WO2023130774A1

WO2023130774A1 - Data acquisition system for scientific research capability assessment based on subject development

Info

Publication number: WO2023130774A1
Application number: PCT/CN2022/121792
Authority: WO
Inventors: 张颖聪; 武青松; 马鸣; 向璨; 陈实; 吴建才; 金阳; 王征; 罗飞; 王智慧
Original assignee: 华中科技大学同济医学院附属协和医院
Priority date: 2022-01-07
Filing date: 2022-09-27
Publication date: 2023-07-13
Also published as: CN114358611A

Abstract

The present invention specifically relates to the technical filed of scientific research capability assessment, and disclosed is a data acquisition system for scientific research capability assessment based on subject development. The method comprises: determining basic important features of class clustering data by means of a random algorithm; clustering the features by means of a clustering algorithm to obtain clustered feature data; performing deep calculation on average distinguishing values of category features; performing dimensionality reduction arrangement to obtain data features which are clearly expressed by means of a line chart, such that accurate assessment data can be obtained; establishing confidence and support degree to find out given and satisfied minimum support degree and minimum credibility, and mining and analyzing association rules; and performing quantitative calculation to obtain a quantitative result index value. By means of the clustering processing, dimensionality reduction arrangement, and data association analysis processing, accurate information is ensured to be extracted, such that a practical foundation can be provided for subsequent assessment, and the credibility and authenticity of scientific research result assessment are improved.

Description

A Data Acquisition System for Evaluation of Scientific Research Ability Based on Subject Development

Cross References to Related Applications

This application claims the priority of a Chinese patent application with application number 202210015385.3 and titled "Data Collection for Evaluation of Scientific Research Capabilities Based on Disciplinary Development" filed on January 07, 2022, the entire contents of which are hereby incorporated by reference.

technical field

The disclosure relates to the technical field of scientific research capability assessment, and in particular to a data collection system for scientific research capability assessment based on subject development.

Background technique

Science and technology is the primary productive force, and scientific research is one of the basic functions of medical colleges and universities. As an important part of medical colleges, affiliated hospitals, in addition to undertaking the function of treating diseases and saving lives, scientific research is also an important part of it. At the same time, scientific research capabilities are One of the important indicators to measure the comprehensive strength of medical colleges. Medicine is a highly differentiated and comprehensive discipline. Strong discipline construction is one of the keys to the sustainable development of hospitals. Discipline construction is a basic project for hospital construction and development. , undertaking long-term strategic tasks.

Scientific research capability evaluation is a scientific and technological consulting activity aimed at improving the level of scientific and technological management and decision-making. According to specific purposes, follow certain principles, procedures and indicators, and use scientific, fair and feasible methods to comprehensively analyze scientific and technological activities and their related environments. and judgments, and make qualitative and quantitative evaluations and estimates. The evaluation of scientific research capabilities should not only be limited to scientific and technological activities themselves, but also consider the environmental conditions and effects of scientific and technological activities. It is necessary to comprehensively describe scientific research strengths, accurately To grasp scientific research strength accurately, several indicators should be used to describe the internal structure of scientific research strength, and a complete index system should be used to comprehensively describe scientific research strength, so as to give full play to the scientific evaluation and assessment function of the index system and its guiding role in the development of scientific research.

The scientific research level of a hospital depends on the scientific research ability of doctors, among which the scientific research ability evaluation is one of the effective ways to test the scientific research level, and it also provides a reference for strengthening the hospital's scientific research management and formulating scientific research development planning strategies. In the process of evaluation, it is often necessary to collect evaluation data first. In the prior art, according to the characteristics of quantitative evaluation, in order to better describe the qualitative characteristics of scientific research results, it is often necessary to collect data from the scientific research results of the personnel being evaluated. However, Generally, only the quantity of scientific research achievements is considered during collection, and the ranking of scientific research personnel in the scientific research results is not considered, which makes the accuracy of scientific research evaluation low, and due to the diversity of information sources and the complexity of data structures, etc. It is difficult to extract accurate and effective information, and there is no correlation analysis for the collected data, which makes the data scattered and inconvenient for subsequent evaluation. Therefore, a data acquisition system for scientific research ability evaluation based on discipline development is studied. It is important to solve the above problems.

Contents of the invention

In order to overcome the above-mentioned defects of the prior art, the present disclosure provides a data acquisition system for evaluating scientific research capabilities based on subject development, and the technical problem to be solved by utilizing one or more embodiments of the present disclosure is: prior art In general, according to the characteristics of quantitative evaluation, in order to better describe the qualitative characteristics of scientific research results, it is often necessary to collect data on the scientific research results of the evaluated personnel. The ranking in scientific research results makes the accuracy of scientific research evaluation low, and due to the diversity of information sources and the complexity of data structures, it is difficult to extract accurate and effective information, and it will not be correct for the collected data. The correlation analysis of the data makes the data more scattered, which is not convenient for subsequent evaluation and use.

In order to achieve the above purpose, this disclosure provides the following technical solutions: a data collection system for evaluating scientific research capabilities based on subject development, including a data mining module, a data reporting module and the Internet, the output of the Internet and the input of the data mining module Terminals are electrically connected, the output terminal of the data mining module is electrically connected to the input terminal of the data preprocessing module, the output terminal of the data reporting module is electrically connected to the input terminal of the data preprocessing module, and the output terminal of the data preprocessing module end is electrically connected to the input end of the feature extraction module, and the output end of the feature extraction module is electrically connected to the input end of the investigation and correction module; the output end of the investigation and correction module is electrically connected to the input end of the cluster analysis module, and the The output end of the clustering analysis module is electrically connected to the input end of the association module, the output end of the association module is electrically connected to the input end of the quantization calculation module, and the output end of the quantization calculation module is electrically connected to the input end of the data dimension reduction module. connected, the output end of the data dimensionality reduction module is electrically connected to the input end of the database, and the output end of the database is electrically connected to the input end of the feature extraction module.

As a further solution of the present disclosure: the data reporting module includes a text entry module, a voice entry module and an image entry module, the output end of the text entry module, the output end of the speech entry module and the output end of the image entry module are connected with the data The input end of the preprocessing module is electrically connected.

As a further solution of the present disclosure, the data mining module is used to mine the following various related data: data related to scientific research capabilities, the scientific research capabilities include technological innovation capabilities, technological transformation capabilities, technological competitiveness and technological support capabilities, The scientific and technological innovation capability includes three elements of theoretical innovation, technological innovation and collaborative innovation, the scientific and technological transformation ability includes two elements of military benefit and economic benefit, and the scientific and technological competitiveness includes three elements of academic competition, talent competition and development potential , the technological support capability includes two elements of platform support and management support.

As a further solution of the present disclosure, the data mining module is used for: crawling data with the help of a web crawler tool, grabbing data on the network across screens with the help of the Scrapy grabbing framework, and grabbing structured data from the page at the same time, using the Python's Scrapy technical framework implements data crawling in the website; and then according to the needs of evaluating data, the captured data is analyzed by association algorithm through data mining.

As a further solution of the present disclosure, the data preprocessing module is configured to: convert the data captured by the data mining module into a data set for computer identification and calculation; perform on the data set: remove abnormal data , and check data spelling errors and remove duplicate data records, calculate missing data by derivation and fill incomplete record data, remove interference and noise in data through filtering technology and data cleaning, and strengthen useful information.

As a further solution of the present disclosure, the cluster analysis module is used to implement the following steps: S1. Randomly select the threshold of clustering, determine the category of each cluster through a random algorithm, and use the clustering algorithm to analyze the research-corrected Cluster the data to obtain clusters with categories: C={C ₁ ,C ₂ ,…,C _k }, calculate the degree of discrimination between any cluster in each feature and other clusters; S2, calculate its different The average value Mean _i of the discrimination between categories, calculate the maximum value Max _i and the minimum value Min _i of the average discrimination between different categories on each feature, and calculate the discrimination f _i of each feature on different categories: f _i ＝(Max _i -Min _i )/Mean _i , arrange the features in descending order according to the degree of discrimination f _i to obtain f _i ^* (i=1,2,...,m); S3, express the results obtained in S2 through a line graph , and find the sharply changing point or inflection point i ₀ in the line chart,

is the selected feature subset, and this feature subset together constitutes a feature cluster analysis document.

As a further solution of the present disclosure: the association steps and rules executed by the association module are: Let I={i ₁ ,i ₂ ,...i _m } be a collection of m different items, given a transaction database D, Each transaction T is a set of items in I, namely

T has a unique identifier TD, and association rules are of the form

the implication of

X∩Y=Φ, the condition for association rules to be established is the support degree S and the confidence degree C, in the support degree S, at least S% of transactions in D contain X∪Y, that is

In the confidence level C, among the transactions of X contained in D, at least C% of the transactions also contain Y, that is

The mining problem of association rules is to find the association rules in the transaction database D that satisfy the minimum support S _min and the minimum confidence C _min given by the user.

As a further solution of the present disclosure, the calculation formula of the quantization calculation module is:

Among them, A is the quantitative score of scientific research personnel performance evaluation, t _h and S _h are the number and ranking of scientific research personnel in scientific research activities _h that meet the quantitative indicators K _ij... The actual number of scientific research results of K _ij...x , where K _ij...x is the quantitative index value of a scientific researcher satisfying the quantitative index K _ij...x .

As a further solution of the present disclosure, the data dimension reduction module is used to: reduce the dimension of multi-dimensional data through OLAP, convert it into a report form and store it in the database; query in the database, and finally use the data as evaluation basis for scientific research Capability Assessment.

As a further solution of the present disclosure, the text input module is used to input data information by means of text input; the voice input module is used to input data information by voice input; the image input module The module is used to input data by means of image input, and can also perform text data; the Internet is used to search, obtain and share data information; the database is used to collect various data information in the system Management and classification and sorting, and at the same time realize the storage of data information; the feature extraction module: used to further screen and extract important features and characteristics of the mined data information.

The beneficial effects of the disclosure are:

In this disclosure, the clustering threshold is randomly selected, and the basic important features of the cluster-like data are determined through a random algorithm, and then the features are clustered through the clustering algorithm to obtain the feature data that has been classified, and then the in-depth calculation of the category features is performed. The average value is distinguished, and the data characteristics are obtained through dimensionality reduction. The line chart can be clearly expressed to obtain accurate evaluation data, and the minimum support and minimum credibility of the given satisfaction can be found through the establishment of confidence and support. And carry out the mining analysis of association rules, and then obtain the quantitative result index value through quantitative calculation, then through the above clustering processing, dimension reduction arrangement and data association analysis processing to ensure the extraction of accurate information, and then lay a solid foundation for subsequent evaluation, Improve the credibility and authenticity of the evaluation of scientific research results;

This disclosure provides a wide range of data information through the Internet, which is convenient for the search and acquisition of big data, so that the data mining module uses web crawler tools to obtain information, grab network data, and grab structured data from pages at the same time. Use the Python-based Scrapy technical framework to achieve data capture in the website, so that data information can be obtained in real time through the network, and the real-time performance of data association updates can be improved.

Description of drawings

FIG. 1 shows a data acquisition system for evaluating scientific research capabilities based on subject development according to one or more embodiments of the present disclosure;

Fig. 2 shows a schematic structural diagram of the data reporting module of the present disclosure.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present disclosure with reference to the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only some of the embodiments of the present disclosure, not all of them. Based on the embodiments in the present disclosure, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present disclosure.

As shown in Figure 1-2, this disclosure provides a data collection system for evaluating scientific research capabilities based on discipline development, including a data mining module, a data reporting module, and the Internet, and the output end of the Internet is connected to the input end of the data mining module. connection, the output end of the data mining module is electrically connected to the input end of the data preprocessing module, the output end of the data reporting module is electrically connected to the input end of the data preprocessing module, and the output end of the data preprocessing module is connected to the input end of the feature extraction module Electrically connected, the output end of the feature extraction module is electrically connected to the input end of the investigation and correction module.

The output end of the research and correction module is electrically connected to the input end of the cluster analysis module, the output end of the cluster analysis module is electrically connected to the input end of the association module, the output end of the association module is electrically connected to the input end of the quantitative calculation module, and the quantitative calculation The output end of the module is electrically connected to the input end of the data dimensionality reduction module, the output end of the data dimensionality reduction module is electrically connected to the input end of the database, and the output end of the database is electrically connected to the input end of the feature extraction module.

Data reporting module comprises text entry module, voice entry module and image entry module, text entry module, speech entry module and image entry module, the output end of described text entry module, the output end of described speech entry module and image entry module The output end is electrically connected with the input end of the data preprocessing module.

The text entry module is used to enter data information by means of text input. The voice input module is used for inputting data information by means of voice input. The image input module is used to input data by means of image input, and can also input text data. The Internet is used to search, acquire and share data information. The database is used to manage, classify and sort various data information in the system, and realize the storage of data information at the same time. The feature extraction module is used to further screen and extract important features and characteristics of the mined data information.

In data mining, the following kinds of relevant data are mined: data related to scientific research capabilities, scientific research capabilities include technological innovation capabilities, technological transformation capabilities, technological competitiveness and technological support capabilities, and technological innovation capabilities Theoretical innovation, technological innovation and collaborative innovation are three elements , S&T transfer capability includes two elements: military benefit and economic benefit; S&T competitiveness includes three elements: academic competition, talent competition, and development potential; S&T support capability includes two elements: platform support and management support.

The data mining module is used for: crawling data with the help of web crawler tools, fast and high-level cross-screen crawling of data on the network with the help of the Scrapy crawling framework, and crawling structured data from the page at the same time, using the Python-based Scrapy technical framework to achieve Carry out data capture in the website, and then according to the needs of evaluating data, carry out association algorithm analysis on the captured data through data mining. This disclosure provides a wide range of data information through the Internet, which is convenient for the search and acquisition of big data, so that the data mining module uses web crawler tools to obtain information, grab network data, and grab structured data from pages at the same time. Use the Python-based Scrapy technical framework to achieve data capture in the website, so that data information can be obtained in real time through the network, and the real-time performance of data association updates can be improved.

Data preprocessing is used to: convert the crawled data into a data set that can be recognized and operated by the computer, and perform on the data set: remove abnormal data, check data spelling errors, remove duplicate data records, and then calculate missing data by derivation And fill in the incomplete record data, remove the interference and noise in the data through filtering technology and data cleaning, and strengthen the useful information.

The clustering analysis module is used to implement the following steps S1-S3: S1, randomly select the clustering threshold, determine the category of each cluster through a random algorithm, and use the clustering algorithm to cluster the survey-corrected data to obtain Clustering with categories: C={C ₁ ,C ₂ ,...,C _k }, and then calculate the degree of discrimination between any one cluster and other clusters in each feature. S2, and calculate the average Mean _i of its discrimination between different categories, and then further calculate the maximum value Max _i and minimum value Min _i of the average discrimination between different categories on each feature, and then calculate each feature in Distinguishing degrees of different categories, that is, f _i =(Max _i -Min _i )/Mean _i , and then arrange the features in descending order of f _i to obtain f _i ^* (i=1,2,...,m). S3. Express the above-mentioned results through a line graph, and find a sharply changing point or inflection point i ₀ in the line graph,

The association steps and rules executed by the association module are: Let I={i ₁ ,i ₂ ,…i _m } be a set of m different items, given a transaction database D, each transaction T in it is a A collection of group items, i.e.

T has a unique identifier TD, and association rules are of the form

the implication of

The calculation formula of the quantitative calculation module is:

Among them, A is the quantitative score of scientific research personnel performance evaluation, t _h , S _h are the number and ranking of scientific research personnel in scientific research activities _h that meet the quantitative indicators K _ij... The actual number of scientific research results of K _ij...x , where K _ij...x is the quantitative index value of a scientific researcher satisfying the quantitative index K _ij...x .

The data dimension reduction module is used to reduce the dimension of multi-dimensional data through OLAP, convert it into a report form and store it in the database, which can be queried in the database, and finally use the data as the evaluation basis to evaluate scientific research capabilities.

This disclosure works by:

S1. First, enter the data information through the data reporting module. At the same time, you can also perform network capture and structured data capture operations on the information on the Internet through the data mining module. At this time, the data information obtained by mining will be passed to the data preprocessing Module, the data preprocessing module performs error screening on data information, deletes abnormal data, and obtains missing data through derivation and calculation, so that data information can be strengthened;

S2. Then, important basic features can be extracted through the feature extraction module, or similar basic features can be extracted from the database, and the credibility of the data features can be checked and checked through the research and correction module;

S3. Finally, the clustering analysis module randomly selects the clustering threshold, and obtains the data category through the clustering algorithm, distinguishes and obtains the dimensionality reduction ranking, and establishes support and confidence through the association module, and analyzes the integrated association regularity. At this time, the evaluation index value is calculated by the quantitative calculation module, and the accurate result evaluation value is obtained. Finally, the multidimensional data is reduced by the OLAP of the data dimension reduction module, and it is converted into a report form and stored in the database.

The beneficial effects of the disclosure are:

The last points that should be explained are: Although the present disclosure has been described in detail above with general descriptions and specific examples, on the basis of the present disclosure, the above embodiments are only used to illustrate the present disclosure. The technical solution of the present invention is not limited thereto; although the present disclosure has been described in detail with reference to the aforementioned embodiments, those of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, Or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present disclosure.

Claims

A data collection system for evaluating scientific research capabilities based on discipline development, including:

The data mining module, the data reporting module and the Internet are characterized in that: the output end of the Internet is electrically connected to the input end of the data mining module, and the output end of the data mining module is electrically connected to the input end of the data preprocessing module. The output end of the data reporting module is electrically connected to the input end of the data preprocessing module, the output end of the data preprocessing module is electrically connected to the input end of the feature extraction module, and the output end of the feature extraction module is connected to the research correction module. The input terminal is electrically connected;

The output end of the research and correction module is electrically connected to the input end of the cluster analysis module, the output end of the cluster analysis module is electrically connected to the input end of the association module, and the output end of the association module is electrically connected to the input end of the quantitative calculation module Terminals are electrically connected, the output terminal of the quantitative calculation module is electrically connected to the input terminal of the data dimensionality reduction module, the output terminal of the data dimensionality reduction module is electrically connected to the input terminal of the database, and the output terminal of the database is electrically connected to the feature extraction module The input terminals are electrically connected.
The data collection system for evaluating scientific research capabilities based on subject development according to claim 1, wherein the data reporting module includes:

A text entry module, a voice entry module and an image entry module, the output of the text entry module, the output of the speech entry module and the output of the image entry module are electrically connected to the input of the data preprocessing module .
The data collection system for evaluating scientific research capabilities based on subject development according to claim 1, wherein the data mining module is used to mine the following multiple related data:

Data related to scientific research capabilities. The scientific research capabilities include technological innovation capabilities, technological transformation capabilities, technological competitiveness, and technological support capabilities. The technological innovation capabilities include three elements: theoretical innovation, technological innovation, and collaborative innovation. The technological transformation Capabilities include two elements: military benefit and economic benefit. The technological competitiveness includes three elements: academic competition, talent competition and development potential. The technological support capability includes two elements: platform support and management support.
The data collection system for evaluating scientific research capabilities based on subject development according to claim 3, wherein the data mining module is used for:

Crawl data with the help of web crawler tools, use the Scrapy crawling framework to crawl data on the network across screens, and grab structured data from pages at the same time, and use the Python-based Scrapy technical framework to achieve data crawling in the website;

According to the needs of the evaluation data, the association algorithm analysis is carried out on the captured data through data mining.
The data collection system for evaluating scientific research capabilities based on subject development according to claim 4, wherein the data preprocessing module is used for:

converting the data captured by the data mining module into a data set for computer identification and calculation;

For the data set: eliminate abnormal data, check data spelling errors and remove duplicate data records, calculate missing data through derivation and fill incomplete record data, and remove interference and noise in data through filtering technology and data cleaning , and strengthen the useful information.
The data collection system for evaluating scientific research capabilities based on subject development according to claim 1, wherein the cluster analysis module is used to implement the following steps:

S1. Randomly select the clustering threshold, determine the category of each cluster through a random algorithm, and use the clustering algorithm to cluster the data after the survey correction to obtain a cluster with categories: C={C 1 ,C 2 ,...,C k }, calculate the discrimination between any cluster and other clusters in each feature;

S2. Calculate the average Mean i of its discrimination between different categories, calculate the maximum value Max i and minimum value Min i of the average discrimination between different categories on each feature, and calculate the mean value of each feature on different categories Distinguishing degree f i : f i =(Max i -Min i )/Mean i , arrange features in descending order according to distinguishing degree f i to obtain f i * (i=1,2,...,m);

S3. Express the result obtained in S2 through a line graph, and find a sharply changing point or inflection point i 0 in the line graph,
is the selected feature subset, and this feature subset together constitutes a feature cluster analysis document.
The data acquisition system for scientific research capability assessment based on subject development according to claim 1, wherein the association steps and rules executed by the association module are:

Let I={i 1 ,i 2 ,…i m } be a set of m different items, given a transaction database D, each transaction T is a set of items in I, that is
T has a unique identifier TD, and association rules are of the form
the implication of
X∩Y=Φ, the condition for association rules to be established is the support degree S and the confidence degree C, in the support degree S, at least S% of transactions in D contain X∪Y, that is
In the confidence level C, among the transactions of X contained in D, at least C% of the transactions also contain Y, that is
The mining problem of association rules is to find the association rules in the transaction database D that satisfy the minimum support S min and the minimum confidence C min given by the user.
The data collection system for evaluating scientific research capabilities based on subject development according to claim 1, wherein the calculation formula of the quantitative calculation module is:

Among them, A is the quantitative score of scientific research personnel performance evaluation, t h and S h are the number and ranking of scientific research personnel in scientific research activities h that meet the quantitative indicators K ij... The actual number of scientific research results of K ij...x , where K ij...x is the quantitative index value of a scientific researcher satisfying the quantitative index K ij...x .
The data collection system for evaluating scientific research capabilities based on subject development according to claim 1, wherein the data dimensionality reduction module is used for:

Reduce the dimensionality of multidimensional data through OLAP, convert it into a report form and store it in the database;

Inquire in the database, and finally use the data as the evaluation basis to evaluate scientific research capabilities.
The data collection system for evaluating scientific research capabilities based on subject development according to claim 2, wherein:

The text input module is used to input data information by means of text input;

The voice input module is used to input data information by means of voice input;

The image input module is used to input data by means of image input, and can also perform text data;

The Internet is used for searching, obtaining and sharing data information;

The database is used for the management and classification of various data information in the system, and at the same time realizes the storage of data information;

The feature extraction module is used for screening and extracting important features and characteristics of the mined data information.