in FIG. 1 thereon
Device) Method for Sales Forecast model to be used can be determined according to the category information in predictions request received in step 201, then will
The feature vector generated in step 202 is input to identified Method for Sales Forecast model, generates the sales volume information (example of target product
Such as, sales volume of the target product at following one week).
Above-mentioned Method for Sales Forecast model can be using machine learning algorithm, based on training sample to the mould for numerical prediction
Type is (for example, regression tree (Regression Decision Tree), promotion decision tree (Boosting Decision
Tree) etc.) it is trained rear obtained model.
As an example, above-mentioned Method for Sales Forecast model can be regression tree (Regression Decision
Tree) model.Regression tree is a kind of machine learning algorithm of structure progress decision (that is, prediction) based on tree, generally,
One regression tree includes root node (root node), multiple internal nodes (also referred to as non-leaf nodes, a non-leaf
) and multiple leaf nodes (leaf nodes) nodes.The input space is divided into different regions by regression tree, and each region has
Independent parameter, each node of regression tree and a region of the input space are associated, and internal node continue by
Region is divided into the subregion (splitting region usually using reference axis) under child node.Thus space is subdivided into nonoverlapping region,
One-to-one relationship is formed between leaf node and input area.Regression tree each feature (example of exhaustion when carrying out branch
Such as, extracted key feature) each threshold value find best cut-point, standards of measurement (that is, loss function) can be with
It is error sum of squares, most reliable branch foundation can be found by minimizing square error.
With reference to Fig. 3 A, it illustrates showing according to a Method for Sales Forecast model of the method for generating information of the application
It is intended to.In the regression tree of Fig. 3 A, whether " completely subtracting " is root node, " coupon ", " present ", " range of decrease >=20% ", " black
Color ", " suit ", " unit price >=99 " etc. are internal node, " 300 ", " 200 ", " 260 ", " 180 ", " 68 ", " 36 ", " 24 ", " 4 "
Deng for leaf node (that is, output valve).
In some optional implementations of the present embodiment, Method for Sales Forecast model corresponding with category information can pass through
Following steps training obtains: above-mentioned electronic equipment can obtain the affiliated category of target product (can determine by category information) first
Under sales volume data of at least one product in preset time period (for example, 1 year, 2 years etc.), wherein preset time period include
At least one sub- period；Later, for each product at least one product, extracting from the sales volume data of the product should
Attribute information, price related information and history sales volume of the product within each sub- period (for example, daily, weekly, monthly etc.)
Information, and the key feature for influencing sales volume is extracted (as previously mentioned, including extremely from above-mentioned attribute information and price related information
A few attribute information and at least one price related information), the key feature based on extraction generates the product in each period of the day from 11 p.m. to 1 a.m
Between feature vector in section；Using feature vector of each product within each sub- period as input, by each product in phase
Answer the history sales volume information in the sub- period as output, training obtains Method for Sales Forecast model.
As another example, above-mentioned Method for Sales Forecast model, which can be, promotes decision tree (Boosting Decision Tree)
Model.Promoting decision tree is also a kind of machine learning algorithm, and by more regression trees of iteration come Shared Decision Making, each is returned
Return decision tree learning is the residual error of the sum of predicted value of all regression trees before, and fitting obtains current residual error recurrence and determines
Therefore plan tree promotes the cumulative of the predicted value that decision tree is the regression tree generated in entire iterative process.By cumulative more
Regression tree can reduce Method for Sales Forecast model over-fitting, improve the accuracy of sales volume information.
With reference to Fig. 3 B, it illustrates showing according to another Method for Sales Forecast model of the method for generating information of the application
It is intended to.In figure 3b, the structure of every regression tree is similar with regression tree shown in Fig. 3 A, the difference is that, often
The target value of regression tree (in addition to first regression tree) is all the sum of the output valve of all regression trees in front
Residual error, for example, being the sample of " 300 " for sales volume information, the output valve of first regression tree is " 280 ", then second
The target value of regression tree is " 20 " (that is, 300-280=20), further, third regression tree (not shown)
Target value be " 2 " (that is, 300-280-18=2, wherein " 18 " be second regression tree output valve).Pass through iteration
More regression trees, it is possible to reduce the internal node quantity of every regression tree, to effectively avoid over-fitting.
In some optional implementations of the present embodiment, Method for Sales Forecast model corresponding with category information includes more
Regression tree, above-mentioned Method for Sales Forecast model can be trained as follows and be obtained: above-mentioned electronic equipment can obtain first
Feature vector (generating process of feature vector as previously described) and each product of each product within each sub- period are in phase
Answer the history sales volume information in the sub- period；Later the feature vector using each product within each sub- period as input,
Using history sales volume information of each product within the corresponding sub- period as output, training obtains first regression tree；So
Afterwards, the feature vector using each product within each sub- period is as input, by the output valve of first regression tree (or
Claim predicted value) residual error (for example, it is desirable to the difference for being worth (i.e. history sales volume information) Yu output valve) as second recurrence decision
The output of tree, training obtain second regression tree；Using feature vector of each product within each sub- period as defeated
Enter, by the residual error of first regression tree to the sum of the output valve of second regression tree (for example, it is desirable to be worth (i.e. history
Sales volume information) subtract the difference of the sum of output valve) output as third regression tree, training obtains third recurrence and determines
Plan tree；..., and so on, training obtains more regression trees, to obtain Method for Sales Forecast model.
Here, the output (that is, target value) of kth (k is the integer greater than 1) regression tree is true in the following way
It is fixed: for each feature vector of each product in the feature vector in each sub- period, based on this feature vector the
The output valve of one regression tree to -1 regression tree of kth the corresponding history sales volume of output valve and this feature vector
Information (for example, the output valve of history sales volume information and first regression tree to -1 regression tree of kth output valve it
The difference of sum) determine this feature vector in the target value of kth regression tree.
Specifically, for any feature vector X of inputi, in the predicted value f (X of K regression treei) are as follows:
Wherein, K is the quantity of regression tree, and k is the natural number less than K, fk(Xi) it is feature vector XiIt is returned in kth
Return the output valve of decision tree, fK(Xi) it is feature vector XiIn the predicted value of the K regression tree,It is returned for first
Decision tree to the K-1 regression tree integrated forecasting value.
Then training sample { (X1,y1)；(X2,y2)；…；(Xn,yn) the sum of loss function value are as follows:
Wherein, n is training samples number, and i is the natural number less than n, X1、X2、……、XnFor feature vector, y1、
y2、……、ynFor sales volume information, L (f (Xi),yi) it is training sample (Xi,yi) loss function.
Using Taylor's formula, obtains formula (2) and existThe expansion at place:
Wherein, gi、hiRespectivelyFirst derivative and second dervative.
To can determine the optimal solution of loss function, the Sale Forecasting Model of K regression tree composition is obtained.
As another example, above-mentioned Method for Sales Forecast model can be gradient and promote decision tree (Gradient Boosting
Decision Tree) model.It is similar with decision tree is promoted that gradient promotes decision tree, and passes through more regression trees of iteration
Carry out Shared Decision Making, the difference is that, it is promoted in decision tree in gradient, in order to further avoid over-fitting, only adds up every and return
The sub-fraction for returning the output of decision tree is covered the shortage more by learning (training) several regression trees, slowly approaches expectation
Value.
When Method for Sales Forecast model is that gradient promotes decision-tree model, the training step of above-mentioned Method for Sales Forecast model and promotion
Decision-tree model is similar, the difference is that, it is promoted in decision-tree model in gradient, the output of kth regression tree passes through
As under type determines: for each feature vector of each product in the feature vector in each sub- period, by this feature
The synthesis of the corresponding history sales volume information of vector and this feature vector in first regression tree to -1 regression tree of kth
The difference of output valve is determined as this feature vector in the target value of kth regression tree, wherein this feature vector is at first
The synthesis output valve of regression tree to -1 regression tree of kth is this feature vector in the defeated of -1 regression tree of kth
It is worth out defeated in the synthesis of first regression tree to -2 regression trees of kth with sum of products this feature vector of learning coefficient
The sum being worth out, this feature vector are this feature vector in first recurrence decision in the synthesis output valve of first regression tree
The output valve of tree, learning coefficient are the constant greater than 0 less than 1.
In some optional implementations of the present embodiment, the training step of Method for Sales Forecast model corresponding with category information
Suddenly further include: obtain the feature vector and history sales volume information of different product under the affiliated category of target product；By every kind of product
Feature vector is input to k regression tree, obtains the prediction sales volume information of every kind of product；Based on history sales volume information and prediction
Sales volume information determines the predictablity rate of every kind of product；Be greater than in response to the difference between the predictablity rate of different product etc.
Difference between the predictablity rate that k-1 regression tree determines different product stops the training of regression tree, by k-
1 regression tree is determined as Method for Sales Forecast model corresponding with category information.
With continued reference to the signal that Fig. 4, Fig. 4 are according to the application scenarios of the method for generating information of the present embodiment
Figure.In the application scenarios 400 of Fig. 4, user sends target product (for example, XX air-conditioning) to server 402 by terminal 401
Predictions request, server 402 are primarily based on the attribute information of target product in predictions request (for example, so-and-so brand, black, suitable
With area 70m2, the information such as level-one efficiency) and price related information (for example, unit price≤3000, often expiring 1000 and subtracting 50, land vertically
500, give quality guarantee in 2 years, the information such as coupon can be used) feature vector of target product is generated, and belonging to selection target product
The Method for Sales Forecast model of category (for example, " air-conditioning " category)；Then the feature vector of generation is input to the Method for Sales Forecast of selection
Model generates the sales volume information (for example, following one week sales volume of target product is 320) of target product；Finally by the pin of generation
It measures information and returns to terminal 401.
Method provided by the embodiments of the present application for generating information, passes through the attribute information and price according to target product
Relevant information generates the feature vector of target product, and the feature vector of target product is then input to the affiliated category of target product
Method for Sales Forecast model obtain the sales volume information of target product, to improve the accuracy of sales volume information.
With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, this application provides one kind for generating letter
One embodiment of the device of breath, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, which can specifically answer
For in server.
As shown in figure 5, the device 500 for generating information of the present embodiment includes: request reception unit 501, vector life
At unit 502 and information generating unit 503.Wherein, request reception unit is configured to receive predictions request, and predictions request includes
Attribute information, price related information and the category information of target product；Vector generation unit 502 is configured to according to target product
Attribute information and price related information, generate the feature vector of target product；And be configured to will be special for information generating unit 503
Sign vector is input to training in advance, corresponding with category information Method for Sales Forecast model, generates the sales volume information of target product,
Middle Method for Sales Forecast model is used to characterize the corresponding relationship between the feature vector of product and the sales volume information of product.
It in the present embodiment, can be by wired connection side for generating the request reception unit 501 of the device 500 of information
It includes the pre- of the attribute information of target product, price related information and category information that formula or radio connection are received from terminal
Survey request.Wherein, target product can be products in kind and be also possible to virtual product (for example, service)；The category information of product
The description information of product category, product category, that is, product category, a product category refer to one group it is associated and/or can phase
The product of trans-substitution, " mobile phone " category, " tablet computer " category, " air-conditioning " category etc.；The attribute information of product includes producing
The attribute information of each attribute of product, the product of different categories can be different, for example, the attribute information of the product of " air-conditioning " category
It may include brand, model, color, classification, be applicable in area, refrigerating capacity, refrigeration work consumption, heating capacity, heats power, efficiency grade
Deng；Price related information refers to information associated with the price of product, for example, unit price, knock-down price, completely subtract, land vertically, completely give,
The information such as present, coupon, suit.
In the present embodiment, it is based on the received predictions request of request reception unit 501, above-mentioned vector generation unit 502 can
To extract the key feature for influencing product sales volume from the attribute information and price related information of target product (for example, so-and-so product
Board, is applicable in area 70m at black2, the attribute informations such as level-one efficiency, and unit price≤3000, often expire 1000 subtract 50, land vertically 500,
Give quality guarantee in 2 years, the price related informations such as coupon can be used), the key feature based on extraction generates target product later
Feature vector.
In the present embodiment, above- mentioned information generation unit 503 can be according to the received predictions request of request reception unit 501
In category information determine Method for Sales Forecast model to be used, the feature vector for then generating vector generation unit 502 inputs
To identified Method for Sales Forecast model, the sales volume information (for example, the sales volume of target product at following one week) of target product is generated.
In some optional implementations of the present embodiment, Method for Sales Forecast model corresponding with category information can pass through
Following steps training obtains: above-mentioned electronic equipment can obtain the affiliated category of target product (can determine by category information) first
Under sales volume data of at least one product in preset time period (for example, 1 year, 2 years etc.)；Later, at least one
Each product in product, extracted from the sales volume data of the product product each sub- period (for example, daily, weekly,
Attribute information, price related information and history sales volume information in monthly etc.), and from above-mentioned attribute information letter related to price
The key feature for influencing sales volume is extracted in breath (as previously mentioned, including at least one attribute information letter related at least one price
Breath), the key feature based on extraction generates feature vector of the product within each sub- period；By each product in every height
Feature vector in period is as input, using history sales volume information of each product within the corresponding sub- period as exporting,
Training obtains Method for Sales Forecast model.
In some optional implementations of the present embodiment, Method for Sales Forecast model corresponding with category information includes more
Regression tree, training obtains kth regression tree as follows: for each product within each sub- period
Each feature vector in feature vector, based on this feature vector first regression tree output valve to kth -1 return
The output valve and the product for returning decision tree in the corresponding history sales volume information of this feature vector determine this feature vector in kth
The target value of regression tree, wherein k is the integer greater than 1；Feature vector of each product within each sub- period is made
For input, using each feature vector kth regression tree target value as output, training obtain kth recurrence decision
Tree.
In some optional implementations of the present embodiment, target of each feature vector in kth regression tree
Value be the corresponding history sales volume information of this feature vector and this feature vector first regression tree output valve to kth -1
The difference of the sum of the output valve of regression tree.
In some optional implementations of the present embodiment, target of each feature vector in kth regression tree
Value is that the corresponding history sales volume information of this feature vector is determined with this feature vector in first regression tree to kth -1 recurrence
The difference of the synthesis output valve of plan tree, wherein this feature vector is in first regression tree to -1 regression tree of kth
Comprehensive output valve is this feature vector in the output valve of -1 regression tree of kth and sum of products this feature vector of learning coefficient
In the sum of the synthesis output valve in first regression tree to -2 regression trees of kth, this feature vector is returned at first
The synthesis output valve for returning decision tree is output valve of this feature vector in first regression tree, and learning coefficient is small greater than 0
In 1 constant.
In some optional implementations of the present embodiment, the training step of Method for Sales Forecast model corresponding with category information
Suddenly further include: obtain the feature vector and history sales volume information of different product under the affiliated category of target product；By every kind of product
Feature vector is input to k regression tree, obtains the prediction sales volume information of every kind of product；Based on history sales volume information and prediction
Sales volume information determines the predictablity rate of every kind of product；Be greater than in response to the difference between the predictablity rate of different product etc.
Difference between the predictablity rate that k-1 regression tree determines different product stops the training of regression tree, by k-
1 regression tree is determined as Method for Sales Forecast model corresponding with category information
The device provided by the embodiments of the present application for being used to generate information, passes through the attribute information and price according to target product
Relevant information generates the feature vector of target product, and the feature vector of target product is then input to the affiliated category of target product
Method for Sales Forecast model obtain the sales volume information of target product, to improve the accuracy of sales volume information.
Below with reference to Fig. 6, it illustrates the computer systems for the device/server for being suitable for being used to realize the embodiment of the present application
600 structural schematic diagram.Device/server shown in Fig. 6 is only an example, should not function to the embodiment of the present application and
Use scope brings any restrictions.
As shown in fig. 6, computer system 600 includes central processing unit (CPU) 601, it can be read-only according to being stored in
Program in memory (ROM) 602 or be loaded into the program in random access storage device (RAM) 603 from storage section 608 and
Execute various movements appropriate and processing.In RAM 603, also it is stored with system 600 and operates required various programs and data.
CPU 601, ROM 602 and RAM 603 are connected with each other by bus 604.Input/output (I/O) interface 605 is also connected to always
Line 604.
I/O interface 605 is connected to lower component: the importation 606 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 607 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 608 including hard disk etc.；
And the communications portion 609 of the network interface card including LAN card, modem etc..Communications portion 609 via such as because
The network of spy's net executes communication process.Driver 610 is also connected to I/O interface 605 as needed.Detachable media 611, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 610, in order to read from thereon
Computer program be mounted into storage section 608 as needed.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communications portion 609, and/or from detachable media
611 are mounted.When the computer program is executed by central processing unit (CPU) 601, limited in execution the present processes
Above-mentioned function.
It should be noted that computer-readable medium described herein can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires
Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In this application, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this
In application, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium
Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned
Any appropriate combination.
The calculating of the operation for executing the application can be write with one or more programming languages or combinations thereof
Machine program code, described program design language include object oriented program language-such as Java, Smalltalk, C+
+, it further include conventional procedural programming language-such as " C " language or similar programming language.Program code can
Fully to execute, partly execute on the user computer on the user computer, be executed as an independent software package,
Part executes on the remote computer or executes on a remote computer or server completely on the user computer for part.
In situations involving remote computers, remote computer can pass through the network of any kind --- including local area network (LAN)
Or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as utilize Internet service
Provider is connected by internet).
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard
The mode of part is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor packet
Include request reception unit, vector generation unit and information generating unit.Wherein, the title of these units is not under certain conditions
The restriction to the unit itself is constituted, for example, request reception unit is also described as " receiving the unit of predictions request ".
As on the other hand, present invention also provides a kind of computer-readable medium, which be can be
Included in device described in above-described embodiment；It is also possible to individualism, and without in the supplying device.Above-mentioned calculating
Machine readable medium carries one or more program, when said one or multiple programs are executed by the device, so that should
Device: predictions request is received, predictions request includes the attribute information, price related information and category information of target product；According to
The attribute information and price related information of target product, generate the feature vector of target product；Feature vector is input in advance
Method for Sales Forecast model trained, corresponding with category information, generates the sales volume information of target product, and wherein Method for Sales Forecast model is used
Corresponding relationship between the feature vector of characterization product and the sales volume information of product.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from foregoing invention design, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed herein
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (10)
1. a kind of method for generating information, comprising:
Predictions request is received, the predictions request includes the attribute information, price related information and category information of target product；
According to the attribute information and price related information of the target product, the feature vector of the target product is generated；
Described eigenvector is input to training in advance, corresponding with category information Method for Sales Forecast model, described in generation
The sales volume information of target product, wherein Method for Sales Forecast model is for characterizing between the feature vector of product and the sales volume information of product
Corresponding relationship.
2. according to the method described in claim 1, wherein, Method for Sales Forecast model corresponding with the category information by walking as follows
Rapid training obtains:
The sales volume data of at least one product within a preset period of time under the affiliated category of the target product are obtained, wherein described
Preset time period includes at least one sub- period；
For each product at least one described product, the product is extracted from the sales volume data of the product in each period of the day from 11 p.m. to 1 a.m
Between attribute information, price related information and history sales volume information in section, and it is raw based on attribute information and price related information
At feature vector of the product within each sub- period；
Using feature vector of each product within each sub- period as input, by each product within the corresponding sub- period
History sales volume information obtains Method for Sales Forecast model as output, training.
3. kth recurrence is determined according to the method described in claim 2, wherein, Method for Sales Forecast model includes more regression trees
Training obtains plan tree as follows:
For each feature vector of each product in the feature vector in each sub- period, based on this feature vector the
The output valve of one regression tree to -1 regression tree of kth the corresponding history sales volume of output valve and this feature vector
Information determines this feature vector in the target value of kth regression tree, wherein k is the integer greater than 1；
Using feature vector of each product within each sub- period as input, each feature vector is determined in kth recurrence
The target value of plan tree obtains kth regression tree as output, training.
4. according to the method described in claim 3, wherein, the training step of Method for Sales Forecast model corresponding with the category information
Further include:
Obtain the feature vector of different product and history sales volume information under the affiliated category of the target product；
The feature vector of every kind of product is input to k regression tree, obtains the prediction sales volume information of every kind of product；
Based on history sales volume information and prediction sales volume information, the accuracy rate of the sales volume information of every kind of product is determined；
It is more than or equal to k-1 regression tree in response to the difference between the accuracy rate of the sales volume information of different product and determines difference
K-1 regression tree is determined as corresponding with the category information by the difference between the accuracy rate of the sales volume information of product
Method for Sales Forecast model.
5. a kind of for generating the device of information, comprising:
Request reception unit is configured to receive predictions request, and the predictions request includes the attribute information of target product, price
Relevant information and category information；
Vector generation unit is configured to attribute information and price related information according to the target product, generates the mesh
Mark the feature vector of product；
Information generating unit is configured to for described eigenvector being input to training in advance, corresponding with the category information
Method for Sales Forecast model generates the sales volume information of the target product, wherein Method for Sales Forecast model be used to characterize the feature of product to
Corresponding relationship between amount and the sales volume information of product.
6. device according to claim 5, wherein Method for Sales Forecast model corresponding with the category information by walking as follows
Rapid training obtains:
The sales volume data of at least one product within a preset period of time under the affiliated category of the target product are obtained, wherein described
Preset time period includes at least one sub- period；
For each product at least one described product, the product is extracted from the sales volume data of the product in each period of the day from 11 p.m. to 1 a.m
Between attribute information, price related information and history sales volume information in section, and it is raw based on attribute information and price related information
At feature vector of the product within each sub- period；
Using feature vector of each product within each sub- period as input, by each product within the corresponding sub- period
History sales volume information obtains Method for Sales Forecast model as output, training.
7. device according to claim 6, wherein Method for Sales Forecast model includes more regression trees, and kth recurrence is determined
Training obtains plan tree as follows:
For each feature vector of each product in the feature vector in each sub- period, based on this feature vector the
The output valve of one regression tree to -1 regression tree of kth the corresponding history sales volume of output valve and this feature vector
Information determines this feature vector in the target value of kth regression tree, wherein k is the integer greater than 1；
Using feature vector of each product within each sub- period as input, each feature vector is determined in kth recurrence
The target value of plan tree obtains kth regression tree as output, training.
8. device according to claim 7, wherein the training step of Method for Sales Forecast model corresponding with the category information
Further include:
Obtain the feature vector of different product and history sales volume information under the affiliated category of the target product；
The feature vector of every kind of product is input to k regression tree, obtains the prediction sales volume information of every kind of product；
Based on history sales volume information and prediction sales volume information, the accuracy rate of the sales volume information of every kind of product is determined；
It is more than or equal to k-1 regression tree in response to the difference between the accuracy rate of the sales volume information of different product and determines difference
K-1 regression tree is determined as corresponding with the category information by the difference between the accuracy rate of the sales volume information of product
Method for Sales Forecast model.
9. a kind of device/server, comprising:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now method as described in any in claim 1-4.
10. a kind of computer readable storage medium, is stored thereon with computer program, wherein described program is executed by processor
Method of the Shi Shixian as described in any in claim 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810069211.9A CN110070382A (en) | 2018-01-24 | 2018-01-24 | Method and apparatus for generating information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810069211.9A CN110070382A (en) | 2018-01-24 | 2018-01-24 | Method and apparatus for generating information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110070382A true CN110070382A (en) | 2019-07-30 |
Family
ID=67365665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810069211.9A Pending CN110070382A (en) | 2018-01-24 | 2018-01-24 | Method and apparatus for generating information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110070382A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111127072A (en) * | 2019-11-14 | 2020-05-08 | 浙江大学 | Multi-stage real-time prediction method for new product requirements |
| CN113269574A (en) * | 2020-02-14 | 2021-08-17 | 北京沃东天骏信息技术有限公司 | Information distribution method, device, equipment and storage medium |
| CN113627846A (en) * | 2021-08-10 | 2021-11-09 | 北京沃东天骏信息技术有限公司 | Inventory adjusting method and device, electronic equipment and computer readable medium |
| CN113807758A (en) * | 2020-12-02 | 2021-12-17 | 北京京东振世信息技术有限公司 | Data generation method and device |
| CN113837492A (en) * | 2021-10-26 | 2021-12-24 | 北京沃东天骏信息技术有限公司 | Method, apparatus, storage medium, and program product for predicting supply amount of article |
| WO2022199383A1 (en) * | 2021-03-26 | 2022-09-29 | 北京京东拓先科技有限公司 | Article replenishment method and device, and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217355A (en) * | 2014-09-11 | 2014-12-17 | 北京京东尚科信息技术有限公司 | Method and device for predicting sales volume of promotion items |
| CN106408341A (en) * | 2016-09-21 | 2017-02-15 | 北京小米移动软件有限公司 | Goods sales volume prediction method and device, and electronic equipment |
| CN106447397A (en) * | 2016-09-30 | 2017-02-22 | 浪潮软件集团有限公司 | Tobacco retail customer pricing method based on decision tree algorithm |
| CN107038190A (en) * | 2016-10-28 | 2017-08-11 | 厦门大学 | A kind of intelligent promotion plan modeling method applied to Taobao |
| US20170262900A1 (en) * | 2016-03-11 | 2017-09-14 | Wipro Limited | System and method for generating promotion data |
-
2018
- 2018-01-24 CN CN201810069211.9A patent/CN110070382A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217355A (en) * | 2014-09-11 | 2014-12-17 | 北京京东尚科信息技术有限公司 | Method and device for predicting sales volume of promotion items |
| US20170262900A1 (en) * | 2016-03-11 | 2017-09-14 | Wipro Limited | System and method for generating promotion data |
| CN106408341A (en) * | 2016-09-21 | 2017-02-15 | 北京小米移动软件有限公司 | Goods sales volume prediction method and device, and electronic equipment |
| CN106447397A (en) * | 2016-09-30 | 2017-02-22 | 浪潮软件集团有限公司 | Tobacco retail customer pricing method based on decision tree algorithm |
| CN107038190A (en) * | 2016-10-28 | 2017-08-11 | 厦门大学 | A kind of intelligent promotion plan modeling method applied to Taobao |
Non-Patent Citations (2)
| Title |
|---|
| 王建伟;: "基于商品聚类的电商销量预测", 计算机系统应用, no. 10 * |
| 费斐 等: "决策树算法在团购商品销售预测中的应用", 《计算机系统应用》, vol. 22, no. 2, pages 1 - 2 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111127072A (en) * | 2019-11-14 | 2020-05-08 | 浙江大学 | Multi-stage real-time prediction method for new product requirements |
| CN113269574A (en) * | 2020-02-14 | 2021-08-17 | 北京沃东天骏信息技术有限公司 | Information distribution method, device, equipment and storage medium |
| CN113807758A (en) * | 2020-12-02 | 2021-12-17 | 北京京东振世信息技术有限公司 | Data generation method and device |
| CN113807758B (en) * | 2020-12-02 | 2023-11-03 | 北京京东振世信息技术有限公司 | Data generation method and device |
| WO2022199383A1 (en) * | 2021-03-26 | 2022-09-29 | 北京京东拓先科技有限公司 | Article replenishment method and device, and storage medium |
| CN113627846A (en) * | 2021-08-10 | 2021-11-09 | 北京沃东天骏信息技术有限公司 | Inventory adjusting method and device, electronic equipment and computer readable medium |
| CN113837492A (en) * | 2021-10-26 | 2021-12-24 | 北京沃东天骏信息技术有限公司 | Method, apparatus, storage medium, and program product for predicting supply amount of article |
| CN113837492B (en) * | 2021-10-26 | 2024-05-21 | 北京沃东天骏信息技术有限公司 | Method, apparatus, storage medium, and program product for predicting article supply amount |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110070382A (en) | Method and apparatus for generating information | |
| CN109064278B (en) | Target object recommendation method and device, electronic equipment and storage medium | |
| CN111652657A (en) | Commodity sales prediction method and device, electronic equipment and readable storage medium | |
| CN109426885A (en) | Order allocation method and device | |
| CN110135901A (en) | A kind of enterprise customer draws a portrait construction method, system, medium and electronic equipment | |
| CN110400103A (en) | Replenishment quantity determines method, apparatus, computer installation and storage medium | |
| US11620593B2 (en) | Systems and methods for industry chain graph generation | |
| CN109214730A (en) | Information-pushing method and device | |
| CN108805594A (en) | Information-pushing method and device | |
| CN109961198B (en) | Associated information generation method and device | |
| CN109684624A (en) | A kind of method and apparatus in automatic identification Order Address road area | |
| CN109634833A (en) | A kind of Software Defects Predict Methods and device | |
| CN109544076A (en) | Method and apparatus for generating information | |
| CN110020143A (en) | A kind of landing page generation method and device | |
| CN108388563A (en) | Information output method and device | |
| CN110033337A (en) | The method and apparatus of order production | |
| CN112215448A (en) | Method and device for distributing customer service | |
| CN111292149A (en) | Method and device for generating return processing information | |
| CN110019367A (en) | A kind of method and apparatus of statistical data feature | |
| CN109993544A (en) | Data processing method, system, computer system and computer readable storage medium | |
| CN110348921A (en) | The method and apparatus that shops's article is chosen | |
| CN113204712A (en) | Information pushing method, device, medium and program product based on community service | |
| CN109711917A (en) | Information-pushing method and device | |
| CN109948773A (en) | The method and apparatus for generating information | |
| CN112348590A (en) | Method and device for determining value of article, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |