CN114328461A - Big data analysis-based enterprise innovation and growth capacity evaluation method and system - Google Patents

Abstract

The method and system for evaluating the innovation and growth capability of an enterprise based on big data analysis provided by the present invention relate to the field of quantitative evaluation. ;Train the index weight model with the sample data after cleaning and classification, so as to obtain the relative importance score of the index item and the priority map to calculate the final weight value of each index item; Combine the final weight of each index item according to the enterprise data of each enterprise to be evaluated The final evaluation score of each enterprise to be evaluated is calculated separately; among them, the enterprise dimension includes industry and commerce, employees, qualifications, intellectual property, projects, financing loans, assets and R&D expenses. The invention accurately and reliably measures the growth ability of the enterprise by performing accurate and reliable quantitative evaluation on the intangible assets of the scientific and technological enterprise, including innovation ability and scientific and technological projects, and calculates the proportion weight, and provides data support for the future development planning and decision-making of the enterprise.

Description

An evaluation method and system for enterprise innovation and growth ability based on big data analysis

technical field

The invention relates to the field of quantitative evaluation, in particular to a method and system for evaluating the innovation and growth capability of an enterprise based on big data analysis.

Background technique

The traditional evaluation method of enterprise innovation and growth ability based on big data analysis is mainly financial analysis method, which analyzes the financial data, business model and business plan of the enterprise, such as assets and liabilities, sales profit, etc., this type of enterprise grows. The capability evaluation method can only be evaluated or updated when the company raises funds, audits or releases financial reports, and is generally aimed at larger-scale companies or companies that have been established for a long time; for entrepreneurial technology companies that have been established for a short time, their characteristics are: With a short operating time, less financial data, and a large proportion of intangible assets, it is not applicable for such enterprises to use the financial analysis method to evaluate the growth ability of the enterprise.

Another enterprise evaluation technology is enterprise portrait, but enterprise portrait only displays the existing data of the enterprise from different dimensions, and does not conduct in-depth analysis of it, nor does it quantitatively evaluate the business behavior of the enterprise.

Due to the inability to accurately and reliably assess the growth capability of technology-based SMEs with a large proportion of intangible assets, it will be very difficult for such enterprises to obtain financing loans and policy support, which is not conducive to the development and growth of enterprises and technological innovation.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide an evaluation method and system for enterprise innovation and growth ability based on big data analysis. The evaluation of the dimensions is strengthened, and more suitable data support and investment projects can be obtained according to this evaluation structure during policy implementation or enterprise investment and financing.

In order to achieve the above-mentioned purpose, the present invention proposes the following technical solution: a method for evaluating the innovation and growth capability of an enterprise based on big data analysis, including:

Collect the sample data of each enterprise dimension of the sample enterprise, and perform data cleaning, calculation and label classification on the sample data according to the time series; among which, the enterprise dimension includes industry and commerce, employees, qualifications, intellectual property rights, projects, financing loans, assets and R&D expenses;

Using the cleaned sample data and label classification of the sample companies, the index weight model is trained, so as to obtain the relative importance score and priority map of the index items according to the index weight model, and then calculate the final weight value of each index item;

According to the enterprise data of each enterprise to be evaluated and the final weight value of each index item, the final evaluation score and ranking of each index item of each to-be-evaluated enterprise are calculated respectively; wherein, the enterprise data is a sample of each enterprise dimension after data cleaning of the to-be-evaluated enterprise data.

Further, the process of performing data cleaning, calculation and label classification on the sample data according to the time series includes:

Obtain the industrial and commercial information of the sample companies, as well as the number of researchers and the proportion of educational backgrounds in each of the past three years, qualification certification information, the number of Class I intellectual property rights, the number of Class II intellectual property rights, project application information, loan amount, and financing data , R&D expenses and technical contract information, where the loan amount includes the maximum loan amount, the minimum loan amount, the average loan amount and the total loan amount;

Clean and normalize the content of the obtained data above, replace the Chinese content with digital representation, and obtain numerical enterprise data;

Label and classify the numerical enterprise data to determine each index item, put the data of each index item into the object of the enterprise data matrix, and calculate the data value of the sample enterprise under each index item respectively;

Calculate the sum, average, maximum value, minimum value, 25% value and 75% value of the data values of all sample companies under each indicator, and deal with abnormal data values.

Further, the calculation process of the final weight value of the index item is:

Predefine a multi-level data classification label as the calculation index item;

Receive all the sample data of the sample enterprise, use the gradient boosting decision tree algorithm of the machine learning xgboost library to extract the features of each sample data and classify, sort and group, and combine the input calculation index items to calculate the softmax loss function plus the regular term, and obtain each sample data one by one. The relative importance score and ranking diagram of each index item;

The average value of the relative importance score of the index item and the proportion score of the preset corresponding index item is taken as the final weight value of the index item.

Further, the calculation process of the final evaluation score of each index item of each enterprise to be evaluated is as follows:

Obtain the sample data of each enterprise dimension of each to-be-evaluated enterprise and process the data to obtain the enterprise data matrix of each to-be-evaluated enterprise;

Each index item calculated by the index weight model is divided into five groups of evaluation dimensions, including endogenous innovation, collaborative investment, management and operation ability, achievement transformation ability and sustainable growth ability. Perform matrix dot product on the corresponding value in and its corresponding final weight value to obtain the evaluation score of the corresponding index item;

According to the characteristics of normal distribution, the scores of each group of evaluation dimensions are converted into a 100-point system as the final evaluation score of the enterprise index item.

Further, the indicator items of the sample companies include the total number of employees, the proportion of researchers, the proportion of college education, the proportion of postgraduates, the number of enterprise qualifications, the qualifications of production, education and research, the number of subordinate institutions, the number of research and development activities, and the proportion of research and development expenses. , the company's research and development expenses in the previous year, last year's research and development expenses, the previous year's project establishment cost, the last year's project establishment cost, the maximum project amount, the minimum project amount, the average project amount, the total project amount, the total number of approved projects, the total number of applications for projects, and equity investment Total amount, total number of equity investments, total amount of claims in the previous year, total amount of claims in the previous year, maximum value of claims, minimum value of claims, average value of claims, total amount of claims, total number of claims obtained, total assets in the previous year, total assets in the previous year, net in the previous year Assets, last year's net assets, last year's main business income, last year's main business income, last year's cost, last year's cost, last year's net profit, last year's net profit, previous year's profits and taxes, last year's profits and taxes, the proportion of the previous year's main business, last year Proportion of main business, net profit margin of the previous year, net profit margin of the previous year, management certification, intellectual property rights of type I, intellectual property rights of type II, the number of technical contracts in the previous year, the number of technical contracts in the previous year, the largest technical transaction volume of technical contracts, and the smallest technical contract technology Transaction value, average technical transaction value of technical contracts, total technical transaction value of technical contracts, total number of technical transactions of technical contracts, number of national science and technology awards, number of established standards, number of transformation of scientific and technological achievements, number of high-tech products, high-tech income, previous year Growth rate of total assets, growth rate of net assets, growth rate of main business income, rate of change of costs and expenses, growth rate of net profit, growth rate of profits and taxes paid, growth rate of R&D expenses, rate of change of main business proportion, growth rate of net profit rate , the growth rate of the proportion of R&D expenses, the growth rate of the number of intellectual property rights, and the growth rate of the proportion of R&D personnel.

Further, the processing method of the abnormal data value is as follows: for the data value whose proportion data is not in the range of 0 to 1, when it is greater than 1, it is set as 75% of the value of the corresponding item, and when it is less than 1, it is set as 25% of the corresponding item. % value; for data with a growth rate greater than 20, it is set to the 75% value of the corresponding item, and data with a growth rate less than -20 is set to the 25% value of the corresponding item.

Another solution of the present invention is to disclose an evaluation system for enterprise innovation and growth capability based on big data analysis, the system comprising:

The enterprise data collection and processing module is used to collect sample data of each enterprise dimension of the sample enterprise, and perform data cleaning, calculation and label classification of the sample data according to the time series; among which, the enterprise dimension includes business, employee, qualification, intellectual property, project, financing Loans, assets and research and development expenses;

The indicator weight module is used to train the indicator weight model according to the cleaned sample data and label classification of the sample enterprises, so as to obtain the relative importance score and priority map of the indicator items according to the indicator weight model, and then calculate the final weight value of each indicator item. ;

The score assigning module is used to calculate the final evaluation score and ranking of each index item of each to-be-evaluated enterprise according to the enterprise data of each to-be-evaluated enterprise and the final weight value of each index item.

Further, the enterprise data collection and processing module includes:

The first acquisition unit is used to obtain the industrial and commercial information of the sample enterprises, as well as the number of researchers and the proportion of educational backgrounds in each of the past three years, information on qualification certification, the number of Class I intellectual property rights, the number of Class II intellectual property rights, and project declarations Information, loan amount, financing data, research and development expenses and technical contract information, wherein the loan amount includes the maximum loan amount, the minimum loan amount, the average loan amount and the total loan amount;

The cleaning and normalization unit is used to clean and normalize the data acquired by the acquisition unit, including deleting duplicate data, evaluating irrelevant data, and replacing Chinese content with digital representation to obtain numerical enterprise data;

The classification unit is used to label and classify the numerical enterprise data to determine each index item, put the data of each index item into the object of the enterprise data matrix, and calculate the data value of the sample enterprise under each index item respectively;

The calculation unit is used to calculate the sum, average, maximum value, minimum value, 25% value and 75% value of data values of all sample enterprises under each index item, and process abnormal data values.

Further, the indicator weight module includes:

The predefined unit is used to predefine a multi-level data classification label as the calculation index item;

The machine learning unit is used to receive all the sample data of the sample enterprise, use the gradient boosting decision tree algorithm of the machine learning xgboost library to extract the features of each sample data and classify, sort and group, and combine the input calculation index items to perform softmax loss function plus regularization Item calculation, obtain the relative importance score and ranking diagram of each index item one by one;

The weight calculation unit is configured to calculate the mean value of the relative importance score of the index item and the proportion score of the preset corresponding index item, and use the mean value as the final weight value of the index item.

Further, the score assignment module includes:

The second obtaining unit is used to obtain sample data of each enterprise dimension of each enterprise to be evaluated and perform data processing to obtain the enterprise data matrix of each enterprise to be evaluated;

The evaluation unit is divided, and each index item used for the calculation of the index weight model is divided into five groups of evaluation dimensions, including endogenous innovation, collaborative investment, management and operation ability, achievement transformation ability and sustainable growth ability. The corresponding value of the index item in the enterprise data matrix is multiplied by the matrix dot with its corresponding final weight value to obtain the evaluation score of the corresponding index item;

The score conversion unit is used to convert the scores of each group of evaluation dimensions into a 100-point system according to the normal distribution characteristics as the final evaluation score of the enterprise index item.

As can be seen from the above technical solutions, the technical solutions of the present invention have obtained the following beneficial effects:

The method and system for evaluating the innovation and growth ability of an enterprise based on big data analysis disclosed in the present invention not only considers the financial indicators of the enterprise, but also conducts accurate quantitative evaluation on the intangible assets of the enterprise, such as innovation ability and scientific and technological projects. Measure the ability of a business to grow. Specifically, the method includes: collecting sample data of each enterprise dimension of the sample enterprise, and performing data cleaning, calculation and label classification on the sample data according to the time series, wherein the enterprise dimension includes industry and commerce, employees, qualifications, intellectual property rights, projects, financing loans , assets and R&D expenses; use the sample data and label classification after cleaning from the sample companies to train the index weight model, so as to obtain the relative importance score and priority map of the index items according to the index weight model, and then calculate the final weight value of each index item ;According to the enterprise data of each enterprise to be evaluated and the final weight value of each index item, the final evaluation score of each index item of each to-be-evaluated enterprise is calculated respectively; the final evaluation score is based on five aspects, namely endogenous innovation, collaborative investment, management and operation Ability, achievement transformation ability and continuous growth ability.

The final evaluation score of the enterprise obtained by the present invention is more suitable for small, medium and micro enterprises to evaluate the growth ability of the enterprise compared with the financial analysis method that focuses on financial index items, and provides data support for the future development planning, policy adaptation and decision-making of the enterprise; For enterprise portraits, it can not only display the development data of the enterprise from different dimensions, but also analyze the impact of the enterprise's business behavior and innovation ability on the enterprise's growth ability more deeply.

It is to be understood that all combinations of the foregoing concepts, as well as additional concepts described in greater detail below, are considered to be part of the inventive subject matter of the present disclosure to the extent that such concepts are not contradictory.

The foregoing and other aspects, embodiments and features of the present teachings can be more fully understood from the following description when taken in conjunction with the accompanying drawings. Other additional aspects of the invention, such as features and/or benefits of the exemplary embodiments, will be apparent from the description below, or learned by practice of specific embodiments in accordance with the teachings of this invention.

Description of drawings

The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures may be represented by the same reference numeral. For clarity, not every component is labeled in every figure. Embodiments of various aspects of the present invention will now be described by way of example and with reference to the accompanying drawings, wherein:

Fig. 1 is the flow chart of the evaluation method of enterprise innovation and growth ability based on big data analysis of the present invention;

Fig. 2 is the flow chart of sample data cleaning preparation in the embodiment;

Fig. 3 is the sample data cleaning flow chart 1 in the embodiment;

4 is a second flow chart of sample data cleaning in the embodiment;

FIG. 5 shows the weight of each evaluation dimension for generating the evaluation index in the embodiment.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, embodiments of the present invention. Based on the described embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention. Unless otherwise defined, technical or scientific terms used herein should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention belongs.

The terms "first", "second" and similar terms used in the description of the patent application and the claims of the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Also, unless the context clearly dictates otherwise, the singular forms "a," "an," or "the" and the like do not denote a limitation of quantity, but rather denote the presence of at least one. Words like "including" or "comprising" mean that the elements or items appearing before "including" or "including" cover the features, integers, steps, operations, elements and/or recited after "including" or "including" or components, does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or sets thereof.

Based on the existing technology, the evaluation method of enterprise growth ability is mainly financial analysis method, which focuses on the financial indicators of the enterprise. As a result, the enterprise growth ability evaluated by this method can only be used in enterprise financing, auditing or issuing financial reports, and can only be applied to Large-scale or long-established enterprises are not applicable to small and medium-sized micro-enterprises and entrepreneurial technology enterprises that have been established for a short time, because the innovation ability and achievement transformation ability of such enterprises in their business activities are included in this evaluation method. Quantitative assessments were not performed. Therefore, the present invention aims to propose a method and system for evaluating the innovation and growth capability of an enterprise based on big data analysis, which, in addition to common indicators such as financial indicators, also quantitatively evaluates the non-quantifiable innovation capability and achievement transformation capability of the intangible assets of the enterprise. , to accurately and reliably evaluate the growth ability of the enterprise, and provide data support for the future development and decision-making of the enterprise.

The method and system for evaluating the innovation and growth capability of an enterprise based on big data analysis disclosed in the present invention will be further described in detail below with reference to the embodiments shown in the accompanying drawings.

With reference to the flowchart shown in FIG. 1 , the method for evaluating the innovation and growth capability of an enterprise based on big data analysis disclosed in the embodiment of the present application includes the following steps during execution:

Step S102, collect the sample data of each enterprise dimension of the sample enterprise, and perform data cleaning, calculation and label classification on the sample data according to the time series; wherein, the enterprise dimension includes industry and commerce, employees, qualifications, intellectual property, projects, financing loans, assets and research and development cost;

The specific execution process of step S102 is as follows: first, obtain the industrial and commercial information of the sample enterprise, as well as the number of researchers and the proportion of educational backgrounds, qualification certification information, the number of Type I intellectual property rights, and the number of Type II intellectual property rights in each of the past three years. , project application information, loan amount, financing data, R&D expenses, and technical contract information, where the loan amount includes the maximum loan amount, the minimum loan amount, the average loan amount, and the total loan amount; then, the content of the obtained data is cleaned and normalized. , replace the Chinese content with digital representation, and obtain numerical enterprise data; secondly, label and classify the numerical enterprise data to determine each index item, put the data of each index item into the object of the enterprise data matrix, and separately calculate the sample enterprises in Data values under each indicator; finally, calculate the sum, average, maximum, minimum, 25% and 75% of the data values of all sample companies under each indicator, and deal with abnormal data values.

The indicators of the above sample companies include the total number of employees, the proportion of researchers, the proportion of college graduates, the proportion of graduate students, the number of enterprise qualifications, the qualifications of production, education and research, the number of subordinate institutions, the number of research and development activities, the proportion of research and development expenses, the year before the company. R&D expenses, last year's R&D expenses, last year's project establishment expenses, last year's project establishment expenses, maximum project amount, minimum project amount, average project amount, total project amount, total number of approved projects, total number of applications for projects, total equity investment, Total number of equity investments, total amount of claims in the previous year, total amount of claims last year, maximum value of claims, minimum value of claims, average value of claims, total amount of claims, total number of claims obtained, total assets in the previous year, total assets in the previous year, net assets in the previous year, last year Net assets, main business income in the previous year, main business income in the previous year, cost in the previous year, cost in the previous year, net profit in the previous year, net profit in the previous year, profits and taxes paid in the previous year, profits and taxes paid in the previous year, the proportion of the main business in the previous year, and the main business in the previous year Proportion, net profit margin of the previous year, net profit margin of the previous year, management certification, intellectual property rights of type I, intellectual property rights of type II, the number of technical contracts in the previous year, the number of technical contracts in the last year, the largest technical transaction value of technical contracts, the smallest technical transaction value of technical contracts, Average technical transaction value of technical contracts, total technical transaction value of technical contracts, total number of technical transactions of technical contracts, number of national science and technology awards, number of established standards, number of transformation of scientific and technological achievements, number of high-tech products, high-tech income, total asset growth in the previous year rate, growth rate of net assets, growth rate of main business income, rate of change of cost and expense, growth rate of net profit, growth rate of profits and taxes paid, growth rate of research and development expenses, rate of change of main business proportion, rate of change of net profit rate, research and development expenses Proportion growth rate, intellectual property quantity growth rate, and R&D personnel ratio growth rate.

The process of acquiring, cleaning, calculating and classifying the sample data of each enterprise dimension of the sample enterprise includes at least three stages as shown in Figures 2 to 4.

As shown in Figure 2, the data preparation stage is entered in the early stage, the operating income data is processed, and the relevant data required by all sample companies to evaluate the index model is derived from the database. "Enterprise.xls" file is stored in memory for backup; obtain a collection of all unique sample company names from the "All-enterprise.xls" file, and for any sample company in the collection: first obtain the company's latest science and technology loan The data is used as the data basis for the enterprise in this evaluation method; then calculate the 1) maximum, minimum, average and total amount of science and technology innovation loans obtained by the enterprise in the past three years, 2) the average proportion of researchers in each year, 3) The average number of Class I intellectual property rights in each year, 4) The average number of Class II intellectual property rights in each year, forming the data object CA; the latest piece of science and technology loan data for any sample enterprise merger in the collection and its corresponding The data object CA constitutes the new enterprise data and is transferred to the initial table "date1.xls" file in the first step.

Normalize the data content in the initial table of the first step, for example, remove the Chinese in the columns of the proportion of researchers, the proportion of college degree or above, and the total number of employees in the table, and convert them into float type data; Whether it is a high-tech enterprise, a technologically advanced enterprise, or a technology-based small and medium-sized enterprise will be converted to 0/1; The enterprise information and the corresponding enterprise data of the enterprise are merged into the first step result table "sheet1.xlsx" file.

As shown in Figure 3, in the mid-term, the label classification and quantitative evaluation of the achievement conversion ability of the sample enterprises includes obtaining five parts of data; one is the result table "sheet1.xlsx" file of the first step; the second is, from the first step The enterprise data obtained from the initial form "date1.xls" file about certain qualification certification information and the intellectual property rights and researcher data of the enterprise; The certification type is converted to a type column. If the value of the column certified by the company is 1; the third is to obtain the equity financing data of all companies from the database, and use this data with all relevant data in the above "All companies.xls" document. Take the intersection to obtain the first intersection data, and calculate the investment times and total investment amount obtained by the enterprise for any sample enterprise according to the first intersection data; the fourth is to obtain the technical contract data of all enterprises from the database, and combine the data with Take the intersection of all relevant data in the above "All Enterprises.xls" document to obtain the second intersection data, and calculate the technology transaction volume of the enterprise in each of the past three years and the largest enterprise technology transaction volume for any sample enterprise in turn according to the second intersection data. , minimum, average, total data and times; the fifth is to obtain the project declaration data of all enterprises from the database, and intersect the data with all relevant data of the above "All enterprises.xls" document to obtain the third intersection data, According to the third intersection data, for any sample enterprise, calculate the project approval amount of the enterprise in each year of the past three years and the maximum, minimum, average, total data and times of the enterprise project approval amount.

Finally, combine the above five parts of data according to the above enterprise dimensions, and output the result table "date99.xlsx" file of the second step.

As shown in Figure 4, the data values and abnormal data values of the sample enterprises under each indicator are calculated in the later stage, and the data in the early and mid-term are combined in this stage; that is, the first step result table "date1.xls" file The operating income in the second step is combined with the "date99.xlsx" file of the second step result table, and the total amount of subsidy and subsidies obtained by each sample enterprise (industry-university-research), the number of qualification certifications (enterprise qualification) of each sample enterprise, and the The number of management certifications (management certification) obtained by each enterprise, the growth rate of assets, profits, income, tax payment and R&D expenses of the sample enterprises, the proportion of talents, expenses, main business income, zero profits, etc. of the sample enterprises in the past/previous year , the calculation result data is merged with the data in the second step result table "date99.xlsx" file, and recorded as the "bb_date" table object of the indicator item data for all sample companies; For a column of data, calculate the count, mean, std, min, 25%, 50%, 75%, max, null_% of the column data (the ratio of the number of columns and rows with data to the total number of rows), and the calculated value is recorded as " n_df" table object.

Finally, the abnormal data values in the table object of the indicator item data "bb_date" are processed. The processing method of the abnormal data values proposed in the embodiment is: for the data values whose proportion data is not in the range of 0 to 1, the setting when it is greater than 1 It is the 75% value of the corresponding item, and if it is less than 1, it is set to the 25% value of the corresponding item; for the data whose growth rate is greater than 20, it is set to the 75% value of the corresponding item, and the data whose growth rate is less than -20 is set to the 25% value of the corresponding item. % value; set the value of some label columns that exceed the maximum value to the corresponding maximum value, including the total number of approved projects, the number of subordinate institutions, the total number of application projects, the maximum amount of claims, the average value of claims, and the total amount of claims ; After the abnormal data value is processed, store the processed data in the third step result data "data_72.xlsx" file.

Step S104, using the sample data cleaned by the sample enterprise and the label classification, and training the index weight model, so as to obtain the relative importance score and the priority map of the index items according to the index weight model, and then calculate the final weight value of each index item;

The specific execution process of step S104 is: predefining a multi-level data classification label as a calculation index item; receiving all the sample data of the sample enterprise, using the gradient boosting decision tree algorithm of the machine learning xgboost library to extract the characteristics of each sample data and carry out the process. Categorize and sort into groups, perform softmax loss function plus regular term calculation in combination with the input calculation index items, and obtain the relative importance score and priority map of each index item one by one; take the relative importance score of the index item and the preset corresponding index item to account for The average of the ratio scores and the sum is used as the final weight value of this indicator item.

FIG. 5 shows the calculation process of calculating the final weight value of the index item by taking the total amount of the project amount as the calculation index item. Specifically, first, copy the column of total item amount in the result data "data_72.xlsx" of the third step as column Y, and preprocess the data in column Y, including excluding all rows with null values in column Y, Retain rows with at least 50 non-null columns; secondly, use the gradient boosting decision tree algorithm of the xgboost library to train the weight model, the feature matrix is data other than column Y, and the label is column Y; then, get the importance of the index item Score the result, and bind the result to the column name of the feature matrix source, that is, the binding result is the column name of the third step result data "data_72.xlsx", and the value is the importance score; then in the binding result On the basis of , add the feature matrix source column index number to form numbered column 1, and arrange the numbered column 1 in ascending order according to the importance score; add the feature matrix source row index number to the numbered column 1 to form the numbered column 2, and according to the numbered column Sort by 1, get the column name and value of the numbered column 2.

Finally, merge the data in column No. 2 and the column data of "Comprehensive Weight" and "Index" read from the preset weight allocation table "zonghe.xlsx" for all indicators, and calculate the final weight value of the indicator item. The calculation method is: (No. The value in column 2 + comprehensive weight)/2, calculate the index items of the enterprise dimension in turn, and save all the results as the weight data "qz.xlsx" file.

Step S106, calculate the final evaluation score and ranking of each index item of each to-be-evaluated enterprise according to the enterprise data of each to-be-evaluated enterprise and the final weight value of each index item; wherein, the enterprise data is each enterprise after data cleaning of the to-be-evaluated enterprise dimension of sample data.

The specific execution process of step S108 is: obtaining sample data of each enterprise dimension of each enterprise to be evaluated and performing data processing to obtain an enterprise data matrix of each to-be-evaluated enterprise; dividing each index item calculated by the index weight model into five groups of evaluation dimensions, Including endogenous innovation, collaborative investment, management and operation ability, achievement transformation ability and sustainable growth ability, the corresponding value of any index item in each group of evaluation dimensions in the enterprise data matrix and the corresponding final weight value are matrix dot multiplied, Obtain the evaluation scores of the corresponding index items; according to the normal distribution characteristics, convert the scores of each group of evaluation dimensions into a 100-point system as the final evaluation score of the enterprise index items.

The embodiment of the present application further provides an electronic device, the device includes a processor and a memory, the memory is used to store program instructions, and transmit the program instructions to the processor; when the program instructions are executed by the processor, the processor Run the method for evaluating the innovation and growth capability of an enterprise based on big data analysis in the above embodiment.

The above program can be executed in a processor, or can also be stored in a computer-readable storage medium. The computer-readable storage medium includes permanent and non-permanent, removable and non-removable media. Information can be realized by any method or technology. storage. Information may be computer readable instructions, data structures, modules of programs, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), Flash Memory or other memory technology, Compact Disc Read Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage , magnetic tape cartridges, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable storage media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

These computer programs can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby executing instructions on the computer or other programmable device Steps for implementing the functions specified in one or more flows of the flowchart and/or one or more blocks of the block diagram are provided, and corresponding and different steps may be implemented by different modules.

Exemplarily, this embodiment provides such a device or system, that is, an evaluation system for enterprise innovation and growth capabilities based on big data analysis. The system includes the following program modules: an enterprise data collection and processing module for collecting data. The sample data of each enterprise dimension of the sample enterprise is used for data cleaning, calculation and label classification according to the time series; among which, the enterprise dimension includes industry and commerce, employees, qualifications, intellectual property rights, projects, financing loans, assets and R&D expenses; indicator weights The module is used to train the index weight model according to the sample data and label classification after cleaning of the sample enterprise, so as to obtain the relative importance score and priority map of the index items according to the index weight model, and then calculate the final weight value of each index item; score; The endowment module is used to calculate the final evaluation score and ranking of each index item of each to-be-evaluated enterprise according to the enterprise data of each to-be-evaluated enterprise and the final weight value of each index item.

The system is used to implement the functions of the evaluation method in the above-mentioned embodiment, and each module in the system corresponds to each step in the method, which has been described in the method, and will not be repeated here.

For example, the enterprise data collection and processing module includes: the first acquisition unit, which is used to obtain the industrial and commercial information of the sample enterprises, as well as the number of researchers and the proportion of educational backgrounds in each of the past three years, qualification certification information, the number of I intellectual property rights, The number of Class II intellectual property rights, project application information, loan amount, financing data, R&D expenses and technical contract information, where the loan amount includes the maximum loan amount, the minimum loan amount, the average loan amount and the total loan amount; cleaning and normalization unit , used to clean and normalize the data acquired by the acquisition unit, including deleting duplicate data, evaluating irrelevant data, and replacing Chinese content with digital representation to obtain numerical enterprise data; classification unit, used to label numerical enterprise data Determine each index item by classification, put the data of each index item into the object of the enterprise data matrix, and calculate the data value of the sample enterprise under each index item respectively; the calculation unit is used to calculate the data of all sample enterprises under each index item Sum, mean, maximum, minimum, 25%, and 75% of values, and handle outlier data values.

For another example, the indicator weight module includes: a pre-defined unit for pre-defining a multi-level data classification label as a calculation indicator item; a machine learning unit for receiving all sample data of the sample enterprise, using the gradient boosting of the machine learning xgboost library The decision tree algorithm extracts the features of each sample data and performs classification, sorting and grouping. Combined with the input calculation index items, the softmax loss function plus regular term calculation is performed, and the relative importance score and priority map of each index item are obtained one by one; weight calculation unit , which is used to calculate the mean value of the relative importance score of the index item and the proportion score of the preset corresponding index item, and use the mean value as the final weight value of the index item.

And for example, the score giving module includes: a second obtaining unit for obtaining sample data of each enterprise dimension of each to-be-evaluated enterprise and performing data processing to obtain an enterprise data matrix of each to-be-evaluated enterprise; dividing the evaluation unit for dividing indicators Each index item calculated by the weighting model is divided into five groups of evaluation dimensions, including endogenous innovation, collaborative investment, management and operation ability, achievement transformation ability and sustainable growth ability. Perform matrix dot product with the corresponding final weight value to obtain the evaluation score of the corresponding index item; the score conversion unit is used to convert the scores of each group of evaluation dimensions into a 100-point system according to the normal distribution characteristics as the enterprise index The final evaluation score for the item.

The embodiment of the present application solves the problem that the entrepreneurial technology enterprise with a relatively short establishment time cannot use the financial analysis method to evaluate the enterprise growth ability, and quantitatively evaluates the unquantifiable innovation ability and achievement transformation ability in the business activities of the enterprise. , avoid focusing on the financial indicators of enterprises, and apply to the growth ability assessment of large, medium, small and micro enterprises, and provide more accurate and reliable data support for enterprise development planning, policy matching and enterprise investment and financing decisions.

Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Those skilled in the art to which the present invention pertains can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be determined according to the claims.

Claims (10)

1. An evaluation method for enterprise innovation and growth capability based on big data analysis, characterized in that it comprises: Collect the sample data of each enterprise dimension of the sample enterprise, and perform data cleaning, calculation and label classification on the sample data according to the time series; among which, the enterprise dimension includes industry and commerce, employees, qualifications, intellectual property rights, projects, financing loans, assets and R&D expenses; Using the cleaned sample data and label classification of the sample companies, the index weight model is trained, so as to obtain the relative importance score and priority map of the index items according to the index weight model, and then calculate the final weight value of each index item; According to the enterprise data of each enterprise to be evaluated and the final weight value of each index item, the final evaluation score and ranking of each index item of each to-be-evaluated enterprise are calculated respectively; wherein, the enterprise data is a sample of each enterprise dimension after data cleaning of the to-be-evaluated enterprise data. 2. The evaluation method of enterprise innovation and growth capability based on big data analysis according to claim 1, wherein the process of performing data cleaning, calculation and label classification on the sample data according to time series comprises: Obtain the industrial and commercial information of the sample companies, as well as the number of researchers and the proportion of educational backgrounds in each of the past three years, qualification certification information, the number of Class I intellectual property rights, the number of Class II intellectual property rights, project application information, loan amount, and financing data , R&D expenses and technical contract information, where the loan amount includes the maximum loan amount, the minimum loan amount, the average loan amount and the total loan amount; Clean and normalize the content of the obtained data above, replace the Chinese content with digital representation, and obtain numerical enterprise data; Label and classify the numerical enterprise data to determine each index item, put the data of each index item into the object of the enterprise data matrix, and calculate the data value of the sample enterprise under each index item respectively; Calculate the sum, average, maximum value, minimum value, 25% value and 75% value of the data values of all sample companies under each indicator, and deal with abnormal data values. 3. the evaluation method of the enterprise innovation and growth ability based on big data analysis according to claim 2, is characterized in that, the calculation process of the final weight value of described index item is: Predefine a multi-level data classification label as the calculation index item; Receive all the sample data of the sample enterprise, use the gradient boosting decision tree algorithm of the machine learning xgboost library to extract the features of each sample data and classify, sort and group, and combine the input calculation index items to calculate the softmax loss function plus the regular term, and obtain each sample data one by one. The relative importance score and ranking diagram of each index item; The average value of the relative importance score of the index item and the proportion score of the preset corresponding index item is taken as the final weight value of the index item. 4. the evaluation method of the enterprise innovation and growth ability based on big data analysis according to claim 3, is characterized in that, the calculation process of the final evaluation score of each index item of each said enterprise to be evaluated is: Obtain the sample data of each enterprise dimension of each to-be-evaluated enterprise and process the data to obtain the enterprise data matrix of each to-be-evaluated enterprise; Each index item calculated by the index weight model is divided into five groups of evaluation dimensions, including endogenous innovation, collaborative investment, management and operation ability, achievement transformation ability and sustainable growth ability. Perform matrix dot product on the corresponding value in and its corresponding final weight value to obtain the evaluation score of the corresponding index item; According to the characteristics of normal distribution, the scores of each group of evaluation dimensions are converted into a 100-point system as the final evaluation score of the enterprise index item. 5. The evaluation method of enterprise innovation and growth ability based on big data analysis according to claim 2, wherein the index items of the sample enterprise include the total number of employees, the proportion of researchers, the proportion of college education, and the proportion of graduate students. Proportion, number of enterprise qualifications, industry-university-research qualifications, number of subordinate institutions, number of R&D activities, ratio of R&D expenses, R&D expenses in the previous year, last year's R&D expenses, project establishment expenses in the previous year, last year's project establishment expenses, maximum project amount, and minimum project amount Value, average value of project amount, total amount of project amount, total number of approved projects, total number of application projects, total amount of equity investment, total number of equity investment, total amount of debt in the previous year, total amount of debt in the last year, maximum amount of debt, minimum amount of debt, Average amount of claims, total amount of claims, total number of claims, total assets in the previous year, total assets in the previous year, net assets in the previous year, net assets in the previous year, main business income in the previous year, main business income in the previous year, cost in the previous year, cost in the previous year, Net profit of the previous year, net profit of the previous year, profit and tax paid in the previous year, profit and tax paid in the previous year, the proportion of the main business in the previous year, the proportion of the main business in the previous year, the net profit rate of the previous year, the net profit rate of the previous year, management certification, class I intellectual property, class II Intellectual property rights, the number of technical contracts in the previous year, the number of technical contracts in the last year, the largest technical transaction value of technical contracts, the smallest technical transaction value of technical contracts, the average technical transaction value of technical contracts, the total technical transaction value of technical contracts, the total number of technical transactions of technical contracts, the country The number of scientific and technological awards, the number of standards formed, the number of scientific and technological achievements transformed, the number of high-tech products, high-tech income, the growth rate of total assets in the previous year, the growth rate of net assets, the growth rate of main business income, the rate of change in costs and expenses, and the growth rate of net profit , the growth rate of profits and taxes paid, the growth rate of R&D expenses, the rate of change in the proportion of main business, the growth rate of net profit margin, the growth rate of R&D expenses, the growth rate of intellectual property rights, and the growth rate of R&D personnel. 6 . The method for evaluating enterprise innovation and growth capability based on big data analysis according to claim 2 , wherein the processing method of the abnormal data value is: for the data value whose proportion type data is not in the range of 0 to 1. 7 . , when it is greater than 1, it is set as the 75% value of the corresponding item, and when it is less than 1, it is set as the 25% value of the corresponding item; for the data whose growth rate is greater than 20, it is set as the 75% value of the corresponding item, and the data whose growth rate is less than -20 Set to the 25% value of the corresponding item. 7. An evaluation system for enterprise innovation and growth capability based on big data analysis, characterized in that it includes: The enterprise data collection and processing module is used to collect sample data of each enterprise dimension of the sample enterprise, and perform data cleaning, calculation and label classification of the sample data according to the time series; among which, the enterprise dimension includes business, employee, qualification, intellectual property, project, financing Loans, assets and research and development expenses; The indicator weight module is used to train the indicator weight model according to the cleaned sample data and label classification of the sample enterprises, so as to obtain the relative importance score and priority map of the indicator items according to the indicator weight model, and then calculate the final weight value of each indicator item. ; The score assigning module is used to calculate the final evaluation score and ranking of each index item of each to-be-evaluated enterprise according to the enterprise data of each to-be-evaluated enterprise and the final weight value of each index item. 8. The evaluation system of enterprise innovation and growth capability based on big data analysis according to claim 7, wherein the enterprise data collection and processing module comprises: The first acquisition unit is used to obtain the industrial and commercial information of the sample enterprises, as well as the number of researchers and the proportion of educational backgrounds in each of the past three years, information on qualification certification, the number of Class I intellectual property rights, the number of Class II intellectual property rights, and project declarations Information, loan amount, financing data, research and development expenses and technical contract information, wherein the loan amount includes the maximum loan amount, the minimum loan amount, the average loan amount and the total loan amount; The cleaning and normalization unit is used to clean and normalize the data acquired by the acquisition unit, including deleting duplicate data, evaluating irrelevant data, and replacing Chinese content with digital representation to obtain numerical enterprise data; The classification unit is used to label and classify the numerical enterprise data to determine each index item, put the data of each index item into the object of the enterprise data matrix, and calculate the data value of the sample enterprise under each index item respectively; The calculation unit is used to calculate the sum, average, maximum value, minimum value, 25% value and 75% value of data values of all sample enterprises under each index item, and process abnormal data values. 9. The evaluation system of enterprise innovation and growth capability based on big data analysis according to claim 8, wherein the indicator weight module comprises: The predefined unit is used to predefine a multi-level data classification label as the calculation index item; The machine learning unit is used to receive all the sample data of the sample enterprise, use the gradient boosting decision tree algorithm of the machine learning xgboost library to extract the features of each sample data and classify, sort and group, and combine the input calculation index items to perform softmax loss function plus regularization Item calculation, obtain the relative importance score and ranking diagram of each index item one by one; The weight calculation unit is configured to calculate the mean value of the relative importance score of the index item and the proportion score of the preset corresponding index item, and use the mean value as the final weight value of the index item. 10. The evaluation system for enterprise innovation and growth capability based on big data analysis according to claim 9, wherein the score giving module comprises: The second obtaining unit is used to obtain sample data of each enterprise dimension of each enterprise to be evaluated and perform data processing to obtain the enterprise data matrix of each enterprise to be evaluated; The evaluation unit is divided, and each index item used for the calculation of the index weight model is divided into five groups of evaluation dimensions, including endogenous innovation, collaborative investment, management and operation ability, achievement transformation ability and sustainable growth ability. The corresponding value of the index item in the enterprise data matrix is multiplied by the matrix dot with its corresponding final weight value to obtain the evaluation score of the corresponding index item; The score conversion unit is used to convert the scores of each group of evaluation dimensions into a 100-point system according to the normal distribution characteristics as the final evaluation score of the enterprise index item.

Images