CN108229804A - Brand promotion project management method, device, terminal device and storage medium - Google Patents
Brand promotion project management method, device, terminal device and storage medium Download PDFInfo
- Publication number
- CN108229804A CN108229804A CN201711429863.0A CN201711429863A CN108229804A CN 108229804 A CN108229804 A CN 108229804A CN 201711429863 A CN201711429863 A CN 201711429863A CN 108229804 A CN108229804 A CN 108229804A
- Authority
- CN
- China
- Prior art keywords
- project
- kpi
- current
- information
- brand promotion
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of brand promotion project management method, device, terminal device and storage mediums.The brand promotion project management method obtains project acceptance inspection information, and the project acceptance inspection information includes item id, project channel and at least two current project indexs;Based on the project channel, corresponding KPI assessment models are obtained;Based on the KPI assessment models and at least two current project indexs, current KPI value is obtained;Based on the item id, the target KPI value of corresponding brand promotion project is obtained;Based on the current KPI value and the target KPI value, the project acceptance inspection result of the brand promotion project is obtained.The brand promotion project management method can intuitively, effectively evaluate the completion effect of brand promotion project.
Description
Technical field
The present invention relates to project management field more particularly to a kind of brand promotion project management method, device, terminal devices
And storage medium.
Background technology
Project management generally comprises project initiation, approves and initiate a project, the stages such as the project implementation and the project acceptance inspection, the project management
Including but not limited to brand promotion project management.In brand promotion project management, project initiation refers to formulate brand promotion item
The processes such as purpose decision or plan approve and initiate a project and refer to according to the decision or plan formulated work out the work of brand promotion project
Make the processes such as target or plan, the project implementation refers to carry out performing brand promotion project according to the target or plan worked out
Process, the project acceptance inspection refer to the process of carry out an acceptance inspection to the executive condition and effect of brand promotion project.In brand promotion project
The project acceptance inspection stage, it is corresponding since the project channel (i.e. the circulation way of brand promotion) of brand promotion project is different
The project indicator (such as environmental disturbances degree, spectators' concentration degree) differs, moreover, collected multiple projects in same project channel
Index influences each other, and can not determine to complete effect quality for evaluating brand promotion project, so as to can not intuitively evaluate brand
Whether publicity project reaches expectation when approving and initiate a project.
Invention content
The embodiment of the present invention provides a kind of brand promotion project management method, to solve current brand promotion project management
Cheng Wufa visual evaluation brand promotions project completes the problem of effect is fine or not.
In a first aspect, the embodiment of the present invention provides a kind of brand promotion project management method, including:
Project acceptance inspection information is obtained, the project acceptance inspection information includes item id, project channel and at least two currentitems
Mesh index;
Based on the project channel, corresponding KPI assessment models are obtained;
Based on the KPI assessment models and at least two current project indexs, current KPI value is obtained;
Based on the item id, the target KPI value of corresponding brand promotion project is obtained;
Based on the current KPI value and the target KPI value, the project acceptance inspection result of the brand promotion project is obtained.
Second aspect, the embodiment of the present invention provide a kind of brand promotion project management device, including:
Project acceptance inspection data obtaining module, for obtaining project acceptance inspection information, the project acceptance inspection information include item id,
Project channel and at least two current project indexs;
KPI assessment models acquisition modules for being based on the project channel, obtain corresponding KPI assessment models;
Current KPI value acquisition module, for being based on the KPI assessment models and at least two current project indexs,
Obtain current KPI value;
Target KPI value acquisition module for being based on the item id, obtains the target KPI of corresponding brand promotion project
Value;
Project acceptance inspection result acquisition module for being based on the current KPI value and the target KPI value, obtains the product
Board publicizes the project acceptance inspection result of project.
The third aspect, the embodiment of the present invention provide a kind of terminal device, including memory, processor and are stored in described
In memory and the computer program that can run on the processor, the processor are realized when performing the computer program
The step of brand promotion project management method.
Fourth aspect, the embodiment of the present invention provide a kind of computer readable storage medium, the computer-readable storage medium
Matter is stored with computer program, and the computer program realizes the brand promotion project management method when being executed by processor
Step.
It is first in brand promotion project management method, device, terminal device and storage medium that the embodiment of the present invention is provided
Project acceptance inspection information is first obtained, project acceptance inspection information includes item id, project channel and at least two current project indexs, passes through
Project acceptance inspection information is obtained, effectively provides the data basis for carrying out brand promotion project management.Then it based on project channel, obtains
Corresponding KPI assessment models are taken, by project channel, corresponding KPI assessment models is obtained according to the project channel, are worked as to obtain
Preceding KPI provides suitable assessment models.KPI assessment models and at least two current project indexs are then based on, are obtained current
KPI value by using KPI assessment models and current project index, can get current KPI value, to obtain brand promotion
The project acceptance inspection result of project provides important data basis.Then based on item id, corresponding brand promotion project is obtained
Target KPI value, by obtaining the target KPI value of corresponding brand promotion project, be conducive to realize to working as by target KPI value
Whether preceding KPI value inspection brand promotion project meets the effect of desired design.Current KPI value and target KPI value are finally based on, is obtained
The project acceptance inspection of brand promotion project is taken as a result, by the comparison to current KPI value and target KPI value, to brand promotion project
Performance is assessed, to determine whether the brand promotion project reaches desired value, you can with using the current KPI calculated
Value intuitively evaluates the quality that brand promotion project completes effect.
Description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is a flow chart of brand promotion project management method in the embodiment of the present invention 1.
Fig. 2 is the particular flow sheet before step S10 in Fig. 1.
Fig. 3 is a particular flow sheet of step S11 in Fig. 2.
Fig. 4 is the particular flow sheet before step S20 in Fig. 1.
Fig. 5 is a particular flow sheet of step S22 in Fig. 4.
Fig. 6 is another flow chart of brand promotion project management method in the embodiment of the present invention 1.
Fig. 7 is a functional block diagram of brand promotion project management device in the embodiment of the present invention 2.
Fig. 8 is a schematic diagram of terminal device in the embodiment of the present invention 4.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is part of the embodiment of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, the every other implementation that those of ordinary skill in the art are obtained without creative efforts
Example, shall fall within the protection scope of the present invention.
Embodiment 1
Fig. 1 shows a flow chart of brand promotion project management method in the present embodiment.The brand promotion project management side
Method can be applicable in the terminal device of the financial institutions such as bank, insurance and security or the configuration of other mechanisms, for carrying out brand a surname
Biography project is managed, and specifically can be applicable in the brand promotion project management system of installation on the terminal device.The brand is declared
Project management system is passed to refer to for the system to the progress project management of brand promotion project.Wherein, the terminal device be can be with
The equipment that user carries out human-computer interaction, the including but not limited to equipment such as computer, smart mobile phone and tablet.As shown in Figure 1, the brand
Publicity project management method includes the following steps:
S10:Project acceptance inspection information is obtained, project acceptance inspection information includes item id, project channel and at least two currentitems
Mesh index.
Wherein, project acceptance inspection information refers to information included in project acceptance inspection process.Item id refers to uniquely know
Other brand promotion item purpose mark.Project channel refers to that brand promotion project carries out the mode of publicity propagation.Wherein, project channel
Can include outdoor advertising (as outdoor big shot, advertisement of whitewashing a wall, LED advertisements, community lamp box, subway, bus, building, airport,
High ferro and cinema etc.), the web advertisement (such as video ads, search advertisements, brand picture-text advertisement and information flow advertisement), plane
The channels such as advertisement (such as newspaper advertisement, magazine advertising) and television advertising (such as television advertising piece, product placement).Current project
Index refers to the project indicator for weighing the corresponding brand promotion project quality of item id.The project indicator include specific dimension and
Corresponding numerical value, and current project index includes the actual value determined when specific dimension and the project acceptance inspection.The project indicator specifically may be used
To include propagating range (flow of the people, light exposure, circulation, covering person-time), propagate depth (geographical location, location advertising, advertisement
Duration), the specific dimension such as cost and recruitment evaluation.Specifically, the project indicator should be associated with project channel, for example, working as project
When channel is outdoor advertising, it can be flow of the people and displaying duration etc. to propagate the specific targets in range dimension.When project canal
When road is the web advertisement, propagating the specific targets in range dimension can be light exposure and launch frequency etc. rather than project
Channel is the indexs such as the flow of the people of outdoor advertising and displaying duration.It is to be appreciated that each project indicator should be according to specific
Depending on project channel, and may not be able to be general on other project channels.
In the present embodiment, in the project acceptance inspection stage, terminal device obtains project acceptance inspection information, which includes
Item id, project channel and at least two current project indexs can be follow-up acquisition KPI by obtaining project acceptance inspection information
Assessment models and current KPI value provide data and support, base is provided for the final project acceptance inspection result for obtaining brand promotion project
The guarantee of plinth information.
In a specific embodiment, as shown in Fig. 2, before project acceptance inspection information (i.e. step S10) is obtained, it is specific to go back
Include the following steps:
S11:Acquisition is approved and initiate a project information, and it is pre- that information of approving and initiate a project includes item id, item types, project channel and project
Calculation funds.
Wherein, information of approving and initiate a project refers to information included in the process of approving and initiate a project.Item id refers to uniquely know
Other item purpose mark.Item types refer to the specific rating-type of the project, as item types be belong to it is large, medium and small or other
The item types of grade.Project channel refers to that brand promotion project carries out the mode of publicity propagation.Project budget funds refer to this
Project pre-set funds value in project verification, it is total for pre-estimating the funds needed in entire brand promotion Project Process
Volume.In addition, information of approving and initiate a project is corresponding with project acceptance inspection information, that is, information of approving and initiate a project should also refer to including at least two projects
Mark, the project indicator is as the specific dimension where current project index, but the numerical value of the project indicator is the stage of approving and initiate a project
Preset value rather than the actual value in project acceptance inspection stage.
S12:Based on item types and project channel, project funds range and target KPI computational charts are obtained.
Wherein, project funds range refers to the funds determined during approving and initiate a project according to item types and project channel
Range is for judging whether project budget funds meet the judge index of the project verification requirement of respective item type.Target KPI is counted
Calculation table refers to be used for the table for calculating target KPI value during project verification.Target KPI value refers to that brand promotion is approved and initiate a project the stage
It determines its institute's KPI value to be achieved, is for the desired value of brand evaluation publicity project performance quality.Wherein, KPI (Key
Performance Indicator) refer to KPI Key Performance Indicator, it is the evaluation index that assessment item completes effect quality.
In the present embodiment, if the grade of the item types for large size, project channel be outdoor advertising, then terminal device according to
The item types of acquisition and project channel search corresponding project funds range and target KPI computational charts, i.e., in terminal
The grade that equipment query obtains item types is large-scale corresponding project funds range, and inquire acquisition project channel as open air
The target KPI computational charts of advertisement.Specifically, project funds range and KPI computational charts are stored in database, and and item class
Type and project channel are associated, and can inquire by query sentence of database and obtain corresponding project funds range and target KPI meters
Table is calculated, process is simple and convenient.
S13:If project budget funds are in the range of project funds, by the application for information of approving and initiate a project, and based on mesh
KPI computational charts and project budget funds are marked, obtain target KPI value corresponding with item id.
In the present embodiment, the application for information of approving and initiate a project is needed by audit, to judge whether to meet information of approving and initiate a project
Application condition after could pass through.Specifically, project budget funds are made comparisons with project funds range, if the project budget passes through
Take in the range of project funds, then pass through the application for information of approving and initiate a project;It is calculated according to target KPI corresponding with project channel
Table and project budget funds obtain target KPI value corresponding with item id.Target KPI value can be by self-defined Set scale
It calculates or other KPI calculations obtains, suitable calculation should be selected in KPI computational charts as the case may be.
S14:If project budget funds not in the range of project funds, obtain project verification application not by prompt message.
Specifically, the item types in project verification application information and project channel are identified project after funds range, if than
Assert that project budget funds not in the range of the project funds, then assert that the project verification application information does not meet application slip compared with judgement
Part, shown on the display interface of terminal device project verification application not by prompt message, met so that user is prompted to re-enter
The application for application information of setting up the project.
In a specific embodiment, in step S11, information of approving and initiate a project further includes project fund information table.Wherein, item
Mesh funds application information refers to be formed the specifying information of project budget funds during approving and initiate a project.As shown in figure 3, the project of acquisition
Project verification information specifically comprises the following steps:
S111:Acquisition project funds application information, project funds application information include current cost subject and specific expense
Value.
Specifically, project funds application information is it is emphasised that the current tools such as cost subject and specific cost value in the process of setting up the project
Body information.Current cost subject refers to the specific class categories of process funds expense of approving and initiate a project, and such as current cost subject can be with
It is the subjects such as advertising expense, provider service expense and publicity articles for use expense.Specific cost value refers to the current cost subject specifically in item
Mesh project initiation phase is expected to the expense spent.
S112:If current cost subject is Standard Rate subject, determined based on current cost subject and specific cost value
Project fund information table.
Wherein, Standard Rate subject refers to by cost subject seek unity of talking to manage and the cost subject of setting.This reality
Apply in example, when enterprise possess it is multiple management subsystem when, often will appear it is each management subsystem between bore do not have alignment feelings
Condition, such as buying, finance, the inconsistent situation of business bore, will lead to not support to KPI Key Performance Indicator analysis design with
Follow-up optimization operation, and Standard Rate subject is exactly to be set to solve such case.It such as can be by brand promotion project
Various expenses be all set to advertising expense, provider service expense and publicity articles for use and take the Standard Rate subject of this three categories, often
The Standard Rate section of one major class is further subdivided into the Standard Rate subject of multiple groups now, during approving and initiate a project, makes current
Cost subject is corresponding with the Standard Rate subject of group.In the present embodiment, if current cost subject is Standard Rate subject,
Subsequent KPI Key Performance Indicator analysis design and optimization operation is not interfered with, then is based on current cost subject and specific cost value
It can determine project fund information table.
S113:If current cost subject is not Standard Rate subject, taken using preset term subject conversion table to current
Conversion process is carried out with subject, obtains corresponding Standard Rate subject, item is determined based on Standard Rate subject and specific cost value
Mesh fund information table.
Wherein, preset term subject translation table refers to pre-set be used to non-standard cost subject being converted into standard expense
With section's purpose translation table, i.e. the preset term subject translation table recites current brand a surname in current management subsystem or mechanism
The cost subject that biography project uses and the Standard Rate section purpose transformational relation, so that the brand promotion of all brand promotion projects
Expense bore is consistent.Media advertisement, media such as non-standard cost subject obtain visitor and brand cooperation and are turned by preset term subject
Change the transformational relation in table, media advertisement, media are obtained into visitor and brand cooperation is converted in corresponding Standard Rate subject
Advertising expense.
In the present embodiment, if current cost subject is not Standard Rate subject, need to obtain storage in the database pre-
If project subject translation table, and conversion process is carried out to current cost subject using preset term subject conversion table, it obtains and corresponds to
Standard Rate subject, then can be identified project fund information table based on Standard Rate subject and specific cost value.
S20:Based on project channel, corresponding KPI assessment models are obtained.
Wherein, KPI assessment models refer to comment with the relevant calculating KPI value of specific targets for being obtained according to specific targets
Estimate model.In the present embodiment, terminal device can obtain according to the project channel and assess mould with the corresponding KPI of the project channel
Type, detailed process are to search and call the corresponding model of the KPI assessment models in the database for being stored in and being connected with terminal device
File, wherein, which refers to that KPI assessment models are stored in the form of model file in the database.Different project canals
The corresponding KPI assessment models in road are different, such as when project channel is outdoor advertising, then the KPI assessment models obtained are
For calculating the KPI value that the project channel is outdoor advertising;When project channel is the web advertisement, then the KPI obtained assesses mould
Type is for calculating the KPI value that the project channel is the web advertisement.Corresponding KPI is obtained by a certain specific project channel to comment
Estimate model, be capable of providing on the project channel and calculate KPI value accuracy and all very high assessment models of the property of can refer to, be follow-up
KPI value calculate, obtain check and acceptance result, both provide the reliable and accurate KPI assessment models of result of calculation.
In a specific embodiment, as shown in figure 4, based on project channel, the step of corresponding KPI assessment models is obtained
Suddenly before (i.e. step S20), brand promotion project management method further includes training KPI assessment models in advance, and training KPI is commented in advance
Estimate model to specifically comprise the following steps:
S21:The original item information of same project channel is obtained, original item information includes original KPI value and at least two
A original item index.
Wherein, original item information refers to create the project information of project, it can be understood as is the item in history item
Mesh information, here using original item information as training data to train KPI assessment models.Original KPI value and at least two originals
The beginning project indicator contributes to train the training data needed for KPI assessment models, i.e., original KPI value and at least two original items
Index is all the data that the project information in history item is included, and has very big side for the accurate KPI assessment models of training
It helps.The specific targets content phase one of the specific targets content of at least two current project indexs and at least two original item indexs
It causes, as current project index is flow of the people, shows duration, geographical location, area, visibility, environmental disturbances degree and stay time
Etc. indexs, then original item index should include the corresponding specific targets content of above-mentioned current project index.
In the present embodiment, by obtaining the original item information of same project channel, included based on the original item information
Original KPI value and at least two original item indexs, the follow-up data basis for carrying out KPI assessment models training is provided.
S22:Based on original KPI value and at least two original item indexs, obtained using convolutional neural networks model training
KPI assessment models.
Wherein, convolutional neural networks (Convolutional Neural Network, abbreviation CNN) model, before being one kind
Present neural network.Convolutional neural networks generally include at least two non-linear trainable convolutional layers, and at least two is nonlinear
Pond layer and at least one full articulamentum, i.e., including at least five hidden layers, additionally including input layer and output layer.
In the present embodiment, by using convolutional neural networks model to original KPI value and at least two original item indexs
It is trained, obtains the weights of updated each layer in convolutional neural networks model, the weights based on updated each layer obtain
KPI assessment models, the KPI assessment models obtained by convolutional neural networks model training.It is obtained by the KPI assessment models
The accuracy of KPI value and the property of can refer to are very high.
In a specific embodiment, as shown in figure 5, in step S22, based on original KPI value and at least two primitive terms
Mesh index obtains KPI assessment models using convolutional neural networks model training and specifically comprises the following steps:
S221:Initialize convolutional neural networks model.
In the present embodiment, initialization convolutional neural networks refer to initialize convolution kernel (i.e. weights) and the biasing of convolutional layer,
Assign weights and biasing initial value in convolutional neural networks model.If initial weight be in error surface one is relatively flat
During slow region, the convergence rate of convolutional neural networks model training may extremely slowly.Under normal circumstances, the weights of network
It is initialized in a relatively small section with 0 mean value and is uniformly distributed, such as section as [- 0.10 ,+0.10]
It is interior.
S222:Input is by original KPI value and at least two original item Index Establishments in convolutional neural networks model
Characteristic pattern.
In the present embodiment, input is by original KPI value and at least two original item indexs in convolutional neural networks model
The characteristic pattern of foundation.Wherein, original item index includes multiple dimensions, as original item of the project channel for outdoor advertising refers to
Mark can include propagating range, propagate the dimensions such as depth, cost and recruitment evaluation.It can include multiple indexs in per dimension,
Flow of the people, displaying duration original item index can be included by such as propagating range dimension.By by the original item of different dimensions
The operation that index is normalized in advance so that different dimensions, the data in source are unified under a referential, and ability is right in this way
Original item index under different dimensions is compared and treated, according to the original item index of different dimensions and the project channel
Original KPI value, establish corresponding characteristic pattern (this feature figure essence embody be relationship between different dimensions data), that is, exist
Original item index and corresponding original KPI value in hyperspace based on different dimensions form original image, then this is original
Image inputs convolutional neural networks model to calculate the output of each layer.
S223:Feature based figure calculates the output of each layer of convolutional neural networks model.
The output of each layer of convolutional neural networks model is calculated according to this feature figure, the output of each layer uses propagated forward algorithm
It can obtain.Wherein, different from the neural network model generally connected entirely, locally-attached convolutional neural networks model is also needed
The characteristic pattern of each output of the characteristic pattern and pond layer of each output of convolutional layer in computation model, to be carried out to weights
Update.Specifically, the characteristic pattern x of each output of convolutional layerjForWherein, l is current layer,
Mj represents the input feature vector figure combination of selection,It is the output of i.e. l-1 layers of i-th kind of characteristic pattern of input,It is l layers of input
Convolution kernel used is connected between i-th kind of characteristic pattern and the jth kind characteristic pattern of output,It is that l layers of jth kind characteristic pattern is corresponding to add
Property biasing, f is activation primitive, which can be sigmoid activation primitives.The characteristic pattern of each output of pond layer
xjForWherein, down represents that down-sampling calculates, hereL layers of correspondence of jth kind characteristic pattern
The biasing of multiplying property, b is that the corresponding additivity of l layer of jth kind characteristic pattern biases.The present embodiment is mainly provided in convolutional neural networks model
The convolutional layer of the general neural network model connected entirely and the output of pond layer are different from, the output of remaining each layer with generally connecting entirely
Neural network model calculate identical, can be obtained using propagated forward algorithm, therefore the citing that differs, in order to avoid repeat.
S224:Error-duration model is carried out to each layer of convolutional neural networks model according to the output of each layer of convolutional neural networks model
Update obtains the weights of updated each layer of convolutional neural networks model.
In step S223, error is certainly existed between the output by each layer of convolutional neural networks model, is needed this
Control information successively returns to each layer, allows each layer of weights for updating them, could obtain the better face of recognition effect and know
Other model.In the present embodiment, error-duration model update is carried out to each layer of convolutional neural networks model according to the output of each layer, is obtained more
The weights of each layer after new specifically include the control information for calculating each layer of convolutional neural networks model, and use gradient descent method
Update each layer of weights.Wherein, gradient descent method update weights mainly utilize error cost function to the gradient of parameter, institute
Target with right value update is exactly that each layer is allowed to obtain such gradient, is then updated.
In a specific embodiment, step S224 specifically comprises the following steps:According to the table of n-th of error cost function
Up to formulaN-th of error cost function is obtained, wherein, E is error cost function, and n is individualized training sample
This, the output in convolutional neural networks model isReality output isC is the dimension of reality output.It is single in order to ask for
For the error cost function of sample to the local derviation of parameter, it is error to the change rate of output, the table of sensitivity to define sensitivity δ here
It is up to formulaWherein E is error cost function, and wherein u is ul=Wlxl-1+bl, current l layers of l expressions, WlRepresent the layer
Weights, xl-1Represent the input of this layer, blRepresent the additivity biasing of this layer.Return error letter layer by layer by meter sensitivity
Backpropagation can be realized in breath, and the process of wherein backpropagation refers to miss each layer of convolutional neural networks model
Poor anti-pass update obtains the process of the weights of updated each layer.The then sensitivity of l layers of convolutional layer isWherein,Each element multiplication is represented, because of each neuron connection
Will be there are one sensitivity δ, so each layer of sensitivity is a matrix, l+1 layers refer to pond layer, the essential phase of operation
When in being also to do convolution algorithm.Such as do the down-sampling that characteristic pattern size is 2 and operate, be exactly with each value of 2*2 for 1/4 volume
Product nuclear convolution image, so weights W here is actually the convolution kernel of this 2*2, its value is βj.Up is adopted on representing
Sample calculates, and it is to calculate opposite calculating with down-sampling that up-sampling, which calculates, and when doing down-sampling calculating, decimation factor is n, then on adopt
Sample, which calculates, each pixel to replicate n times on vertical and horizontal direction respectively.Since the sensitivity matrix of l+1 ponds layer is l
The 1/4 of the size of layer sensitivity matrix calculates so up-sampling need to be done to l+1 layers of sensitivity matrix, makes their sizes consistent.
According to the sensitivity of acquisition, calculate error cost function is to the local derviation of additivity biasing bI.e. in layer l
All node summations in sensitivity, wherein (u, v) represents the element position in sensitivity matrix.Multiplying property biases β and is passed with forward direction
The pond layer for broadcasting middle current layer is related, therefore first definesError cost function is then calculated to multiplying property
Biasing β local derviation beLocal derviation of the error cost function to convolution kernel k is calculated laterHereIt isWhen doing convolution, with kijIt does small in each characteristic pattern of convolution
Block, (u, v) refer to fritter center, export the value of (u, v) position in characteristic pattern, be by input feature vector figure (u, v) position it is small
Block and convolution kernel kijValue obtained by convolution.According to the operation of above formula, updated convolutional neural networks model can be obtained
The weights of convolutional layer.In the training process of convolutional neural networks model, also cope with pond layer and be updated, for pond layer
The characteristic pattern x of each outputjForWherein, down represents down-sampling, and β here is
Multiplying property biases, and b is additivity biasing.The calculation formula of pond layer sensitivity is in convolutional neural networks modelAnd error cost can be acquired according to δ
Function pair additivity biasing b local derviation beWherein conv2, rot180 and full are to calculate required function,
Remaining parameter of above formula is identical with the meaning of parameters that above-mentioned convolutional layer formula refers to, this will not be detailed here.According to above-mentioned public affairs
Formula can obtain updated pond layer weights;Additionally cope with other each layers (such as full articulamentum) of convolutional neural networks model
Between weights be updated, the renewal process is identical with the weight renewing method of general full Connection Neural Network model, after
Weights are updated to propagation algorithm, to avoid burden, then this is not described in detail one by one.By to each layer of convolutional neural networks model into
Row error-duration model updates, and obtains the weights of updated each layer.
S225:Based on the weights of updated each layer of convolutional neural networks model, KPI assessment models are obtained.
In the present embodiment, by the weights of updated each layer of acquisition, being applied in convolutional neural networks model can obtain
Take the KPI assessment models after training.Further, the weights between each layer of KPI assessment models reflect same project channel
The potential relationship between the project indicator between lower different dimensions, can reflect the project indicator under practical each dimension for whole
The influence of a KPI assessments, the influence can be presented as the weights of disparity items index under each dimension.It is to be appreciated that pass through
Original KPI value corresponds to the original item index under project channel with the original KPI value, corresponding characteristic pattern is established, using convolution
Neural network model is trained this feature figure, and the weights reflection of each layer of convolutional neural networks model of acquisition is each original
The beginning project indicator accounts for the weights of original KPI value, and the high KPI assessment models of assessment accuracy, KPI assessments are obtained according to the weights
Model can help us to be assessed during the project acceptance inspection current project index.
S30:Based on KPI assessment models and at least two current project indexs, current KPI value is obtained.
In the present embodiment, according to get with the corresponding KPI assessment models of project channel, in KPI assessment models
Input at least two current project indexs, using the KPI assessment models at least two current project indexs of input at
Reason, obtains current KPI value.The current KPI value is one and carries out processing acquisition to current project index by KPI assessment models
As a result, by using the KPI assessment models of convolutional neural networks model training, it can preferably reflect that the project of different dimensions refers to
Potential relationship between mark so that can obtain the higher current KPI value of true rate based on the KPI assessment models.
S40:Based on item id, the target KPI value of corresponding brand promotion project is obtained.
In the present embodiment, target KPI value associated with the item id is found in the database according to item id, according to
The item id finds target KPI computational charts to obtain target KPI value, based on the target KPI value got, to examine brand
Whether publicity project meets the effect of desired design.Wherein, target KPI value is pre-set, the setting target KPI value mistake
Journey in step s 13 it is stated that, details are not described herein.
S50:Based on current KPI value and target KPI value, the project acceptance inspection result of brand promotion project is obtained.
In the present embodiment, based on current KPI value and target KPI value, size comparison is carried out to the two;If current KPI value is small
In target KPI value, then the brand promotion project cannot reach the desired value set when approving and initiate a project;If current KPI value is more than mesh
KPI value is marked, then the brand promotion project can reach pre-set target KPI value during project verification, it is believed that the Branded terms publicize
Performance be meet it is desired.Current KPI value is obtained according to KPI assessment models, with reference to acquired in historical data
KPI assessment models compare closing to reality situation, the current KPI value accuracy of acquisition and the property of can refer to are all very high;Target KPI value
It is the brand promotion project " acceptance line " being pre-designed, i.e. the target KPI value is for examining whether current KPI value reaches expectation
Whether the judgment threshold of value, the performance for assessing the brand promotion project meet expectation.
In a specific embodiment, as shown in fig. 6, the brand promotion project management method further includes following steps:
S61:It obtains supplier information and recommends instruction, supplier information recommends instruction to include at least one classification indicators and row
Sequential mode.
Specifically, project acceptance inspection information further includes and the corresponding supplier information of item id.Supplier information recommendation refers to
Order is the instruction for carrying out supplier information recommendation.The supplier information recommends instruction to include at least one classification indicators and row
Sequential mode.Wherein, classification indicators refer to the criteria for classification for distinguishing supplier information, the classification indicators include item size,
Item amount and other classification indicators are 0-50 ten thousand, 51-100 ten thousand, ten thousand and more than 5,010,000 amount of money of 101-500 such as by item amount
Classify.Sortord refers to the aligning method for arranging supplier information, such as ascending order, by sequence aligning method.
In the present embodiment, terminal device obtains supplier information input by user and recommends instruction, which recommends
Instruction includes at least one classification indicators and sortord, for distinguishing and sorting to project, obtains the confession needed for user
Answer quotient's information recommendation information.
S62:According at least one classification indicators, corresponding supplier information table is obtained.
In the present embodiment, terminal device recommends the classification indicators that include in instruction according to the supplier information of acquisition, according to
Classification indicators find the supplier information table that storage corresponds to the classification indicators in the database.Wherein, supplier information table is
Previously according to the table that all classification indicators create in the database, user can directly acquire according to classification indicators and be stored in
Corresponding supplier information table in database.
S63:The current KPI value in supplier information table is ranked up according to sortord, supplier is obtained and recommends row
Table.
In the present embodiment, terminal device is recommended to instruct the sortord included according to the supplier information of acquisition, according to row
Sequential mode is ranked up the current KPI value in supplier information table, to obtain ranked supplier's recommendation list, so as to base
Suitable supplier is selected to cooperate in the supplier's recommendation list to have sorted, that improves follow-up brand promotion project reaches the phase
The probability of prestige.
In the brand promotion project management method that the present embodiment is provided, project acceptance inspection information, the project acceptance inspection are obtained first
Information includes item id, project channel and at least two current project indexs, by obtaining project acceptance inspection information, effectively provides
The data basis of brand promotion project management is carried out, help is provided for follow-up acquisition KPI assessment models and current KPI value.It connects
It based on project channel, obtains corresponding KPI assessment models, by project channel, corresponding KPI is obtained according to the project channel
Assessment models provide the suitable assessment models of project channel corresponding with the current KPI to obtain current KPI.It is then based on KPI
Assessment models and at least two current project indexs, obtain current KPI value, by using KPI assessment models and current project
Index can get current KPI value, and important data basis is provided for the follow-up project acceptance inspection result that obtains.Then it is based on
Item id obtains the target KPI value of corresponding brand promotion project, by the target KPI for obtaining corresponding brand promotion project
Value is conducive to realize the effect for brand promotion project being examined whether to meet desired design current KPI value by target KPI value.
Be finally based on current KPI value and target KPI value, obtain the project acceptance inspection of brand promotion project as a result, by current KPI value and
Brand promotion project performance is assessed in the comparison of target KPI value, to determine whether the brand promotion project reaches the phase
Prestige value, you can intuitively to evaluate the quality that brand promotion project completes effect using the current KPI value calculated.
It should be understood that the size of the serial number of each step is not meant to the priority of execution sequence, each process in above-described embodiment
Execution sequence should determine that the implementation process without coping with the embodiment of the present invention forms any limit with its function and internal logic
It is fixed.
Embodiment 2
Fig. 7 is shown and the one-to-one brand promotion project management device of brand promotion project management method in embodiment 1
Functional block diagram.As shown in fig. 7, the brand promotion project management device includes project acceptance inspection data obtaining module 10, KPI is assessed
Model acquisition module 20, current KPI value acquisition module 30, target KPI value acquisition module 40 and project acceptance inspection result acquisition module
50.Wherein, project acceptance inspection data obtaining module 10, KPI assessment models acquisition module 20, current KPI value acquisition module 30, target
The realization function of KPI value acquisition module 40 and project acceptance inspection result acquisition module 50 and brand promotion project management side in embodiment
The corresponding step of method corresponds, and is repeated to avoid, the present embodiment is not described in detail one by one.
Project acceptance inspection data obtaining module 10, for obtaining project acceptance inspection information, project acceptance inspection information includes item id, item
Mesh channel and at least two current project indexs.
KPI assessment models acquisition module 20 for being based on project channel, obtains corresponding KPI assessment models.
Current KPI value acquisition module 30, for being based on KPI assessment models and at least two current project indexs, obtains and works as
Preceding KPI value.
Target KPI value acquisition module 40 for being based on item id, obtains the target KPI value of corresponding brand promotion project.
Project acceptance inspection result acquisition module 50 for being based on current KPI value and target KPI value, obtains brand promotion project
Project acceptance inspection result.
Brand promotion project management device further includes KPI assessment models training module 60 and supplier's recommendation list obtains mould
Block 70.
KPI assessment models training module 60, for training KPI assessment models in advance.
Supplier's recommendation list acquisition module 70, for obtaining supplier's recommendation list.
Preferably, KPI assessment models training module 60, including original item information acquisition unit 61 and KPI assessment models
Acquiring unit 62.
Preferably, KPI assessment models acquiring unit 62 includes initialization subelement 621, characteristic pattern establishes subelement 622,
Layer output obtains subelement 623, right value update subelement 624 and KPI assessment models subelement 625.
Subelement 621 is initialized, for initializing convolutional neural networks model.
Characteristic pattern establishes subelement 622, for being inputted in convolutional neural networks model by original KPI value and at least two
The characteristic pattern of original item Index Establishment.
Layer output obtains subelement 623, and the output of each layer of convolutional neural networks model is calculated for feature based figure.
Right value update subelement 624, for according to the output of each layer of convolutional neural networks model to convolutional neural networks mould
Each layer of type carries out error-duration model update, obtains the weights of updated each layer of convolutional neural networks model.
KPI assessment models subelement 625 for the weights based on updated each layer of convolutional neural networks model, obtains
KPI assessment models.
Preferably, supplier's recommendation list acquisition module 70 includes instruction acquiring unit 71, supplier information table is recommended to obtain
Take unit 72 and supplier's recommendation list acquiring unit 73.
Recommend instruction acquiring unit 71, recommend instruction for obtaining supplier information, supplier information recommends instruction to include
At least one classification indicators and sortord.
Supplier information table acquiring unit 72, for according at least one classification indicators, obtaining corresponding supplier information
Table.
Supplier's recommendation list acquiring unit 73, for according to sortord to the current KPI value in supplier information table
It is ranked up, obtains supplier's recommendation list.
Brand promotion project management device further includes information acquisition unit 81 of approving and initiate a project, KPI computational charts acquiring unit 82
With target KPI value computing unit 83.
Preferably, information acquisition unit of approving and initiate a project 81 includes project funds application information acquiring unit 811 and project passes through
Charge information table determination unit 812.
Project funds application information obtains subelement 811, for obtaining project funds application information, project funds letter of application
Breath includes current cost subject and specific cost value.
Project fund information table determination subelement 812 if being Standard Rate subject for current cost subject, is based on working as
Preceding cost subject and specific cost value are identified project fund information table;
If current cost subject is not Standard Rate subject, using preset term subject conversion table to current cost subject
Conversion process is carried out, obtains corresponding Standard Rate subject, is identified project funds based on Standard Rate subject and specific cost value
Information table.
In the brand promotion project management device that the present embodiment is provided, project acceptance inspection data obtaining module 10, for obtaining
Project acceptance inspection information is taken, project acceptance inspection information includes item id, project channel and at least two current project indexs, passes through acquisition
Project acceptance inspection information effectively provides the data basis for carrying out brand promotion project management, for it is follow-up obtain KPI assessment models and
Current KPI value provides help.KPI assessment models acquisition module 20 for being based on project channel, obtains corresponding KPI assessments
By project channel, corresponding KPI assessment models are obtained according to the project channel, are provided to obtain current KPI with being somebody's turn to do for model
Current KPI corresponds to the suitable assessment models of project channel.Current KPI value acquisition module 30, for being based on KPI assessment models and extremely
Few two current project indexs, obtain current KPI value, by using KPI assessment models and current project index, can obtain
To current KPI value, important data basis is provided for the follow-up project acceptance inspection result that obtains.Target KPI value acquisition module 40 is used
In based on item id, the target KPI value of corresponding brand promotion project is obtained, is conducive to realize to current by target KPI value
Whether KPI value inspection brand promotion project meets the effect of desired design.Project acceptance inspection result acquisition module 50, for being based on working as
Preceding KPI value and target KPI value obtain the project acceptance inspection of brand promotion project as a result, by current KPI value and target KPI value
Comparison, brand promotion project performance is assessed, with determine the brand promotion project whether reach desired value, you can
Intuitively to evaluate the quality that brand promotion project completes effect using the current KPI value calculated.
Embodiment 3
The present embodiment provides a computer readable storage medium, computer journey is stored on the computer readable storage medium
Sequence, brand promotion project management method in realization embodiment 1, repeats to avoid when which is executed by processor, this
In repeat no more.Alternatively, it is realized when the computer program is executed by processor in embodiment 2 in brand promotion project management device
The function of each module/unit, repeats, which is not described herein again to avoid.
Embodiment 4
Fig. 8 is the schematic diagram of terminal device in the present embodiment.As shown in figure 8, terminal device 90 includes processor 91, storage
Device 92 and it is stored in the computer program 93 that can be run in memory 92 and on processor 91.Processor 91 performs computer
During program 93 realize embodiment 1 in brand promotion project management method each step, such as step S10, S20 shown in FIG. 1,
S30, S40 and S50.Alternatively, processor 91 realizes brand promotion project management device in embodiment 2 when performing computer program 93
The function of each module/unit, it is project acceptance inspection data obtaining module 10 as shown in Figure 7, KPI assessment models acquisition module 20, current
The function of KPI value acquisition module 30, target KPI value acquisition module 40 and project acceptance inspection result acquisition module 50.
Illustratively, computer program 93 can be divided into one or more module/units, one or more mould
Block/unit is stored in memory 92, and is performed by processor 91, to complete the present invention.One or more module/units can
To be the series of computation machine program instruction section that can complete specific function, the instruction segment is for describing computer program 93 at end
Implementation procedure in end equipment 90.For example, computer program 90 can be divided into the project acceptance inspection acquisition of information mould in embodiment 2
Block 10, KPI assessment models acquisition module 20, current KPI value acquisition module 30, target KPI value acquisition module 40 and the project acceptance inspection
As a result acquisition module 50, the concrete function of each module as described in Example 2, are repeated to avoid, are not repeated one by one herein.
Terminal device 90 can be the computing devices such as desktop PC, notebook, palm PC and cloud server.Eventually
End equipment may include, but be not limited only to, processor 91, memory 92.It will be understood by those skilled in the art that Fig. 8 is only eventually
The example of end equipment 90 does not form the restriction to terminal device 90, can include than illustrate more or fewer components or
Combine certain components or different components, for example, terminal device can also include input-output equipment, network access equipment,
Bus etc..
Alleged processor 91 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors, digital signal processor (Digital Signal Processor, DSP), application-specific integrated circuit
(Application Specific Integrated Circuit, ASIC), field programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor can also be any conventional processor
Deng.
Memory 92 can be the internal storage unit of terminal device 90, such as the hard disk or memory of terminal device 90.It deposits
Reservoir 92 can also be the plug-in type hard disk being equipped on the External memory equipment of terminal device 90, such as terminal device 90, intelligence
Storage card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card, flash card (Flash Card)
Deng.Further, memory 92 can also both include the internal storage unit of terminal device 90 or including External memory equipment.It deposits
Reservoir 92 is used to store computer program and other programs and data needed for terminal device.Memory 92 can be also used for temporarily
When store the data that has exported or will export.
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each work(
Can unit, module division progress for example, in practical application, can be as needed and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device are divided into different functional units or module, more than completion
The all or part of function of description.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it can also
That each unit is individually physically present, can also two or more units integrate in a unit.Above-mentioned integrated list
The form that hardware had both may be used in member is realized, can also be realized in the form of SFU software functional unit.
If the integrated module/unit realized in the form of SFU software functional unit and be independent product sale or
In use, it can be stored in a computer read/write memory medium.Based on such understanding, the present invention realizes above-mentioned implementation
All or part of flow in example method, can also instruct relevant hardware to complete, the meter by computer program
Calculation machine program can be stored in a computer readable storage medium, the computer program when being executed by processor, it can be achieved that on
The step of stating each embodiment of the method.Wherein, the computer program includes computer program code, the computer program generation
Code can be source code form, object identification code form, executable file or certain intermediate forms etc..The computer-readable medium
It can include:Any entity of the computer program code or device, recording medium, USB flash disk, mobile hard disk, magnetic can be carried
Dish, CD, computer storage, read-only memory (ROM, Read-Only Memory), random access memory (RAM,
Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..It is it should be noted that described
The content that computer-readable medium includes can carry out appropriate increasing according to legislation in jurisdiction and the requirement of patent practice
Subtract, such as in certain jurisdictions, according to legislation and patent practice, computer-readable medium do not include be electric carrier signal and
Telecommunication signal.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although with reference to aforementioned reality
Example is applied the present invention is described in detail, it will be understood by those of ordinary skill in the art that:It still can be to aforementioned each
Technical solution recorded in embodiment modifies or carries out equivalent replacement to which part technical characteristic;And these are changed
Or replace, the spirit and scope for various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of brand promotion project management method, which is characterized in that including:
Project acceptance inspection information is obtained, the project acceptance inspection information refers to including item id, project channel and at least two current projects
Mark;
Based on the project channel, corresponding KPI assessment models are obtained;
Based on the KPI assessment models and at least two current project indexs, current KPI value is obtained;
Based on the item id, the target KPI value of corresponding brand promotion project is obtained;
Based on the current KPI value and the target KPI value, the project acceptance inspection result of the brand promotion project is obtained.
2. brand promotion project management method according to claim 1, which is characterized in that be based on the project canal described
Road, before the step of obtaining corresponding KPI assessment models, the brand promotion project management method further includes:Training institute in advance
State KPI assessment models;
Described train in advance includes the step of the KPI assessment models:
The original item information of same project channel is obtained, the original item information includes original KPI value and at least two originals
The beginning project indicator;
Based on the original KPI value and at least two original item indexs, obtained using convolutional neural networks model training
The KPI assessment models.
3. brand promotion project management method according to claim 2, which is characterized in that described to be based on the original KPI
Value and at least two original item indexs obtain the KPI assessment models, packet using convolutional neural networks model training
It includes:
Initialize the convolutional neural networks model;
Input is by the original KPI value and at least two original item Index Establishments in the convolutional neural networks model
Characteristic pattern;
The output of each layer of convolutional neural networks model is calculated based on the characteristic pattern;Wherein, each output of convolutional layer
Characteristic pattern xjForWherein, l is current layer, and Mj represents the input feature vector figure combination of selection,
It is i-th kind of characteristic pattern of input,It is that connection is used between the l layers of i-th kind of characteristic pattern inputted and the jth kind characteristic pattern of output
Convolution kernel,It is the corresponding additivity biasing of l layers of jth kind characteristic pattern, f is activation primitive;The feature of each output of pond layer
Scheme xjForWherein, down represents that down-sampling calculates, hereL layers of jth kind characteristic pattern is right
The multiplying property biasing answered,It is the corresponding additivity biasing of l layers of jth kind characteristic pattern;
Error-duration model is carried out to each layer of the convolutional neural networks model according to the output of each layer of convolutional neural networks model
Update obtains the weights of updated each layer of convolutional neural networks model;
Based on the weights of updated each layer of convolutional neural networks model, KPI assessment models are obtained.
4. brand promotion project management method according to claim 1, which is characterized in that the project acceptance inspection information is also wrapped
It includes and the corresponding supplier information of the item id;
Described based on the KPI assessment models and at least two current project indexs, the step of obtaining current KPI value it
Afterwards, the brand promotion project management method further includes:
It obtains supplier information and recommends instruction, the supplier information recommends instruction to include at least one classification indicators and sequence side
Formula;
According at least one classification indicators, corresponding supplier information table is obtained;
The current KPI value in the supplier information table is ranked up according to the sortord, supplier is obtained and recommends row
Table.
5. brand promotion project management method according to claim 1, which is characterized in that the acquisition project acceptance inspection letter
Breath, further includes before:
Acquisition is approved and initiate a project information, and the information of approving and initiate a project includes item id, item types, project channel and project budget warp
Take;
Based on the item types and the project channel, project funds range and target KPI computational charts are obtained;
If the project budget funds are in the range of the project funds, by the application of the information of approving and initiate a project, and base
In the target KPI computational charts and the project budget funds, target KPI value corresponding with the item id is obtained.
6. brand promotion project management method according to claim 5, which is characterized in that the information of approving and initiate a project also is wrapped
Include project fund information table;
It is described to obtain information of approving and initiate a project, including:
Acquisition project funds application information, the project funds application information include current cost subject and specific cost value;
If the current cost subject is Standard Rate subject, true based on the current cost subject and the specific cost value
The fixed project fund information table;
If the current cost subject is not the Standard Rate subject, using preset term subject conversion table to described current
Cost subject carries out conversion process, takes the corresponding Standard Rate subject, based on the Standard Rate subject and described specific
Cost value determines the project fund information table.
7. a kind of brand promotion project management device, which is characterized in that including:
Project acceptance inspection data obtaining module, for obtaining project acceptance inspection information, the project acceptance inspection information includes item id, project
Channel and at least two current project indexs;
KPI assessment models acquisition modules for being based on the project channel, obtain corresponding KPI assessment models;
Current KPI value acquisition module for being based on the KPI assessment models and at least two current project indexs, obtains
Current KPI value;
Target KPI value acquisition module for being based on the item id, obtains the target KPI value of corresponding brand promotion project;
Project acceptance inspection result acquisition module for being based on the current KPI value and the target KPI value, obtains brand a surname
The project acceptance inspection result of biography project.
8. brand promotion project management device according to claim 7, which is characterized in that the brand promotion project management
Device further includes KPI assessment models training modules;The KPI assessment models training module includes:
Original item information acquisition unit, for obtaining the original item information of same project channel, the original item information
Including original KPI value and at least two original item indexs;
KPI assessment models acquiring units, for using convolutional neural networks model to described in the original KPI value and at least two
Original item index is trained, and obtains the KPI assessment models.
9. a kind of terminal device, including memory, processor and it is stored in the memory and can be on the processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1 to 6 when performing the computer program
The step of any one brand promotion project management method.
10. a kind of computer-readable medium, the computer-readable medium storage has computer program, which is characterized in that described
The step of the brand promotion project management method as described in any one of claim 1 to 6 is realized when computer program is executed by processor
Suddenly.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711429863.0A CN108229804A (en) | 2017-12-26 | 2017-12-26 | Brand promotion project management method, device, terminal device and storage medium |
PCT/CN2018/074661 WO2019127822A1 (en) | 2017-12-26 | 2018-01-31 | Brand promotion project management method and apparatus, terminal device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711429863.0A CN108229804A (en) | 2017-12-26 | 2017-12-26 | Brand promotion project management method, device, terminal device and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108229804A true CN108229804A (en) | 2018-06-29 |
Family
ID=62647959
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711429863.0A Pending CN108229804A (en) | 2017-12-26 | 2017-12-26 | Brand promotion project management method, device, terminal device and storage medium |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN108229804A (en) |
WO (1) | WO2019127822A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109002997A (en) * | 2018-07-26 | 2018-12-14 | 郑州云海信息技术有限公司 | Accounting method and system based on SV server product line working hour KPI under MES data |
CN110045798A (en) * | 2019-04-25 | 2019-07-23 | 烟台工程职业技术学院(烟台市技师学院) | A kind of brand promotion project management terminal device |
US11741364B2 (en) * | 2018-04-10 | 2023-08-29 | Hookit, Llc | Deep neural networks modeling |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102521706A (en) * | 2011-12-16 | 2012-06-27 | 北京斯泰威网络科技有限公司 | KPI data analysis method and device for the same |
CN104616107A (en) * | 2015-02-03 | 2015-05-13 | 北方信息控制集团有限公司 | Scientific research performance KPI management method and management system |
-
2017
- 2017-12-26 CN CN201711429863.0A patent/CN108229804A/en active Pending
-
2018
- 2018-01-31 WO PCT/CN2018/074661 patent/WO2019127822A1/en active Application Filing
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11741364B2 (en) * | 2018-04-10 | 2023-08-29 | Hookit, Llc | Deep neural networks modeling |
CN109002997A (en) * | 2018-07-26 | 2018-12-14 | 郑州云海信息技术有限公司 | Accounting method and system based on SV server product line working hour KPI under MES data |
CN109002997B (en) * | 2018-07-26 | 2021-11-26 | 郑州云海信息技术有限公司 | SV server product line working hour KPI accounting method and system based on MES data |
CN110045798A (en) * | 2019-04-25 | 2019-07-23 | 烟台工程职业技术学院(烟台市技师学院) | A kind of brand promotion project management terminal device |
Also Published As
Publication number | Publication date |
---|---|
WO2019127822A1 (en) | 2019-07-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20200250511A1 (en) | Artist comprehensive ability evaluation and cultivation assistant system based on artificial intelligence | |
Wang et al. | Predicting construction cost and schedule success using artificial neural networks ensemble and support vector machines classification models | |
CN106407999A (en) | Rule combined machine learning method and system | |
CN108734338A (en) | Credit risk forecast method and device based on LSTM models | |
Branco et al. | Analysing determinants of small and medium‐sized enterprise resilience using fuzzy cognitive mapping | |
CN109242672A (en) | Refund information forecasting method, device and the computer readable storage medium of loan | |
CN110751286A (en) | Training method and training system of neural network model | |
CN111861262A (en) | Enterprise perspective portrait method and terminal based on energy big data | |
Chen et al. | Performance risk assessment in public–private partnership projects based on adaptive fuzzy cognitive map | |
CN107909392A (en) | Stimulation behavior susceptibility Forecasting Methodology, device, equipment and readable storage medium storing program for executing | |
CN108229804A (en) | Brand promotion project management method, device, terminal device and storage medium | |
CN104933428A (en) | Human face recognition method and device based on tensor description | |
CN110264270A (en) | A kind of behavior prediction method, apparatus, equipment and storage medium | |
CN114663198A (en) | Product recommendation method, device and equipment based on user portrait and storage medium | |
TWM582163U (en) | Face recognition financial management planning system | |
CN114997263B (en) | Method, device, equipment and storage medium for analyzing training rate based on machine learning | |
CN114663167A (en) | Evaluation method and device based on shooting information, electronic equipment and storage medium | |
Chatterjee et al. | Calibrating the factors of management quality in banking performance: a mixed method approach | |
CN110033165A (en) | The recommended method of overdue loaning bill collection mode, device, medium, electronic equipment | |
CN109727125A (en) | Borrowing balance prediction technique, device, server, storage medium | |
CN115809837A (en) | Financial enterprise management method, equipment and medium based on digital simulation scene | |
Liu et al. | How to use stock data for data science education: a simulated trading platform in classroom | |
Mortaz Hejri et al. | Factors affecting the readiness of green business process management in the Iranian banking industry | |
CN114462742A (en) | Risk prompting method, device, equipment and computer storage medium | |
CN111179070A (en) | Loan risk timeliness prediction system and method based on LSTM |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180629 |
|
RJ01 | Rejection of invention patent application after publication |