CN110119935A - A kind of science and technology item declares method for procedure tracking and device - Google Patents
A kind of science and technology item declares method for procedure tracking and device Download PDFInfo
- Publication number
- CN110119935A CN110119935A CN201910212882.0A CN201910212882A CN110119935A CN 110119935 A CN110119935 A CN 110119935A CN 201910212882 A CN201910212882 A CN 201910212882A CN 110119935 A CN110119935 A CN 110119935A
- Authority
- CN
- China
- Prior art keywords
- science
- item
- attribute
- technology item
- technology
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Tourism & Hospitality (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 provides a kind of science and technology item and declares method for procedure tracking and device, the insertion matrix for obtaining science and technology item to be tracked first indicates, and the expression of insertionization matrix is input to science and technology item and is declared in flow tracking model, output tracking prediction result, then the attribute coefficients in tracking prediction result between the reference value of every kind of item characteristic attribute of the attribute deviation value and pre-stored science and technology item to be tracked of every kind of item characteristic attribute are calculated, and according to the attribute coefficients between the reference value of every kind of item characteristic attribute of the attribute deviation value and pre-stored science and technology item to be tracked of every kind of item characteristic attribute in the tracking prediction result being calculated, generate science and technology item to be tracked declares flow tracking result, declare the process progress that flow tracking result includes science and technology item to be tracked.It so, it is possible to provide guidance and support that scientific history science and technology project data declares situation, improve the success rate that science and technology item is declared, efficiency is declared in raising.
Description
Technical field
This application involves field of computer technology, in particular to a kind of science and technology item declare method for procedure tracking and
Device.
Background technique
There are numerous studies personnel that can declare country, save ground all types science and technology item every year at present, during science and technology is declared
All multipaths are often related to, the success rate of science and technology item application will be directly related to for the progress of each process.Mesh
Preceding scientific research personnel selection is essentially all the progress by subjective judgement science and technology item process during declaring, and is lacked
The history science and technology project data of science declares the guidance and support of situation, this just frequently can lead to many science and technology items and declares not
Success, and declare that time-consuming, declares inefficiency.
Summary of the invention
In view of this, a kind of science and technology item of being designed to provide of the embodiment of the present application declares method for procedure tracking and dress
It sets, to solve or improve the above problem.
According to the one aspect of the embodiment of the present application, a kind of electronic equipment is provided, may include that one or more storages are situated between
Matter and one or more processors communicated with storage medium.One or more storage mediums are stored with what processor can be performed
Machine readable instructions.When electronic equipment operation, by bus communication between processor and storage medium, processor executes institute
Machine readable instructions are stated, declare method for procedure tracking to execute science and technology item.
According to the another aspect of the embodiment of the present application, a kind of science and technology item is provided and declares method for procedure tracking, is applied to clothes
Business device, which comprises
The insertion matrix for obtaining science and technology item to be tracked indicates, include in insertionization matrix expression and this to
Track the associated project application flow data of science and technology item；
Insertionization matrix expression is input to the science and technology item to declare in flow tracking model, output tracking is pre-
It surveys as a result, the attribute that the tracking prediction result includes at least one item characteristic attribute of the science and technology item to be tracked is biased to
Value；
Calculate in the tracking prediction result attribute deviation value of every kind of item characteristic attribute and it is pre-stored should to
Attribute coefficients between the reference value of every kind of item characteristic attribute of track science and technology item；
According to the attribute deviation value of every kind of item characteristic attribute in the tracking prediction result being calculated with deposit in advance
Attribute coefficients between the reference value of every kind of item characteristic attribute of the science and technology item to be tracked of storage generate described to be tracked
Science and technology item declares flow tracking as a result, described declare the process that flow tracking result includes the science and technology item to be tracked
Progress；
Suggestion is declared according to the corresponding science and technology item of flow tracking result generation of declaring of the science and technology item to be tracked,
And it the science and technology item is declared into suggestion is sent to corresponding science and technology item and declare terminal to prompt associated user according to the section
Skill project application suggestion declares information to science and technology item and carries out supplement confirmation.
In a kind of possible embodiment, the insertion matrix for obtaining science and technology item to be tracked the step of it
Before, the method also includes:
It configures the science and technology item and declares flow tracking model, wherein the science and technology item declares flow tracking model packet
Include input layer, convolutional layer, pond layer, full articulamentum and output layer；
The configuration science and technology item declares the mode of flow tracking model, comprising:
Training sample is obtained, the training sample includes multiple scientific and technological items for marking and having characteristic attribute information
Mesh information；
The training sample is handled, obtaining insertion matrix to be entered indicates；
The insertionization matrix is indicated to be input to the convolutional layer by the input layer；Pass through the convolutional layer pair
The insertionization matrix indicates to carry out convolution algorithm, the corresponding feature extraction information of the pond Xiang Suoshu layer output；
The feature extraction information is converted to by the pond layer to be output to after corresponding multi-C vector and described is connected entirely
Connect layer, wherein the node of the full articulamentum is connect with the node of the pond layer；
The node of the full articulamentum random drop predetermined ratio calculates the feature extraction information, and will meter
It calculates result to export to piecewise linear function progress calculation processing, Xiang Suoshu output layer exports calculation processing result；
The calculation processing result is input in softmax classification function and calculates every kind of project spy by the output layer
Levy the probability value of attribute；
In above-mentioned training process, according to the probability value of every kind of item characteristic attribute and corresponding item characteristic category
Property calculate cross entropy penalty values, and update the scientific and technological item using stochastic gradient descent algorithm according to the cross entropy penalty values
Mesh declares each layer parameter of flow tracking model, when the cross entropy penalty values meet training termination condition, stops instruction
Practice, output objective model parameter and calculating figure；The science and technology item is obtained according to the objective model parameter and calculating figure to declare
Flow tracking model.
It is described that the training sample is handled in a kind of possible embodiment, obtain insertion to be entered
The step of matrix indicates, comprising:
For each science and technology item for marking and having characteristic attribute information, the associated item of the science and technology item is obtained
Mesh number evidence, the project data include project application data, project category data, item source data, project publication data,
One of which or multiple combinations in project appraisal data；
The project data is filtered, and respectively to project name in filtered project data and corresponding
Item characteristic attribute carries out numeralization processing, generates dictionary file；
Insertion matrix to be entered is obtained after handling the dictionary file to be indicated.
In a kind of possible embodiment, it is described the dictionary file is handled after obtain insertion to be entered
Change the step of matrix indicates, comprising:
The project name quantity and item attribute quantity in the dictionary file are counted, statistical result is generated；According to the system
Meter result generates the length to be entered that the science and technology item declares flow tracking model；It is looked into from preconfigured word embeded matrix
Look for the insertion vector of each project name and item attribute in the dictionary file；Based on the length to be entered and find
Each project name and the insertion vector of item attribute generate insertion matrix to be entered and indicate.
In a kind of possible embodiment, it includes input layer, convolution that the science and technology item, which declares flow tracking model,
Layer, pond layer, full articulamentum and output layer, it is described that the insertionization matrix is indicated that being input to the science and technology item declares
In flow tracking model, export prediction result the step of, comprising:
The insertionization matrix is indicated to be input to the convolutional layer by the input layer；Pass through the convolutional layer pair
The insertionization matrix indicates to carry out convolution algorithm, the corresponding feature extraction information of the pond Xiang Suoshu layer output；
The feature extraction information is converted to by the pond layer to be output to after corresponding multi-C vector and described is connected entirely
Connect layer, wherein the node of the full articulamentum is connect with the node of the pond layer；
The full articulamentum calculates the feature extraction information using whole nodes, and calculated result is exported
After carrying out calculation processing to piecewise linear function, Xiang Suoshu output layer exports corresponding calculation processing result；
The calculation processing result is input in softmax classification function and calculates every kind of project spy by the output layer
Levy the attribute deviation value of attribute.
In a kind of possible embodiment, every kind of project in the tracking prediction result that the basis is calculated
The reference value of every kind of item characteristic attribute of the attribute deviation value of characteristic attribute and the pre-stored science and technology item to be tracked it
Between attribute coefficients, generate the step of declaring flow tracking result of the science and technology item to be tracked, comprising:
According to the attribute deviation value of every kind of item characteristic attribute in the tracking prediction result and it is pre-stored should to
Attribute coefficients between the reference value of every kind of item characteristic attribute of track science and technology item, from preconfigured every kind of item characteristic
Corresponding relationship between the attribute coefficients and item attribute progress of attribute, every kind of project spy of the science and technology item to be tracked
The corresponding item attribute progress of attribute coefficients between the reference value of attribute is levied, to obtain the science and technology item to be tracked
Declare flow tracking result.
According to the another aspect of the embodiment of the present application, a kind of science and technology item is provided and declares flow tracking device, is applied to clothes
Business device, described device include:
Module is obtained, the insertion matrix for obtaining science and technology item to be tracked indicates, in the insertionization matrix expression
Include and the associated project application flow data of science and technology item to be tracked；
Input module declares flow tracking model for insertionization matrix expression to be input to the science and technology item
In, output tracking prediction result, the tracking prediction result includes at least one item characteristic category of the science and technology item to be tracked
The attribute deviation value of property；
Computing module, for calculating in the tracking prediction result attribute deviation value of every kind of item characteristic attribute and pre-
Attribute coefficients between the reference value of every kind of item characteristic attribute of the science and technology item to be tracked first stored；
Generation module, for the attribute according to every kind of item characteristic attribute in the tracking prediction result being calculated
Attribute coefficients between the reference value of every kind of item characteristic attribute of deviation value and the pre-stored science and technology item to be tracked,
Generate the science and technology item to be tracked declares flow tracking as a result, the flow tracking result of declaring includes described to be tracked
The process progress of science and technology item；
Sending module, for generating corresponding science and technology according to the flow tracking result of declaring of the science and technology item to be tracked
Project application suggestion, and the science and technology item declared into suggestion be sent to corresponding science and technology item and declare terminal to prompt correlation
User, which according to the science and technology item declares suggestion and declares information to science and technology item, carries out supplement confirmation.
According to the another aspect of the embodiment of the present application, a kind of readable storage medium storing program for executing is provided, is stored on the readable storage medium storing program for executing
There is computer program, above-mentioned science and technology item can be executed when which is run by processor and declares flow tracking side
The step of method.
Based on any of the above-described aspect, the insertion matrix that the embodiment of the present application obtains science and technology item to be tracked first is indicated,
And the expression of insertionization matrix is input to science and technology item and is declared in flow tracking model, then output tracking prediction result calculates
Every kind of the attribute deviation value of every kind of item characteristic attribute and the pre-stored science and technology item to be tracked in tracking prediction result
Attribute coefficients between the reference value of item characteristic attribute, and according to every kind of project spy in the tracking prediction result being calculated
It levies between the attribute deviation value of attribute and the reference value of every kind of item characteristic attribute of the pre-stored science and technology item to be tracked
Attribute coefficients, generate science and technology item to be tracked declares flow tracking as a result, it includes to be tracked for declaring flow tracking result
The process progress of science and technology item.It so, it is possible to provide guidance and branch that scientific history science and technology project data declares situation
Support, improves the success rate that science and technology item is declared, and efficiency is declared in raising.
To enable the above objects, features, and advantages of the embodiment of the present application to be clearer and more comprehensible, below in conjunction with embodiment,
And cooperate appended attached drawing, it elaborates.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be to needed in the embodiment
Attached drawing is briefly described, it should be understood that the following drawings illustrates only some embodiments of the application, therefore is not to be seen as
It is the restriction to range, it for those of ordinary skill in the art, without creative efforts, can be with
Other relevant attached drawings are obtained according to these attached drawings.
Fig. 1 shows the example hardware of server provided by the embodiment of the present application and the schematic diagram of component software；
Fig. 2 shows the flow diagrams that science and technology item provided by the embodiment of the present application declares method for procedure tracking；
Fig. 3 shows the functional block diagram that science and technology item provided by the embodiment of the present application declares flow tracking device
One of；
Fig. 4 shows the functional block diagram that science and technology item provided by the embodiment of the present application declares flow tracking device
Two.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
In attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it should be understood that attached drawing in the application
The purpose of illustration and description is only played, is not used to limit the protection scope of the application.In addition, it will be appreciated that schematical attached
Figure does not press scale.Process used herein shows some embodiments according to the embodiment of the present application
The operation of realization.It should be understood that the operation of flow chart can be realized out of order, not no the step of context relation of logic
It can implement with reversal order or simultaneously.In addition, those skilled in the art are under the guide of teachings herein, it can be to process
Other one or more operations of figure addition, can also remove one or more operations from flow chart.
In addition, described embodiments are only a part of embodiments of the present application, instead of all the embodiments.Usually
The component of the embodiment of the present application being described and illustrated herein in the accompanying drawings can be arranged and be designed with a variety of different configurations.
Therefore, this claimed Shen is not intended to limit to the detailed description of the embodiments herein provided in the accompanying drawings below
Range please, but it is merely representative of the selected embodiment of the application.Based on embodiments herein, those skilled in the art exist
Every other embodiment obtained under the premise of creative work is not made, shall fall in the protection scope of this application.
Fig. 1 shows the exemplary of the server 100 that the application thought may be implemented that some embodiments of the application provide
The schematic diagram of hardware and software component.For example, processor 120 can be used on electronic equipment 100, and for executing this Shen
Please in function.
Electronic equipment 100 can be the computer of general purpose computer or specific use, both can be used to implement this Shen
Science and technology item please declares method for procedure tracking.The application is although illustrate only a computer, for convenience's sake,
Function described herein can be realized in a distributed way on multiple similar platforms, loaded with equilibrium treatment.
For example, electronic equipment 100 may include the network port 110 for being connected to network, one for executing program instructions
A or multiple processors 120, communication bus 130 and various forms of storage mediums 140, for example, disk, ROM or RAM, or
Any combination thereof.Illustratively, computer platform can also include being stored in ROM, RAM or other kinds of non-transitory is deposited
Storage media, or any combination thereof in program instruction.The present processes may be implemented according to these program instructions.Electronics is set
Standby 100 further include the input/output (Input/ between computer and other input-output equipment (such as keyboard, display screen)
Output, I/O) interface 150.
For ease of description, a processor is only described in electronic equipment 100.It should be noted, however, that the application
In electronic equipment 100 can also include multiple processors, therefore a processor described in this application execute the step of
It can be combined by multiple processors and execute or be individually performed.For example, if the processor of electronic equipment 100 executes step A and step
B, then it should be understood that step A and step B can also be executed or in a processor jointly by two different processors
It is individually performed.For example, first processor executes step A, second processor executes step B or first processor and second
Processor executes step A and B jointly.
Fig. 2 shows the flow diagram that the science and technology item that some embodiments of the application provide declares method for procedure tracking,
The science and technology item declare method for procedure tracking can the server 100 as shown in Fig. 1 execute.It should be appreciated that in other embodiments
In, the sequence that science and technology item described in the present embodiment declares method for procedure tracking part step can be according to actual needs
It is exchanged with each other or part steps therein also can be omitted or delete.The science and technology item declares the detailed of method for procedure tracking
Step is described below.
Step S110, the insertion matrix for obtaining science and technology item to be tracked indicate.
In the present embodiment, include and the associated project Shen of science and technology item to be tracked in the insertionization matrix expression
Report flow data.As an implementation, in the present embodiment, the associated project application process of science and technology item to be tracked
Data can refer to the data of each flow elements involved in the science and technology item to be tracked, such as elementary item data, project
The audit data of current process node during declaring, the demographic data of project related personnel etc..
Insertionization matrix expression is input to the science and technology item and declared in flow tracking model by step S120, defeated
Tracking prediction result out.
As an implementation, before step S120, the present embodiment can also be pre-configured with the science and technology item
Declare flow tracking model, wherein the science and technology item declare flow tracking model include input layer, convolutional layer, pond layer,
Full articulamentum and output layer.
Optionally, configuring the science and technology item and declaring flow tracking model can be accomplished in that
Firstly, obtaining training sample, the training sample includes multiple sections for marking and having characteristic attribute information
Skill project information, for example, item characteristic attribute information can be project contribution, project experiences, project development prospect, project money
Golden strength etc., is not specifically limited herein.
Then, the training sample is handled, obtaining insertion matrix to be entered indicates.Such as it can be directed to
It is each to mark the science and technology item for having characteristic attribute information, obtain the associated project data of the science and technology item, the item
Mesh number is according to including in project application data, project category data, item source data, project publication data, project appraisal data
One of which or multiple combinations.Then the project data is filtered, and respectively in filtered project data
Project name and corresponding item characteristic attribute carry out numeralization processing, generate dictionary file.
On the basis of the above, insertion matrix to be entered is obtained after handling the dictionary file to be indicated.Example
Such as, the project name quantity and item attribute quantity in the dictionary file can be counted, generates statistical result, then basis
The statistical result generates the length to be entered that the science and technology item declares flow tracking model, and embedding from preconfigured word
Enter the insertion vector that matrix searches each project name and item attribute in the dictionary file, finally based on described to be entered
The insertion vector of length and each project name found and item attribute generates insertion matrix to be entered and indicates.
Then, the insertionization matrix is indicated to be input to the convolutional layer by the input layer；Pass through the convolution
Layer indicates that progress convolution algorithm, the pond Xiang Suoshu layer export corresponding feature extraction information, then lead to the insertionization matrix
It crosses after the feature extraction information is converted to corresponding multi-C vector by the pond layer and is output to the full articulamentum, wherein
The node of the full articulamentum is connect with the node of the pond layer.
Then, the node of the full articulamentum random drop predetermined ratio calculates the feature extraction information, and
Calculated result is exported to piecewise linear function and carries out calculation processing, Xiang Suoshu output layer exports calculation processing result.It is described defeated
The calculation processing result is input to the probability value that every kind of item characteristic attribute is calculated in softmax classification function by layer out.
In above-mentioned training process, according to the probability value of every kind of item characteristic attribute and corresponding item characteristic category
Property calculate cross entropy penalty values, and update the scientific and technological item using stochastic gradient descent algorithm according to the cross entropy penalty values
Mesh declares each layer parameter of flow tracking model, when the cross entropy penalty values meet training termination condition, stops instruction
Practice, output objective model parameter and calculating figure；The science and technology item is obtained according to the objective model parameter and calculating figure to declare
Flow tracking model.
The science and technology item obtained based on above-mentioned training declares flow tracking model, it is known that the science and technology item is declared
Flow tracking model includes input layer, convolutional layer, pond layer, full articulamentum and output layer.Thus, it is possible to by the insertion
Changing matrix indicates to be input to the convolutional layer by the input layer；The insertionization matrix is indicated by the convolutional layer
Convolution algorithm is carried out, the pond Xiang Suoshu layer exports corresponding feature extraction information.Then, by the pond layer by the spy
Sign extract information be converted to corresponding multi-C vector after be output to the full articulamentum, wherein the node of the full articulamentum with
The node of the pond layer connects.Then, the full articulamentum counts the feature extraction information using whole nodes
It calculates, and calculated result is exported to piecewise linear function progress calculation processing, Xiang Suoshu output layer exports at corresponding calculating
Manage result.Then, the calculation processing result is input in softmax classification function and calculates every kind of item by the output layer
The attribute deviation value of mesh characteristic attribute.The tracking prediction result includes at least one item of the science and technology item to be tracked as a result,
The attribute deviation value of mesh characteristic attribute.
Step S130, calculates in the tracking prediction result attribute deviation value of every kind of item characteristic attribute and deposits in advance
Attribute coefficients between the reference value of every kind of item characteristic attribute of the science and technology item to be tracked of storage.
Step S140 is biased to according to the attribute of every kind of item characteristic attribute in the tracking prediction result being calculated
Attribute coefficients between the reference value of every kind of item characteristic attribute of value and the pre-stored science and technology item to be tracked, generate
The science and technology item to be tracked declares flow tracking result.
For example, can according to the attribute deviation value of every kind of item characteristic attribute in the tracking prediction result with deposit in advance
Attribute coefficients between the reference value of every kind of item characteristic attribute of the science and technology item to be tracked of storage, from preconfigured every
Corresponding relationship between the attribute coefficients and item attribute progress of kind item characteristic attribute, the science and technology item to be tracked
The corresponding item attribute progress of attribute coefficients between the reference value of every kind of item characteristic attribute, with obtain it is described to
Track science and technology item declares flow tracking result.That is, the flow tracking result of declaring includes the science and technology item to be tracked
Process progress.
Step S150 generates corresponding science and technology item according to the flow tracking result of declaring of the science and technology item to be tracked
Suggestion is declared, and the science and technology item declared into suggestion is sent to corresponding science and technology item and declare terminal to prompt associated user
Suggestion is declared according to the science and technology item information is declared to science and technology item carry out supplement confirmation.
Based on above-mentioned design, the insertion matrix that the present embodiment obtains science and technology item to be tracked first is indicated, and will insertion
Changing matrix indicates that being input to science and technology item declares in flow tracking model, then output tracking prediction result calculates tracking prediction
As a result every kind of item characteristic of the attribute deviation value of every kind of item characteristic attribute and the pre-stored science and technology item to be tracked in
Attribute coefficients between the reference value of attribute, and according to every kind of item characteristic attribute in the tracking prediction result being calculated
Attribute system between the reference value of every kind of item characteristic attribute of attribute deviation value and the pre-stored science and technology item to be tracked
Number, generate science and technology item to be tracked declares flow tracking as a result, declaring flow tracking result includes science and technology item to be tracked
Process progress.It so, it is possible to provide guidance and support that scientific history science and technology project data declares situation, improve
Efficiency is declared in the success rate that science and technology item is declared, raising.
Fig. 3 shows the functional module that the science and technology item that some embodiments of the application provide declares flow tracking device 300
Figure, the function which declares the realization of flow tracking device 300 can correspond to the step of above method executes.The science and technology
Project application flow tracking device 300 can be understood as the processor of above-mentioned server 100 or server 100, can also manage
Solution for independently of the component for realizing the application function except above-mentioned server 100 or processor under the control of server 100,
As shown in figure 3, it may include obtaining module 310, input module 320, calculating that the science and technology item, which declares flow tracking device 300,
Module 330, generation module 340 and sending module 350, the science and technology item declares each of flow tracking device 300 separately below
The function of a functional module is described in detail.
Module 310 is obtained, the insertion matrix for obtaining science and technology item to be tracked indicates, the insertionization matrix table
Include in showing and the associated project application flow data of science and technology item to be tracked；
Input module 320 declares flow tracking mould for insertionization matrix expression to be input to the science and technology item
In type, output tracking prediction result, the tracking prediction result includes at least one item characteristic of the science and technology item to be tracked
The attribute deviation value of attribute.
Computing module 330, for calculate in the tracking prediction result attribute deviation value of every kind of item characteristic attribute with
Attribute coefficients between the reference value of every kind of item characteristic attribute of the pre-stored science and technology item to be tracked.
Generation module 340, for the category according to every kind of item characteristic attribute in the tracking prediction result being calculated
Attribute system between the reference value of every kind of item characteristic attribute of property deviation value and the pre-stored science and technology item to be tracked
Number, generate the science and technology item to be tracked declare flow tracking as a result, it is described declare flow tracking result include it is described to
The process progress of track science and technology item.
Sending module 350, for generating corresponding section according to the flow tracking result of declaring of the science and technology item to be tracked
Skill project application suggestion, and the science and technology item declared into suggestion be sent to corresponding science and technology item and declare terminal to prompt phase
It closes user and suggestion is declared according to the science and technology item information is declared to science and technology item and carry out supplement confirmation.
In a kind of possible embodiment, further referring to Fig. 4, the science and technology item declares flow tracking device
300 can also include:
Configuration module 301 declares flow tracking model for configuring the science and technology item, wherein the science and technology item Shen
Reporting flow tracking model includes input layer, convolutional layer, pond layer, full articulamentum and output layer；
The configuration science and technology item declares the mode of flow tracking model, comprising:
Training sample is obtained, the training sample includes multiple scientific and technological items for marking and having characteristic attribute information
Mesh information；
The training sample is handled, obtaining insertion matrix to be entered indicates；
The insertionization matrix is indicated to be input to the convolutional layer by the input layer；Pass through the convolutional layer pair
The insertionization matrix indicates to carry out convolution algorithm, the corresponding feature extraction information of the pond Xiang Suoshu layer output；
The feature extraction information is converted to by the pond layer to be output to after corresponding multi-C vector and described is connected entirely
Connect layer, wherein the node of the full articulamentum is connect with the node of the pond layer；
The node of the full articulamentum random drop predetermined ratio calculates the feature extraction information, and will meter
It calculates result to export to piecewise linear function progress calculation processing, Xiang Suoshu output layer exports calculation processing result；
The calculation processing result is input in softmax classification function and calculates every kind of project spy by the output layer
Levy the probability value of attribute；
In above-mentioned training process, according to the probability value of every kind of item characteristic attribute and corresponding item characteristic category
Property calculate cross entropy penalty values, and update the scientific and technological item using stochastic gradient descent algorithm according to the cross entropy penalty values
Mesh declares each layer parameter of flow tracking model, when the cross entropy penalty values meet training termination condition, stops instruction
Practice, output objective model parameter and calculating figure；The science and technology item is obtained according to the objective model parameter and calculating figure to declare
Flow tracking model.
It is described that the training sample is handled in a kind of possible embodiment, obtain insertion to be entered
The mode that matrix indicates, comprising:
For each science and technology item for marking and having characteristic attribute information, the associated item of the science and technology item is obtained
Mesh number evidence, the project data include project application data, project category data, item source data, project publication data,
One of which or multiple combinations in project appraisal data；
The project data is filtered, and respectively to project name in filtered project data and corresponding
Item characteristic attribute carries out numeralization processing, generates dictionary file；
Insertion matrix to be entered is obtained after handling the dictionary file to be indicated.
In a kind of possible embodiment, it includes input layer, convolution that the science and technology item, which declares flow tracking model,
Layer, pond layer, full articulamentum and output layer, the input module 320 are specifically used for:
The insertionization matrix is indicated to be input to the convolutional layer by the input layer；Pass through the convolutional layer pair
The insertionization matrix indicates to carry out convolution algorithm, the corresponding feature extraction information of the pond Xiang Suoshu layer output；
The feature extraction information is converted to by the pond layer to be output to after corresponding multi-C vector and described is connected entirely
Connect layer, wherein the node of the full articulamentum is connect with the node of the pond layer；
The full articulamentum calculates the feature extraction information using whole nodes, and calculated result is exported
After carrying out calculation processing to piecewise linear function, Xiang Suoshu output layer exports corresponding calculation processing result；
The calculation processing result is input in softmax classification function and calculates every kind of project spy by the output layer
Levy the attribute deviation value of attribute.
Above-mentioned module can be connected to each other or communicate via wired connection or wireless connection.Wired connection may include gold
Belong to cable, optical cable, mixing cable etc., or any combination thereof.Wireless connection may include by LAN, WAN, bluetooth, ZigBee,
Or the connection of the forms such as NFC, or any combination thereof.Two or more modules can be combined into individual module, and any one
A module is segmented into two or more units.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description
With the specific work process of device, the corresponding process in embodiment of the method can be referred to, is repeated no more in the application.In this Shen
Please provided by several embodiments, it should be understood that disclosed systems, devices and methods, can be by another way
It realizes.The apparatus embodiments described above are merely exemplary, for example, the division of the module, only a kind of logic
Function division, there may be another division manner in actual implementation, in another example, multiple module or components can combine or can
To be integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Coupling, direct-coupling or communication connection can be INDIRECT COUPLING or communication by some communication interfaces, device or module
Connection can be electrical property, mechanical or other forms.
The module as illustrated by the separation member may or may not be physically separated, as module
The component of display may or may not be physical unit, it can and it is in one place, or may be distributed over more
In a network unit.Some or all of unit therein can be selected to realize this embodiment scheme according to the actual needs
Purpose.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
To be stored in the executable non-volatile computer-readable storage medium of a processor.Based on this understanding, this Shen
Substantially the part of the part that contributes to existing technology or the technical solution can be with soft in other words for technical solution please
The form of part product embodies, which is stored in a storage medium, including some instructions are to make
It obtains a computer equipment (can be personal computer, server or the network equipment etc.) and executes each embodiment of the application
The all or part of the steps of the method.And storage medium above-mentioned include: USB flash disk, mobile hard disk, ROM, RAM, magnetic disk or
The various media that can store program code such as CD.
The above is only the protection scopes of the specific embodiment of the application, but the application to be not limited thereto, any to be familiar with
Those skilled in the art within the technical scope of the present application, can easily think of the change or the replacement, and should all cover
Within the protection scope of the application.Therefore, the protection scope of the application should be subject to the protection scope in claims.
Claims (10)
1. a kind of science and technology item declares method for procedure tracking, which is characterized in that be applied to server, which comprises
The insertion matrix for obtaining science and technology item to be tracked indicates, includes and the section to be tracked in the insertionization matrix expression
The project application flow data of skill item association；
Insertionization matrix expression is input to the science and technology item to declare in flow tracking model, output tracking prediction knot
Fruit, the tracking prediction result include the attribute deviation value of at least one item characteristic attribute of the science and technology item to be tracked；
Calculate the attribute deviation value of every kind of item characteristic attribute and the pre-stored section to be tracked in the tracking prediction result
Attribute coefficients between the reference value of every kind of item characteristic attribute of skill project；
According to the attribute deviation value of every kind of item characteristic attribute in the tracking prediction result being calculated with it is pre-stored
Attribute coefficients between the reference value of every kind of item characteristic attribute of the science and technology item to be tracked generate the scientific and technological item to be tracked
Purpose declares flow tracking as a result, described declare the process progress feelings that flow tracking result includes the science and technology item to be tracked
Condition；
It generates corresponding science and technology item according to the flow tracking result of declaring of the science and technology item to be tracked and declares suggestion, and by institute
It states science and technology item and declares suggestion and be sent to corresponding science and technology item and declare terminal to prompt associated user according to the science and technology item
It declares suggestion information is declared to science and technology item and carry out supplement confirmation.
2. science and technology item according to claim 1 declares method for procedure tracking, which is characterized in that be tracked in the acquisition
Before the step of insertion matrix of science and technology item, the method also includes:
It configures the science and technology item and declares flow tracking model, wherein it includes defeated that the science and technology item, which declares flow tracking model,
Enter layer, convolutional layer, pond layer, full articulamentum and output layer；
The configuration science and technology item declares the mode of flow tracking model, comprising:
Training sample is obtained, the training sample includes multiple science and technology item letters for marking and having characteristic attribute information
Breath；
The training sample is handled, obtaining insertion matrix to be entered indicates；
The insertionization matrix is indicated to be input to the convolutional layer by the input layer；By the convolutional layer to described embedding
Entering matrix indicates to carry out convolution algorithm, the corresponding feature extraction information of the pond Xiang Suoshu layer output；
It is output to the full articulamentum after the feature extraction information is converted to corresponding multi-C vector by the pond layer,
Wherein, the node of the full articulamentum is connect with the node of the pond layer；
The node of the full articulamentum random drop predetermined ratio calculates the feature extraction information, and by calculated result
Output to piecewise linear function carries out calculation processing, and Xiang Suoshu output layer exports calculation processing result；
The calculation processing result is input in softmax classification function and calculates every kind of item characteristic attribute by the output layer
Probability value；
In above-mentioned training process, calculated according to the probability value of every kind of item characteristic attribute and corresponding item characteristic attribute
Cross entropy penalty values, and update the science and technology item using stochastic gradient descent algorithm according to the cross entropy penalty values and declare stream
Each layer parameter of journey trace model, when the cross entropy penalty values meet training termination condition, deconditioning exports target
Model parameter and calculating figure；The science and technology item, which is obtained, according to the objective model parameter and calculating figure declares flow tracking mould
Type.
3. science and technology item according to claim 2 declares method for procedure tracking, which is characterized in that described to the trained sample
This is handled, and the step of insertion matrix to be entered indicates is obtained, comprising:
For each science and technology item for marking and having characteristic attribute information, the associated item number of the science and technology item is obtained
According to the project data includes project application data, project category data, item source data, project issues data, project is commented
Examine the one of which or multiple combinations in data；
The project data is filtered, and respectively in filtered project data project name and corresponding project it is special
Sign attribute carries out numeralization processing, generates dictionary file；
Insertion matrix to be entered is obtained after handling the dictionary file to be indicated.
4. science and technology item according to claim 3 declares method for procedure tracking, which is characterized in that described to the dictionary text
Part obtains the step of insertion matrix to be entered indicates after being handled, comprising:
The project name quantity and item attribute quantity in the dictionary file are counted, statistical result is generated；It is tied according to the statistics
Fruit generates the length to be entered that the science and technology item declares flow tracking model；Described in the lookup of preconfigured word embeded matrix
The insertion vector of each project name and item attribute in dictionary file；Based on the length to be entered and each item found
The insertion vector of mesh title and item attribute generates insertion matrix to be entered and indicates.
5. science and technology item according to claim 1 declares method for procedure tracking, which is characterized in that the science and technology item is declared
Flow tracking model includes input layer, convolutional layer, pond layer, full articulamentum and output layer, described by the insertionization matrix
The step of expression is input to the science and technology item and declares in flow tracking model, exports prediction result, comprising:
The insertionization matrix is indicated to be input to the convolutional layer by the input layer；By the convolutional layer to described embedding
Entering matrix indicates to carry out convolution algorithm, the corresponding feature extraction information of the pond Xiang Suoshu layer output；
It is output to the full articulamentum after the feature extraction information is converted to corresponding multi-C vector by the pond layer,
Wherein, the node of the full articulamentum is connect with the node of the pond layer；
The full articulamentum calculates the feature extraction information using whole nodes, and calculated result is exported to segmentation
After linear function carries out calculation processing, Xiang Suoshu output layer exports corresponding calculation processing result；
The calculation processing result is input in softmax classification function and calculates every kind of item characteristic attribute by the output layer
Attribute deviation value.
6. science and technology item according to claim 1 declares method for procedure tracking, which is characterized in that the basis is calculated
The tracking prediction result in every kind of item characteristic attribute attribute deviation value and the pre-stored science and technology item to be tracked
Every kind of item characteristic attribute reference value between attribute coefficients, generate the science and technology item to be tracked declares flow tracking
As a result the step of, comprising:
According to the attribute deviation value of every kind of item characteristic attribute in the tracking prediction result and the pre-stored section to be tracked
Attribute coefficients between the reference value of every kind of item characteristic attribute of skill project, from preconfigured every kind of item characteristic attribute
Corresponding relationship between attribute coefficients and item attribute progress, every kind of item characteristic attribute of the science and technology item to be tracked
The corresponding item attribute progress of attribute coefficients between reference value declares process with obtain the science and technology item to be tracked
Tracking result.
7. a kind of science and technology item declares flow tracking device, which is characterized in that be applied to server, described device includes:
Module is obtained, the insertion matrix for obtaining science and technology item to be tracked indicates, includes in the insertionization matrix expression
Have and the associated project application flow data of science and technology item to be tracked；
Input module is declared in flow tracking model for insertionization matrix expression to be input to the science and technology item, defeated
Tracking prediction is as a result, the tracking prediction result includes the category of at least one item characteristic attribute of the science and technology item to be tracked out
Property deviation value；
Computing module, for calculating in the tracking prediction result attribute deviation value of every kind of item characteristic attribute and being stored in advance
The science and technology item to be tracked every kind of item characteristic attribute reference value between attribute coefficients；
Generation module, for the attribute deviation value according to every kind of item characteristic attribute in the tracking prediction result being calculated
Attribute coefficients between the reference value of every kind of item characteristic attribute of the pre-stored science and technology item to be tracked, described in generation
Science and technology item to be tracked declares flow tracking as a result, the flow tracking result of declaring includes the science and technology item to be tracked
Process progress；
Sending module, for generating corresponding science and technology item Shen according to the flow tracking result of declaring of the science and technology item to be tracked
Report suggest, and by the science and technology item declare suggestion be sent to corresponding science and technology item declare terminal with prompt associated user according to
The science and technology item, which declares suggestion and declares information to science and technology item, carries out supplement confirmation.
8. science and technology item according to claim 7 declares flow tracking device, which is characterized in that described device further include:
Configuration module declares flow tracking model for configuring the science and technology item, wherein the science and technology item declare process with
Track model includes input layer, convolutional layer, pond layer, full articulamentum and output layer；
The configuration science and technology item declares the mode of flow tracking model, comprising:
Training sample is obtained, the training sample includes multiple science and technology item letters for marking and having characteristic attribute information
Breath；
The training sample is handled, obtaining insertion matrix to be entered indicates；
The insertionization matrix is indicated to be input to the convolutional layer by the input layer；By the convolutional layer to described embedding
Entering matrix indicates to carry out convolution algorithm, the corresponding feature extraction information of the pond Xiang Suoshu layer output；
It is output to the full articulamentum after the feature extraction information is converted to corresponding multi-C vector by the pond layer,
Wherein, the node of the full articulamentum is connect with the node of the pond layer；
The node of the full articulamentum random drop predetermined ratio calculates the feature extraction information, and by calculated result
Output to piecewise linear function carries out calculation processing, and Xiang Suoshu output layer exports calculation processing result；
The calculation processing result is input in softmax classification function and calculates every kind of item characteristic attribute by the output layer
Probability value；
In above-mentioned training process, calculated according to the probability value of every kind of item characteristic attribute and corresponding item characteristic attribute
Cross entropy penalty values, and update the science and technology item using stochastic gradient descent algorithm according to the cross entropy penalty values and declare stream
Each layer parameter of journey trace model, when the cross entropy penalty values meet training termination condition, deconditioning exports target
Model parameter and calculating figure；The science and technology item, which is obtained, according to the objective model parameter and calculating figure declares flow tracking mould
Type.
9. science and technology item according to claim 8 declares flow tracking device, which is characterized in that described to the trained sample
This is handled, and the mode that insertionization matrix to be entered indicates is obtained, comprising:
For each science and technology item for marking and having characteristic attribute information, the associated item number of the science and technology item is obtained
According to the project data includes project application data, project category data, item source data, project issues data, project is commented
Examine the one of which or multiple combinations in data；
The project data is filtered, and respectively in filtered project data project name and corresponding project it is special
Sign attribute carries out numeralization processing, generates dictionary file；
Insertion matrix to be entered is obtained after handling the dictionary file to be indicated.
10. science and technology item according to claim 7 declares flow tracking device, which is characterized in that the science and technology item Shen
Report flow tracking model includes input layer, convolutional layer, pond layer, full articulamentum and output layer, and the input module is specific to use
In:
The insertionization matrix is indicated to be input to the convolutional layer by the input layer；By the convolutional layer to described embedding
Entering matrix indicates to carry out convolution algorithm, the corresponding feature extraction information of the pond Xiang Suoshu layer output；
It is output to the full articulamentum after the feature extraction information is converted to corresponding multi-C vector by the pond layer,
Wherein, the node of the full articulamentum is connect with the node of the pond layer；
The full articulamentum calculates the feature extraction information using whole nodes, and calculated result is exported to segmentation
After linear function carries out calculation processing, Xiang Suoshu output layer exports corresponding calculation processing result；
The calculation processing result is input in softmax classification function and calculates every kind of item characteristic attribute by the output layer
Attribute deviation value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910212882.0A CN110119935A (en) | 2019-03-20 | 2019-03-20 | A kind of science and technology item declares method for procedure tracking and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910212882.0A CN110119935A (en) | 2019-03-20 | 2019-03-20 | A kind of science and technology item declares method for procedure tracking and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110119935A true CN110119935A (en) | 2019-08-13 |
Family
ID=67520499
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910212882.0A Pending CN110119935A (en) | 2019-03-20 | 2019-03-20 | A kind of science and technology item declares method for procedure tracking and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110119935A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111951132A (en) * | 2020-07-04 | 2020-11-17 | 宁波大学 | College scientific research project data management method |
| CN113077242A (en) * | 2021-04-30 | 2021-07-06 | 重庆市能源投资集团科技有限责任公司 | Science and technology project declaration management system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226751A (en) * | 2013-03-29 | 2013-07-31 | 广西电网公司电力科学研究院 | Post-evaluation method for science and technology projects of grid enterprise |
| US20170098178A1 (en) * | 2015-10-02 | 2017-04-06 | Beyram Belhaj Amor | Project Assessment Tool |
| CN108388996A (en) * | 2018-03-08 | 2018-08-10 | 青岛东方深蓝信息科技股份有限公司 | A kind of scientific research service integration management system |
| CN108765076A (en) * | 2018-05-24 | 2018-11-06 | 蜜芽宝贝（北京）网络科技有限公司 | Mother and baby's content recommendation method, device and readable storage medium storing program for executing |
-
2019
- 2019-03-20 CN CN201910212882.0A patent/CN110119935A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226751A (en) * | 2013-03-29 | 2013-07-31 | 广西电网公司电力科学研究院 | Post-evaluation method for science and technology projects of grid enterprise |
| US20170098178A1 (en) * | 2015-10-02 | 2017-04-06 | Beyram Belhaj Amor | Project Assessment Tool |
| CN108388996A (en) * | 2018-03-08 | 2018-08-10 | 青岛东方深蓝信息科技股份有限公司 | A kind of scientific research service integration management system |
| CN108765076A (en) * | 2018-05-24 | 2018-11-06 | 蜜芽宝贝（北京）网络科技有限公司 | Mother and baby's content recommendation method, device and readable storage medium storing program for executing |
Non-Patent Citations (1)
| Title |
|---|
| 康珺 等: "基于神经网络的项目评估分析方法", 《太原科技》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111951132A (en) * | 2020-07-04 | 2020-11-17 | 宁波大学 | College scientific research project data management method |
| CN113077242A (en) * | 2021-04-30 | 2021-07-06 | 重庆市能源投资集团科技有限责任公司 | Science and technology project declaration management system |
| CN113077242B (en) * | 2021-04-30 | 2024-03-01 | 重庆市能源投资集团科技有限责任公司 | Science and technology project reporting management system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108229314A (en) | Searching method, device and the electronic equipment of target person | |
| CN108197532A (en) | The method, apparatus and computer installation of recognition of face | |
| CN108351881A (en) | The system and method for optimally in distributed mode | |
| CN109388674A (en) | Data processing method, device, equipment and readable storage medium storing program for executing | |
| CN107770783B (en) | Base station capacity expansion transformation scheme design method and related equipment | |
| CN107947951A (en) | Groups of users recommends method, apparatus and storage medium and server | |
| CN108108743A (en) | Abnormal user recognition methods and the device for identifying abnormal user | |
| CN107833140A (en) | fund conversion method, device, terminal and storage medium | |
| CN110119935A (en) | A kind of science and technology item declares method for procedure tracking and device | |
| CN106326904A (en) | Device and method of acquiring feature ranking model and feature ranking method | |
| CN107341668A (en) | A kind of method and device for judging True Data | |
| CN107845016A (en) | information output method and device | |
| CN110168572A (en) | Information processing method, information processing unit, computer readable storage medium | |
| CN109299722A (en) | Characteristic pattern processing method, device and system and storage medium for neural network | |
| CN107958247A (en) | Method and apparatus for facial image identification | |
| CN110264274A (en) | Objective group's division methods, model generating method, device, equipment and storage medium | |
| CN105894028A (en) | User identification method and device | |
| CN108388512A (en) | Data processing method and server | |
| CN108595448A (en) | Information-pushing method and device | |
| CN108427941A (en) | Method, method for detecting human face and device for generating Face datection model | |
| CN107741967A (en) | Method, apparatus and electronic equipment for behavioral data processing | |
| CN112465141A (en) | Model compression method, model compression device, electronic device and medium | |
| CN111047406B (en) | Telecom package recommendation method, device, storage medium and equipment | |
| US20190205361A1 (en) | Table-meaning estimating system, method, and program | |
| CN110490324A (en) | A kind of gradient decline width learning system implementation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190813 |
|
| RJ01 | Rejection of invention patent application after publication |