CN113128979A - Scientific research aid decision-making system based on big data - Google Patents
Scientific research aid decision-making system based on big data Download PDFInfo
- Publication number
- CN113128979A CN113128979A CN202110536149.1A CN202110536149A CN113128979A CN 113128979 A CN113128979 A CN 113128979A CN 202110536149 A CN202110536149 A CN 202110536149A CN 113128979 A CN113128979 A CN 113128979A
- Authority
- CN
- China
- Prior art keywords
- data
- module
- scientific
- management unit
- database
- 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
- 238000011160 research Methods 0.000 title claims abstract description 51
- 238000007726 management method Methods 0.000 claims abstract description 87
- 238000011161 development Methods 0.000 claims abstract description 11
- 230000008447 perception Effects 0.000 claims description 11
- 238000000034 method Methods 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000013070 change management Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 2
- 230000001105 regulatory effect Effects 0.000 claims description 2
- 238000013439 planning Methods 0.000 abstract description 4
- 230000015572 biosynthetic process Effects 0.000 abstract description 2
- 238000013523 data management Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 5
- 230000014509 gene expression Effects 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 4
- 238000005457 optimization Methods 0.000 description 3
- 239000006185 dispersion Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000012550 audit Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013479 data entry Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Images
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
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
-
- 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
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
Abstract
The invention belongs to the technical field of scientific research data management, and particularly relates to a scientific research aid decision-making system based on big data, which comprises a database, wherein the database comprises a scientific and technological ledger management unit, a scientific research management unit and a result reward management unit; the database inputs data through the data acquisition module; and a data auditing module for auditing data is also arranged between the data acquisition module and the database, if the data is qualified after being audited by the data auditing module, the data is smoothly input into the database, otherwise, the data is automatically returned and modified. According to the invention, through the data information of each project or field which is sorted and counted by the scientific research assistant decision-making system, historical information is formed and stored, so that the system has important significance for users to know market conditions and development dynamics, can assist the users to provide effective data technical support in planning and arrangement of future development, and has important guiding significance for the formation of enterprise development strategies.
Description
Technical Field
The invention relates to the technical field of scientific research data management, in particular to a scientific research aid decision-making system based on big data.
Background
The existing scientific and technological data reporting work mainly adopts paper or electronic form reporting to carry out manual summary, and filing and sorting of data consumes a great deal of energy of management department staff. The scientific and technological management department needs a great deal of scientific and technological information when making scientific and technological decisions and planning scientific and research force input directions or decisions.
When the large amount of scientific research data is used specifically, the entering or calling of the data becomes slow along with the increase of the data in the database, manual step-by-step selection operation is needed, when the data is entered manually, once the data is wrong, the use of the whole system is influenced, and particularly, uncontrollable error factors exist in the manual entering during the data entry; meanwhile, when data is called or abnormal increase and decrease occur, correction is difficult to find by people.
Therefore, when a user needs to make decisions and plans on enterprise development and future planning by combining with the existing scientific and technical information, it is difficult to efficiently and quickly obtain accurate relevant data, and the predictability and the rate of decisions are influenced.
Disclosure of Invention
The invention provides a scientific research aid decision-making system based on big data, which is based on the technical problem that the existing scientific research data is easy to obstruct the direction or decision of the scientific research data when being recorded and called or abnormal.
The scientific research assistant decision making system based on big data comprises a database, wherein the database comprises a scientific and technological ledger management unit, a scientific research management unit and a result reward management unit;
the database acquires data corresponding to the scientific and technological standing book management unit, the scientific and technological management unit and the achievement reward management unit through a data acquisition module, and the data acquired by the data acquisition module comprises external data and internal input data;
a data auditing module used for auditing data is also arranged between the data acquisition module and the database, if the data is qualified after being audited by the data auditing module, the data is smoothly input into the database, and if the data is unqualified after being audited by the data auditing module, the data is automatically returned and modified;
the data auditing module marks the data after the data is approved and qualified, and then stores the data into the database through the data marking module;
the marked data in the database is finally displayed and called out by the display module after being sensed by the data sensing module and compared with the comparator, and the data sensing module comprises a sensing machine for solving the second classification of the data in the database.
Preferably, the scientific and technological standing book management unit comprises a plurality of scientific and technological standing book operation modules, and each scientific and technological standing book operation module comprises a patent standing book sub-module, a software copyright standing book sub-module, a result standing book sub-module, a reward standing book sub-module, a standard standing book sub-module, a first-aid technology standing book sub-module, an innovation platform standing book sub-module, a scientific and technological development plan standing book sub-module, a work and comment standing book sub-module, an expert resource library sub-module and a subject library sub-module;
the scientific research management unit comprises a plurality of scientific research modules, and each scientific research module comprises a guide suggestion collection submodule, a subject item establishment submodule, a subject contract signing submodule, a subject progress management submodule, a scientific research expense payment submodule, a subject change management submodule, an outcome check and acceptance submodule and an external subject record declaration submodule;
the result reward management unit comprises a plurality of result reward management modules, and each result reward management module comprises a result evaluation sub-module, a reward evaluation sub-module and a social reward filing and reporting sub-module.
Through above-mentioned technical scheme, to scientific and technological standing book management unit, scientific research management unit and achievement reward management unit and carry out the management of submodule piece, can be more careful carry out classification to scientific research data, careful processing, the different functional module of the operation or the function submodule piece of convenience of customers selectability can both realize intelligent statistics to different operating function, improves the accuracy of data.
Preferably, the data auditing module comprises a comparison module with an adjustable threshold value and a data packet recorded by the data acquisition module, the data in the data packet is compared with the threshold value in the comparison module,
the threshold value comprises an upper limit threshold value and a lower limit threshold value, when the data in the data packet is higher than the upper limit threshold value and lower than the lower limit threshold value, the data is judged to be unqualified, and the data automatically returns to be modified.
Through the technical scheme, once the data in each data packet exceeds the threshold value, the data comprises the upper limit threshold value and the lower limit threshold value, for example, when decimal points in the data are dislocated, the data are automatically returned and modified, so that a large error is prevented, and the accuracy of the whole data is improved.
Preferably, the threshold in the comparison module is adjusted or set by the system configuration.
Through the technical scheme, the threshold value in the comparison module is set by a system administrator through system configuration.
Preferably, the data perception module is further provided with a comparator which is regulated and controlled by system configuration, and when the data perception module perceives that the data in the database is larger than or smaller than a threshold value set by the comparator, the data is displayed and prompted through the display module.
Through the technical scheme, the threshold value in the comparator can be adjusted through system configuration, and the function can be limited or set by a system administrator according to a manager at a specific level.
Preferably, the process of marking data by the data marking module includes the following steps:
acquiring data qualified by the data auditing module, wherein the data comprises data acquired from a scientific ledger management unit, a scientific research management unit and a result reward management unit,
filtering data acquired by a scientific and technological ledger management unit, a scientific and technological management unit and a result reward management unit (13) and setting corresponding content labels, wherein the labels comprise large-class labels and various-class labels to form a label library;
and importing and storing the data subjected to the tags into the database respectively.
Preferably, the marked data in the database is finally displayed and called by the display module after being sensed by the data sensing module and compared with the comparator, and the data calling specifically comprises the following steps:
determining the direction of object data to be called, calling the label of the object data through a label library, transmitting the associated data to a display module through probability statistics,
the display module displays the data statistical result in a report form or a chart form, and corresponding operation buttons are respectively arranged for selecting the report form or the chart form.
The beneficial effects of the invention are as follows:
1. through setting up the data and examining and verifying the module, preliminary automated inspection detects to the data of typeeing that can be abundant, when surpassing the threshold value of setting for in the contrast module, and data is automatic to be beaten back the modification, can effectually prevent that the initial data in the database from being entered into by mistake.
2. Through setting up the data perception module, the extraction speed that causes because of categorised unclear when can be abundant realizing extracting to the data in the database is slow, and accessible system configuration simultaneously sets for the threshold value in the comparator, and when certain or some data are obvious unusual after categorised, can directly show on the display module after the mark through setting for the threshold value is reminded through its mark itself, and the self-checking decision-making of supplementary entire system realization data local error.
Drawings
FIG. 1 is a schematic diagram of a scientific research aid decision-making system based on big data according to the present invention;
FIG. 2 is a data acquisition module audit block diagram of a scientific research aid decision-making system based on big data according to the present invention;
FIG. 3 is a functional diagram of a linearly separable data set in a hyperplane in a perceptron of a scientific research aid decision-making system based on big data according to the present invention;
FIG. 4 is a functional diagram of two overlapping classes of data sets in a perceptron of a big data-based scientific research aid decision-making system according to the present invention;
FIG. 5 is a non-differentiable perceptron function diagram of a scientific research aid decision-making system based on big data according to the present invention.
In the figure: 1. a database; 11. a scientific and technological ledger management unit; 12. a scientific research management unit; 13. a result reward management unit; 2. a data acquisition module; 3. a data auditing module; 31. a comparison module; 32. a data packet; 4. a data marking module; 5. a data perception module; 6. a comparator; 7. a display module; 8. and (5) configuring the system.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
Referring to fig. 1-5, a scientific research aid decision-making system based on big data, in fig. 1, includes a database 1, the database 1 includes three management units, namely a scientific ledger management unit 11, a scientific research management unit 12 and a result reward management unit 13, and the functions of the three management units can also be flexibly adjusted according to actual needs;
the scientific and technological standing book management unit 11 comprises a plurality of scientific and technological standing book operation modules, each scientific and technological standing book operation module comprises a patent standing book sub-module, a software copyright standing book sub-module, a result standing book sub-module, a reward standing book sub-module, a standard standing book sub-module, a first-aid technological standing book sub-module, an innovation platform standing book sub-module, a scientific and technological development plan standing book sub-module, a work and annotation standing book sub-module, an expert resource library sub-module and a subject library sub-module, each sub-module corresponds to corresponding data information and operation steps, a worker can conveniently input, count data, analyze and the like the work of the corresponding sub-module according to needs, the setting of each function is set by a user according to actual conditions, and in addition, the corresponding sub-module can also be adjusted, increased or replaced according;
the scientific research management unit 12 comprises a plurality of scientific research modules, wherein each scientific research module comprises a guide suggestion collection submodule, a subject item establishment submodule, a subject contract signing submodule, a subject progress management submodule, a scientific research expense payment submodule, a subject change management submodule, a subject acceptance check submodule and an external subject record declaration submodule;
the result reward management unit 13 comprises a plurality of result reward management modules, and the result reward management modules comprise a result evaluation sub-module, a reward evaluation sub-module and a social reward filing reporting sub-module, so that scientific research data can be more carefully managed in a classified and refined manner.
According to the technical scheme, a basic database 1 framework is constructed, and a framework of a scientific and technological ledger management unit 11, a scientific research management unit 12 and a result reward management unit 13, a page, a data format and a conversion formula among the units are constructed in advance.
The database 1 acquires data information corresponding to the scientific and technological standing book management unit 11, the scientific and technological management unit 12 and the achievement reward management unit 13 through the data acquisition module 2; the data collected by the data collection module comprise external data and internal input data, so that the source of the data is increased, and the reliability, integrity and authenticity of the data are enriched. The complete database provides a good storage basis for a user when the user needs to analyze historical data, the analyzed result is more practical, and the user can be assisted to make a decision well.
A data auditing module 3 for auditing data is also arranged between the data acquisition module 2 and the database 1, for example, in the process of inputting the relevant patent information in the patent standing book sub-module and the software copyright standing book sub-module, the reasonability of the data information can be audited, if the data is approved and qualified by the data auditing module 3, the data is smoothly input into the database 1, and if the data is not qualified after being approved by the data auditing module 3, the data is automatically returned and modified.
The auditing standard of the data auditing module 3 is set in a targeted manner according to different functional modules in the scientific and technological ledger management unit 11, the scientific research management unit 12 and the achievement reward management unit 13, for example, if the success of a certain research and development in the success reward management unit 13 is broken through in a staged manner, reward is awarded, the reward amount of the reward is preset in a range of 1000-5000 yuan RMB, when the amount of data input through the data acquisition module 2 exceeds 5000 yuan or is lower than 1000 yuan, the data auditing module 3 defaults that the input of the data is abnormal and needs to be returned and modified, the operation has an important meaning for preliminary screening of the data, and the influence of the input of abnormal data on subsequent work is avoided. The auditing reference standard of the data auditing module 3 is preset according to the actual situation.
Further, referring to fig. 2, the data auditing module 3 includes a comparison module 31 capable of adjusting a threshold and a data packet 32 recorded by the data acquisition module 2, data in the data packet 32 is compared with the threshold in the comparison module 31, the threshold includes an upper threshold and a lower threshold, when the data in the data packet 32 is higher than the upper threshold and lower than the lower threshold, the data is determined to be unqualified, and the data is automatically returned and modified.
When data is recorded, once the data in each data packet exceeds a threshold value, for example, decimal points in the data are staggered, and phrase and sentence combination among characters are obviously not smooth, the data is automatically returned and modified, so that a large error is prevented, qualified data is recorded into the database 1 after being marked by the data marking module 4 according to the data type, and the accuracy of the whole data is improved.
Through setting up data and examining module 3, preliminary automated inspection detection is done to the data of typing in that can be abundant, and when exceeding the threshold value that sets for in the comparison module 31, the data is automatic to be put back the modification, can effectually prevent that the initial data in the database 1 from being by wrong typing in into.
Specifically, the process of marking the data by the data marking module 4 includes the following steps:
acquiring data qualified by the data auditing module 3, wherein the data comprises data acquired from a scientific standing book management unit 11, a scientific research management unit 12, a result reward management unit 13 and the like,
filtering data acquired by the scientific and technological ledger management unit 11, the scientific and technological management unit 12 and the achievement reward management unit 13 and setting corresponding content labels, wherein the labels comprise large-class labels and various-class labels to form a label library; the label library can store multiple and multistage labels, a layered three-dimensional label system is constructed, the pertinence is stronger, the classification is more definite, the data after the labels are respectively imported and stored into the database 1, and a user can more quickly find required data information according to the labels in the calling process conveniently at the back.
Further, the threshold value is set or adjusted in the comparison module 31 by the system configuration 8, and the setting of the threshold value can realize automatic preliminary screening. The threshold in the comparison module 31 is set by the system administrator through the system configuration 8.
The data auditing module 3 marks the data after the data is approved and qualified through the data marking module 4 and then stores the data into the database 1;
the marked data in the database 1 are finally displayed and retrieved by the display module 7 after being sensed by the data sensing module 5 and compared with the comparator 6, specifically, the data sensing module 5 is further provided with the comparator 6 which is adjusted and controlled by the system configuration 8, and when the data sensing module 5 senses that the data in the database 1 is larger than or smaller than the threshold value set by the comparator 6, the data are displayed and prompted by the display module 7.
After the marked data in the database 1 is sensed by the data sensing module 5 and compared with the comparator 6, the marked data is finally displayed and called by the display module 7, and the data calling specifically comprises the following steps:
the direction of the object data to be called is determined, the label of the object data is called through a label library, the related data is transmitted to a display module 7 through probability statistics, the data with the same or similar direction to the called data is called through the label library, and the data is provided for a user to select according to the similarity probability statistics and the sequence of the relevance.
The display module 7 displays the data statistical result selected by the user in a report form or a chart form, the selection of the report form or the chart form is respectively provided with a corresponding operation button, and the user can click the corresponding operation button to obtain the data in the chart format or the report form format, so that different users can conveniently select the data according to the preference of the users.
The data perception module 5 comprises a perception machine for solving the two-classification of data in the database 1, the perception machine is a computer intelligent interface system which integrates multiple channels of voice, characters, sign language, human faces, expressions, lip reading, head gestures, body gestures and the like and encodes, compresses, integrates and fuses information of the channels, and the perception machine technology is applied to the technical field of data processing and analysis.
In the invention, the data of the scientific and technological standing book management unit 11, the scientific research management unit 12 and the achievement reward management unit 13 in the database 1 are classified into two categories through the sensing machine. Specifically, the perceptron includes a training set of P successive value input and output data points
Our goal is to learn a hyperplane b + x with parameters b and wTw, to
b+xTw≈yp；
P holds true for all P1. A very different but naive motivation for the linear classification problem has been to search for another ideal hyperplane. Instead of the goal of linear regression being to represent a data set, the goal of classification is to separate two different classes of human output data using a learned hyperplane. That is, one wants to learn a hyperplane b + xTw-0, separating two types of data points as much as possible, one type of data being "above" the hyperplane, i.e., b + xTw > 0, and the other data is located "below" the hyperplane, i.e., b + xTw < 0.
Then, classifying all data, namely classifying the data in the corresponding modules of the scientific and technological ledger management unit 11, the scientific research management unit 12 and the achievement reward management unit 13 one by using a single or a plurality of sensing machines according to specific data types, wherein the specific method comprises the following steps:
p1, establishing a standard line or plane or curved surface of data classification, wherein the data can be data corresponding to each functional module in the scientific and technological ledger management unit 11, the scientific research management unit 12 and the achievement reward management unit 13;
more formally, for the binary problem, there is still a training set with P missing output data points
And each inputs xpAre N-dimensional (each dimension represents an input feature, like a regression problem). However, the output data is no longer a continuous value, but two discrete values or labels, which are used to indicate class membership, i.e., data points belonging to different classes are assigned different labels. Although two values may be optionally chosen to achieve the above goal, it was found that 1 is particularly useful for the values, so it is assumed that for P ═ 1,. P, ypE { -1, +1}, the learned hyperplane is used to separate two different classes of people-losing output data.
As shown in FIGS. 3-4, for linear classification, a hyperplane b + x is learnedTw is 0, the feature space is divided into two parts, and one part satisfies b + xTw > 0, and the other portion satisfies b + xTw < 0, thereby separating the two classes of features. For example, one class is labeled red, the class is denoted "+ 1", another class is labeled blue, and the class is denoted "-1".
b+xTw＝0
In FIG. 3, a linearly separable data set is shown on which a hyperplane can be learned to completely separate the two classes.
In FIG. 4, a data set containing two overlapping classes is shown. Although the distribution of the data does not allow for a complete linear separation, a hyperplane can still be found, minimizing the amount of erroneous data. A misclassification means that the data points are erroneously present in the other half of the space.
P2, whether the data in each data packet falls into a qualified range is measured by a standard line or a plane or a curved surface, and unqualified data is kicked out by the sensing machine when the data is not in or exceeds the threshold value set by the standard line or the plane or the curved surface, so that the data is prevented from not conforming to the preset range or not meeting the user requirement;
learning parameters b and w of the hyperplane to satisfy the first class (where yp+1) maximally distributed over a hyperplane, i.e. b + xTw > 0, and a second class (where yp1) maximally distributed below the hyperplane, i.e. composed of b + xTw < 0. If a given hyperplane can map data points xpPlaced to the right side (or the data points are correctly classified), then it is possible to accurately derive:
The data classification is distinguished by realizing lines or planes, so that the problem that the data is omitted and classified can be fully avoided.
Since 1 is selected as a tag, y can be determined by multiplying two expressions by the inverse of their respective tagspTo express it succinctly, this yields an equivalent expression:
by taking the maximum between this value and the zero value, the hyperplane point pair x can be writtenpThe conditions under which the correct classification is carried out,is equivalent to:
the adjustment of the data abnormality degree can be realized, and when the data abnormality dispersion exceeds the threshold value set by the comparator 6, the data abnormality dispersion is automatically classified as abnormality and displayed.
It should be noted that if xpIs correctly classified, then the expression
The value is zero; and when the point is classified as erroneous, the expression takes on a positive value. This approach is effective not only because it describes the desired hyperplane, but more importantly, by simply adding the expressions for all points, a non-negative rhyme S cost function is obtained:
this is called the perceptron or maximum cost function; when the point data is classified by mistake, the target function has a lower bound, and in the optimization process, if the optimization algorithm can continuously reduce the target function, the optimization algorithm can prove to be effective in convergence according to a monotone bounded criterion. When the objective function is designed to have a lower bound, the data classification is more convenient if the cost function is not negative.
When the objective function presents a lower bound minimization problem:
and then determining the optimal parameters for segmenting the hyperplane, there are two obvious technical problems with respect to minimization itself. First, g1Where b is 0 and w is 0N*1Does not meet the classification requirement but can ensure g1The minimum value is taken (it does make g1 ═ 0). Secondly, although g is1Is continuous (convex in nature), but it is not differentiable everywhere, and therefore the gradient descent method and newton's method cannot be used. A simple solution to both of these problems is to make a specific smooth approximation to the perceptron function.
As shown in fig. 5, the cost g(s) max (0, s) of the imperceptible perceptron or hinge and its smooth softmax approximation g(s) log (1+ e)s)；
That is, when the lower-bound minimum value occurs in the data, it is necessary to predict the lower-bound classification line of the next data, so that the data is constructed into the minimum bound of the lower-bound classification curve or curved surface by the line or surface formed by connecting the predicted data, and the neural network prediction effect is achieved, so that the data classification intelligent effect is achieved, and the method has important application in the data intelligent classification process for the scientific and technical ledger management unit 11, the scientific research management unit 12 and the achievement reward management unit 13.
When the data sensing module 5 senses that the data in the database 1 is larger than or smaller than the threshold value set by the comparator 6, the data is displayed and prompted through the display module 7.
P3, the proposed failure data is compared with the threshold in the comparator 6 and then displayed and flashed on the following module 7, and the result is sent to the person with the authority to modify data in the system configuration 8 for manual judgment, so as to assist in manually auditing the data.
Through setting up data perception module 5, the extraction speed that causes because of categorised unclear when can be abundant realizing extracting the data in database 1 is slow, and threshold value in accessible system configuration 8 contrast ware 6 is set for simultaneously, and when certain or some data are obvious unusual after categorised, can be directly through the mark of itself that sets for the threshold value and show on display module 7 after being reminded, the self-checking decision-making that the whole system realized the local error of data is assisted.
According to the invention, historical information is formed and stored by sorting and counting the data information of each project or field by means of the scientific research aid decision-making system, when a user needs to know the market situation or the historical development situation of the field, the historical situation can be automatically analyzed and displayed by only calling the data information, so that the scientific research aid decision-making system has important significance for the user to know the market situation and the development dynamic, can assist the user to provide effective data technical support in planning and arrangement of future development, and has important guiding significance for the formation of enterprise development strategy.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (7)
1. A scientific research aid decision-making system based on big data comprises a database (1) and is characterized in that: the database (1) comprises a scientific and technological standing book management unit (11), a scientific and technological management unit (12) and a result reward management unit (13);
the database (1) acquires data corresponding to the scientific and technological ledger management unit (11), the scientific research management unit (12) and the achievement reward management unit (13) through a data acquisition module (2), wherein the data acquired by the data acquisition module (2) comprises external data and internal input data;
a data auditing module (3) for auditing data is also arranged between the data acquisition module (2) and the database (1), if the data is approved and qualified by the data auditing module (3), the data is smoothly input into the database (1), and if the data is not qualified after being approved by the data auditing module (3), the data is automatically returned and modified;
the data auditing module (3) marks the data after the data is approved and qualified through the data marking module (4) and then stores the data into the database (1);
the marked data in the database (1) are finally displayed and called out by a display module (7) after being sensed by a data sensing module (5) and compared with a comparator (6), and the data sensing module (5) comprises a sensing neural network for solving the second classification of the data in the database (1).
2. The big data-based scientific aided decision making system according to claim 1, wherein: the scientific and technological standing book management unit (11) comprises a plurality of scientific and technological standing book operation modules, wherein each scientific and technological standing book operation module comprises a patent standing book sub-module, a software copyright standing book sub-module, a result standing book sub-module, a reward standing book sub-module, a standard standing book sub-module, a first-aid technology standing book sub-module, an innovation platform standing book sub-module, a scientific and technological development plan standing book sub-module, a work and annotation standing book sub-module, an expert resource library sub-module and a subject library sub-module;
the scientific research management unit (12) comprises a plurality of scientific research modules, wherein each scientific research module comprises a guide suggestion collection submodule, a subject item establishment submodule, a subject contract signing submodule, a subject progress management submodule, a scientific research expense payment submodule, a subject change management submodule, a subject acceptance check submodule and an external subject record declaration submodule;
the result reward management unit (13) comprises a plurality of result reward management modules, and each result reward management module comprises a result evaluation sub-module, a reward evaluation sub-module and a social reward filing and reporting sub-module.
3. The big data-based scientific aided decision making system according to claim 1, wherein: the data auditing module (3) comprises a comparison module (31) with an adjustable threshold value and a data packet (32) collected by the data collecting module (2), the data auditing module (3) compares the data in the data packet (32) with the threshold value in the comparison module (31),
the threshold value comprises an upper limit threshold value and a lower limit threshold value, when the data in the data packet (32) is higher than the upper limit threshold value and lower than the lower limit threshold value, the data is judged to be unqualified, and the data automatically returns to be modified.
4. The big data-based scientific aided decision making system according to claim 3, wherein: the threshold value in the comparison module (31) is adjusted or set by the system configuration (8).
5. The big data-based scientific aided decision making system according to claim 1, wherein: the data perception module (5) is further provided with a comparator (6) regulated and controlled by a system configuration (8), and when the data perception module (5) perceives that the data in the database (1) is larger than or smaller than a threshold value set by the comparator (6), the data is displayed and prompted through the display module (7).
6. The big data-based scientific aided decision making system according to claim 1, wherein: the process of marking data by the data marking module (4) comprises the following steps:
acquiring data qualified by the data auditing module (3), wherein the data comprises data acquired from a scientific and technological ledger management unit (11), a scientific research management unit (12) and a result reward management unit (13),
filtering data acquired by a scientific and technological ledger management unit (11), a scientific and technological management unit (12) and a result reward management unit (13) and setting corresponding content labels, wherein the labels comprise large-class labels and all-level sub-class labels to form a label library;
the data after the labels are respectively imported and stored into the database (1).
7. The big data-based scientific aided decision making system according to claim 1, wherein: the marked data in the database (1) are finally displayed and called out by a display module (7) after being sensed by a data sensing module (5) and compared with a comparator (6), and the data calling specifically comprises the following steps:
determining the direction of the object data to be called, calling the label of the object data through a label library, transmitting the associated data to a display module (7) through probability statistics,
the display module (7) displays the data statistical result in a report form or a chart form, and corresponding operation buttons are respectively arranged for selecting the report form or the chart form.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110536149.1A CN113128979A (en) | 2021-05-17 | 2021-05-17 | Scientific research aid decision-making system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110536149.1A CN113128979A (en) | 2021-05-17 | 2021-05-17 | Scientific research aid decision-making system based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113128979A true CN113128979A (en) | 2021-07-16 |
Family
ID=76782069
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110536149.1A Pending CN113128979A (en) | 2021-05-17 | 2021-05-17 | Scientific research aid decision-making system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113128979A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115860925A (en) * | 2023-02-19 | 2023-03-28 | 广东德澳智慧医疗科技有限公司 | Intelligent data query and investment management system based on artificial intelligence liability |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104123622A (en) * | 2014-08-18 | 2014-10-29 | 张家港市锦丰镇人民医院 | Performance management system |
| CN110334937A (en) * | 2019-06-28 | 2019-10-15 | 贵州省人民医院 | A kind of hospital research management system |
| CN110490237A (en) * | 2019-08-02 | 2019-11-22 | Oppo广东移动通信有限公司 | Data processing method, device, storage medium and electronic equipment |
| KR102093080B1 (en) * | 2019-12-06 | 2020-04-27 | 주식회사 애자일소다 | System and method for classifying base on generative adversarial network using labeled data and unlabled data |
| CN111581008A (en) * | 2020-06-17 | 2020-08-25 | 荆门汇易佳信息科技有限公司 | Abnormal value rapid and accurate detection method based on parallel cloud computing |
-
2021
- 2021-05-17 CN CN202110536149.1A patent/CN113128979A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104123622A (en) * | 2014-08-18 | 2014-10-29 | 张家港市锦丰镇人民医院 | Performance management system |
| CN110334937A (en) * | 2019-06-28 | 2019-10-15 | 贵州省人民医院 | A kind of hospital research management system |
| CN110490237A (en) * | 2019-08-02 | 2019-11-22 | Oppo广东移动通信有限公司 | Data processing method, device, storage medium and electronic equipment |
| KR102093080B1 (en) * | 2019-12-06 | 2020-04-27 | 주식회사 애자일소다 | System and method for classifying base on generative adversarial network using labeled data and unlabled data |
| CN111581008A (en) * | 2020-06-17 | 2020-08-25 | 荆门汇易佳信息科技有限公司 | Abnormal value rapid and accurate detection method based on parallel cloud computing |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115860925A (en) * | 2023-02-19 | 2023-03-28 | 广东德澳智慧医疗科技有限公司 | Intelligent data query and investment management system based on artificial intelligence liability |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10332007B2 (en) | Computer-implemented system and method for generating document training sets | |
| CN1701324B (en) | Systems, methods, and software for classifying text | |
| US8495002B2 (en) | Software tool for training and testing a knowledge base | |
| US20070050187A1 (en) | Medical billing system and method | |
| US20060089924A1 (en) | Document categorisation system | |
| CN112488507B (en) | Expert classification portrait method and device based on clustering and storage medium | |
| CN110909542A (en) | Intelligent semantic series-parallel analysis method and system | |
| CN109344227A (en) | Worksheet method, system and electronic equipment | |
| CN112184484A (en) | Differentiated service method and system for power users | |
| CN116757808A (en) | Automatic bidding document generation method and system based on big data | |
| CN113128979A (en) | Scientific research aid decision-making system based on big data | |
| CN116578703A (en) | Intelligent identification system and method | |
| CN112579784A (en) | Cloud edge collaborative document classification system and method based on deep reinforcement learning | |
| CN115439919B (en) | Model updating method, device, equipment, storage medium and program product | |
| CN116843175A (en) | Contract term risk checking method, system, equipment and storage medium | |
| CN108460119A (en) | A kind of system for supporting efficiency using machine learning lift technique | |
| CN115544235A (en) | Power grid planning intelligent question-answering system based on text parsing | |
| CN109635254A (en) | Paper duplicate checking method based on naive Bayesian, decision tree and SVM mixed model | |
| TWI761090B (en) | Dialogue data processing system and method thereof and computer readable medium | |
| US20210312319A1 (en) | Artificial Intelligence Process Automation for Enterprise Business Communication | |
| Gaut et al. | Improving Government Response to Citizen Requests Online | |
| CN113269101A (en) | Bill identification method, device and equipment | |
| KR20220005167A (en) | Method for analysis of legal document based on standard document using artificial intelligence, and apparatus for the same | |
| CN117332761B (en) | PDF document intelligent identification marking system | |
| CN117540727B (en) | Subjective question scoring method and system based on ALBERT model and RPA technology |
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 |