Method and system for searching and hunting professional talents

Abstract

The invention relates to the technical field of information processing, in particular to a professional technical talent visiting and hunting method and system, and the method comprises the following steps: carrying out the statistics of the occurrence frequency of words in the answer of each candidate based on the interview answer text of professional technical talents, generating a candidate word frequency matrix, carrying out the normalization processing of the candidate word frequency matrix, and obtaining a candidate word frequency matrix; and obtaining a standardized word frequency matrix. According to the method, interview answers of candidates are converted into the standardized word frequency matrix, the depth and breadth of data processing are ensured, word global probability distribution is calculated, information entropy vectors are generated, and the capacity of capturing and analyzing the information amount is improved. In addition, the combination of the information entropy vector and the standardized word frequency matrix is used for calculating the total information entropy of the candidates, and a quantitative method is provided for evaluating the innovation ability of the candidates. And by comprehensively considering the word frequency and the information entropy, the content quality and originality of the candidates are reflected more accurately, so that the scientificity and effectiveness of the recruitment process are improved.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to a method and system for searching and recruiting professional and technical personnel. Background Art

[0002] Information processing technology is a key branch of computer science that deals with the collection, storage, processing, conversion and management of data. The main goal of this field is to transform raw data into useful information through the use of algorithms and computer systems. The recruitment method is a technical process specifically used to screen and analyze job applicants, especially in finding talents with professional technical capabilities.

[0003] Although existing information processing technology is effective in basic data processing, it is still insufficient in in-depth analysis and comprehensive ability assessment for talent screening. Existing methods fail to fully utilize the potential information in the data, resulting in reliance on simple word frequency statistics or surface data interpretation in the talent evaluation process. This limitation makes the talent evaluation process lack an in-depth understanding of the candidate's innovation ability and professional quality, which may lead to the neglect of high-potential candidates. Therefore, improvements are needed. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a method and system for searching and recruiting professional and technical personnel.

[0005] In order to achieve the above object, the present invention adopts the following technical scheme, a method for searching and recruiting professional and technical personnel, comprising the following steps: Based on the interview answer texts of professional and technical personnel, the number of occurrences of words in each candidate's answer is counted to generate a candidate word frequency matrix, and the candidate word frequency matrix is ​​normalized to obtain a standardized word frequency matrix; Based on the standardized word frequency matrix, the overall frequency of occurrence of each word in all candidates' answers is calculated to obtain a global probability distribution of the word; the information entropy of each word is calculated according to the global probability distribution of the word to generate a word information entropy vector; the word information entropy vector and the standardized word frequency matrix are calculated element by element to obtain a candidate information entropy matrix; Based on the candidate information entropy matrix, sum up the information entropy contribution value of each candidate, calculate the total information entropy, obtain the candidate total entropy value, compare the candidate total entropy value with a preset threshold, and generate an innovation ability score for each candidate; Based on the innovation ability score and the written test results, a comprehensive score for each candidate is calculated, and the candidates are sorted by the comprehensive score to obtain a candidate ranking list.

[0006] Preferably, the steps of obtaining the standardized word frequency matrix are: Based on the interview answer texts of professional and technical personnel, count the number of occurrences of words in each candidate's answer to obtain the candidate word frequency matrix; Based on the candidate word frequency matrix, the relative frequency of each word in the candidate's answer is calculated using the following formula: ; in, For candidates Pair of words The number of occurrences of is the total number of words, is the relative frequency, For candidates The total number of times all words were used; Based on the relative frequencies, a standardized word frequency matrix is ​​generated through normalization processing.

[0007] Preferably, the steps of obtaining the global probability distribution of words are: Extract the sum of the frequencies of each word in all candidates' answers from the standardized word frequency matrix to form a cumulative frequency array; Based on the cumulative frequency array, the global occurrence probability of each word is calculated using the following formula: ; in, For candidates For words The standardized frequency, is the total number of candidates, is the total number of words, is the global occurrence probability; Based on the global occurrence probability, a global probability distribution matrix of the word is constructed to obtain the global probability distribution of the word.

[0008] Preferably, the steps of obtaining the candidate information entropy matrix are: Based on the global probability distribution of the words, the information entropy of each word is calculated one by one, and the word information entropy vector is generated according to the information entropy. The calculation formula of the information entropy is: ; in Expressing words The global occurrence probability of For words Information entropy of Based on the word information entropy vector, extract word information entropy values ​​one by one and store them in an information entropy value array, where each element corresponds to the information entropy of a word; The information entropy value array is multiplied element by element by the standardized word frequency matrix to generate a candidate information entropy matrix.

[0009] Preferably, the steps for obtaining the total entropy value of the candidate are: Check each row in the candidate information entropy matrix, and for each candidate, add up all elements in the row one by one to obtain the total information entropy contribution of each candidate; Based on the sum of each candidate's information entropy contribution, verify the correctness of the calculation of each sum, and confirm that there is no calculation error by comparing the original data with the calculation results to obtain the total entropy value of the candidate.

[0010] Preferably, the steps for obtaining the innovation ability score are: Obtaining the total entropy value of the candidate and setting a preset threshold; Compare the total entropy value of each candidate with a preset threshold, check whether the entropy value of each candidate exceeds the preset threshold, and obtain a comparison result; Based on the comparison, an innovation score is assigned to each candidate that exceeds a preset threshold.

[0011] Preferably, the steps for obtaining the candidate ranking list are: Collect each candidate's innovation ability score and corresponding written test score; Based on the innovation ability score and written test score, a comprehensive score is calculated for each candidate using the following formula: ; in, Indicates the candidate The innovation ability score of Indicates the written test results. and is the weight coefficient of the score and grade, is the adjustment factor, is the comprehensive score; Based on the comprehensive scores, all candidates are sorted from high to low to generate a ranking list, thereby obtaining a candidate ranking list.

[0012] The present invention provides a search and recruitment system, comprising: The data processing and standardization module is used to collect the interview answer texts of professional and technical personnel, count the word frequency in each candidate's answer, obtain the candidate word frequency matrix, and normalize the candidate word frequency matrix to form a standardized word frequency matrix; The global probability and information entropy calculation module uses the standardized word frequency matrix to calculate the frequency of each word in all candidates' answers to obtain the global probability distribution of the word. Based on the global probability distribution of the word, the information entropy of each word is calculated to form a word information entropy vector. The word information entropy vector and the standardized word frequency matrix are calculated element by element to generate a candidate information entropy matrix. The comprehensive scoring and sorting module is used to sum the information entropy contribution value of each candidate in the candidate information entropy matrix to form the candidate's total entropy value, compare the candidate's total entropy value with the preset threshold, generate each candidate's innovation ability score, combine the written test scores, calculate each candidate's comprehensive score, sort the candidates according to the comprehensive score, and obtain a candidate ranking list.

[0013] Compared with the prior art, the advantages and positive effects of the present invention are: The present invention ensures the depth and breadth of data processing by converting the interview answers of candidates into a standardized word frequency matrix, calculates the global probability distribution of words and generates information entropy vectors, and improves the ability to capture and analyze information. In addition, the combination of information entropy vectors and standardized word frequency matrices is used to calculate the total information entropy of candidates, providing a quantitative method for evaluating their innovative ability. And by comprehensively considering word frequency and information entropy, the content quality and originality of candidates are more accurately reflected, thereby improving the scientificity and effectiveness of the recruitment process. The comparison of the total entropy value of the candidate with the preset threshold and the subsequent innovation ability score further enhance the objectivity and fairness of the evaluation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic diagram of the steps of the present invention. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0016] See also Figure 1 The present invention provides a technical solution, a method for searching and recruiting professional and technical personnel, comprising the following steps: Based on the interview answer texts of professional and technical personnel, count the number of occurrences of words in each candidate's answer, generate a candidate word frequency matrix, and normalize the candidate word frequency matrix to obtain a standardized word frequency matrix; Based on the standardized word frequency matrix, calculate the overall frequency of each word in all candidates' answers to obtain the global probability distribution of the word; calculate the information entropy of each word based on the global probability distribution of the word to generate the word information entropy vector; calculate the word information entropy vector and the standardized word frequency matrix element by element to obtain the candidate information entropy matrix; Based on the candidate information entropy matrix, the information entropy contribution value of each candidate is summed up to calculate the total information entropy to obtain the candidate total entropy value, and the candidate total entropy value is compared with the preset threshold to generate the innovation ability score of each candidate; Based on the innovation ability score and the written test results, the comprehensive score of each candidate is calculated, and the candidates are ranked according to the comprehensive score to obtain a candidate ranking list.

[0017] The steps to obtain the standardized word frequency matrix are: Based on the interview answer texts of professional and technical personnel, count the number of occurrences of words in each candidate's answer to obtain the candidate word frequency matrix; Based on the candidate word frequency matrix, calculate the relative frequency of each word in the candidate's answer. The calculation formula is: ; in, For candidates Pair of words The number of occurrences of is the total number of words, is the relative frequency, For candidates The total number of times all words were used; Based on the relative frequency, a standardized word frequency matrix is ​​generated through normalization.

[0018] Specifically, the formula calculation process: In order to obtain the standardized frequency of words First, we need to determine each candidate's Number of times used , which can be obtained by counting the interview text and then calculating the candidate The total number of times all words are used, i.e. ,in is the total number of words in the vocabulary, by Divide by get , which represents the word In the candidate of all words in the candidate's Mentioning the word "technology" in an interview times, and his total vocabulary usage is times, then the normalized frequency of the word “technology” will be ,Right now: ; This result shows that if A higher value for means that the candidate relies more on this vocabulary in his or her answer, which may be an indication of the use of professional terminology or reflect the candidate's proficiency in a certain professional field.

[0019] Identify and extract all the words in the interview text, including common words and professional terms. During the word identification process, pay special attention to distinguishing professional-related and non-related words to avoid bias in the analysis results. Next, count each word mentioned by each candidate to ensure that all words are accurately counted and recorded. This count includes repeated statistics of each word in different answers. Then, the statistical results form a preliminary word frequency matrix, which lists the number of times each candidate uses each word. Normalization is performed on each element in the word frequency matrix. This operation is based on the total number of words used by each candidate. The frequency of each word is calculated by dividing its occurrence times by the total number of times of all the candidate's words to ensure the standardization of the results. The normalization criterion is adjusted according to the frequency of the word in the text relative to the frequency of all words to ensure comparability between different candidates. The result after normalization is the usage ratio of each word relative to the total vocabulary. This ratio reflects each candidate's preference and reliance on specific words, which is crucial for evaluating the candidate's expertise and expression ability.

[0020] The steps to obtain the global probability distribution of words are: Extract the sum of the frequencies of each word in all candidates' answers from the standardized word frequency matrix to form a cumulative frequency array; Based on the cumulative frequency array, the global occurrence probability of each word is calculated using the following formula: ; in, For candidates For words The standardized frequency, is the total number of candidates, is the total number of words, is the global occurrence probability; Based on the global occurrence probability, the global probability distribution matrix of the word is constructed to obtain the global probability distribution of the word.

[0021] Specifically, the word frequency is extracted and counted from the answer text submitted by each candidate, and the total occurrence of each word in all candidates' answers is calculated to construct a cumulative frequency array. This process first involves lexical segmentation of each candidate's text, identifying and counting the number of occurrences of each word, and then summarizing the data for all candidates. For example, if a word appears 5 times, 3 times, and 2 times in the texts of three different candidates, the cumulative frequency of the word is 10 times. This data provides the basis for the calculation of global probability. Each value in the cumulative frequency array represents the frequency of use of a specific word in the entire candidate group, providing the necessary input for further data analysis such as information entropy calculation.

[0022] The benefit of the formula is that by comparing the frequency of each word with the total frequency of all words, it can intuitively show which words have a higher proportion in the candidates' answers, which helps to understand which concepts or technologies are generally concerned by candidates; The steps to obtain the parameters are as follows: For words The number of times used is obtained, and The total number of candidates and the total number of words, respectively, are directly obtained statistics.

[0023] Calculation process: There are 3 candidates, each candidate uses a total of 5 different words, and one specific word is used 10, 15 and 20 times in the three candidates respectively. The total number of times all words are used is 300 times; substitute into the formula to calculate the global probability of the word: ; The results indicate that the word is used in 15% of all candidates, reflecting its importance and frequency in technical discussions.

[0024] By summarizing the global probability of each word, a global probability distribution matrix is ​​constructed. This matrix provides a comprehensive view of the relative importance of each word in all candidate responses. For example, if the global probability of the word "machine learning" is significantly higher than other words, this indicates that machine learning is a hot topic among the current candidate population. In this way, the weight of each word in the candidate's knowledge system can be quantitatively evaluated, thereby evaluating the candidate's knowledge depth and professional focus. This matrix not only helps to identify which technologies or concepts are of general concern to candidates, but also helps to understand the candidate's possible innovation ability and problem-solving strategies.

[0025] The steps to obtain the candidate information entropy matrix are: Based on the global probability distribution of words, the information entropy of each word is calculated one by one to generate the word information entropy vector. The calculation formula of information entropy is: ; in Expressing words The global occurrence probability of For words Information entropy of Based on the word information entropy vector, extract the word information entropy value one by one and store it in the information entropy value array, where each element corresponds to the information entropy of a word; Multiply the information entropy value array and the normalized word frequency matrix element by element to generate the candidate information entropy matrix.

[0026] Specifically, in the formula, the global probability of each word is It is obtained by dividing the frequency of a word appearing in a text by the total number of words in the text. For example, if a word appears 15 times in a set of 1,000 words, Then it is 0.015; Calculation process: global probability of a word , its information entropy The calculation process is: Bit; This result shows that words are more uncommon and provide more information; this result shows that words with higher information entropy values ​​are rarer in the text and therefore provide more information.

[0027] Before constructing the candidate information entropy matrix, first ensure that the frequency of use of each candidate's vocabulary has been accurately calculated and converted into a standardized word frequency matrix through standardization; next, the above-mentioned information entropy value array is multiplied element by element with the standardized word frequency matrix. This operation refines the amount of information used by each candidate for each word. For example, if the candidate's frequency of use of a high information entropy word is 0.05, and the information entropy of the word is 2.3, then the word contributes 0.115 to the information entropy of the candidate. This method allows quantitative evaluation of which high-information words the candidate used in his answer, thereby indirectly evaluating his language richness and innovation ability.

[0028] The steps to obtain the total entropy value of the candidate are: Check each row in the candidate information entropy matrix, and for each candidate, add up all the elements in the row one by one to get the sum of each candidate's information entropy contribution; Based on the sum of each candidate's information entropy contribution, verify the correctness of the calculation of each sum, and confirm that there is no calculation error by comparing the original data with the calculation results to obtain the total entropy value of the candidate.

[0029] Specifically, by checking the candidate information entropy matrix row by row, we first determine that each row in the matrix represents the data of a candidate. For each candidate, we traverse each row of the matrix and add up the information entropy contribution value element by element, thereby reflecting the information complexity displayed by each candidate through his or her answers in the interview. Through this accumulation method, an exact total information entropy contribution is provided for each candidate.

[0030] After obtaining the sum of each candidate's information entropy contribution, the further task is to verify the correctness of the calculation of these sums. This verification process includes comparison with the original matrix data to ensure that no calculation errors or deviations in data input have occurred. By checking each step of the calculation process one by one, it is confirmed that the sum of each candidate is consistent with the result obtained through independent calculation. This comparison not only includes the verification of values, but also involves the review of the calculation method to ensure that all values ​​are based on the same calculation standard, and ultimately ensure that the total information entropy value of each candidate is accurate.

[0031] Based on the sum of each candidate's information entropy contribution, verify the correctness of the calculation of each sum, and confirm that there is no calculation error by comparing the original data with the calculation results to obtain the total entropy value of the candidate.

[0032] Specifically, by checking the candidate information entropy matrix row by row, we first determine that each row in the matrix represents the data of a candidate. For each candidate, we traverse each row of the matrix and add up the information entropy contribution value element by element, thereby reflecting the information complexity displayed by each candidate through his or her answers in the interview. Through this accumulation method, an exact total information entropy contribution is provided for each candidate.

[0033] After obtaining the sum of each candidate's information entropy contribution, the further task is to verify the correctness of the calculation of these sums. This verification process includes comparison with the original matrix data to ensure that no calculation errors or deviations in data input have occurred. By checking each step of the calculation process one by one, it is confirmed that the sum of each candidate is consistent with the result obtained through independent calculation. This comparison not only includes the verification of values, but also involves the review of the calculation method to ensure that all values ​​are based on the same calculation standard, and ultimately ensure that the total information entropy value of each candidate is accurate.

[0034] The steps to obtain the innovation ability score are as follows: Get the total entropy value of the candidates and set the preset threshold; Compare the total entropy value of each candidate with a preset threshold, check whether the entropy value of each candidate exceeds the preset threshold, and obtain a comparison result; Based on the comparison results, an innovation score is assigned to each candidate that exceeds a preset threshold.

[0035] Specifically, the total entropy value of each candidate is identified, and the innovation ability of each candidate is evaluated according to a predetermined preset threshold value, which is precisely set by analyzing past entropy data and consulting with technical personnel, and is intended to identify candidates with high innovation potential; then, the total entropy value of each candidate is compared with the preset threshold value. This process involves sequentially loading and comparing the entropy data of each candidate, checking one by one, recording the results of whether the entropy value exceeds the threshold value, and generating a list of comparison results.

[0036] After confirming the comparison result of the total entropy value of each candidate with the preset threshold, the score allocation stage begins, and an innovation score is allocated to each candidate who exceeds the threshold. The score allocation is based on the specific extent to which the candidate's entropy value exceeds the threshold. The operation includes reading the comparison result list of the previous step, checking the extent to which the entropy value of each candidate exceeds the threshold one by one, and scoring according to the set scoring criteria. This scoring criteria is set according to the percentage of the entropy value exceeding the threshold. For example, candidates who exceed 20% will receive the highest score, and candidates who exceed 10% to 20% will receive the second highest score. In this way, the innovation ability of each candidate is accurately quantified and reflected in the score, ensuring the consistency and objectivity of the score; finally, a detailed list of candidate innovation scores is generated.

[0037] The steps to obtain the candidate ranking list are: Collect each candidate's innovation ability score and corresponding written test score; Based on the innovation ability score and written test score, a comprehensive score is calculated for each candidate using the following formula: ; in, Indicates the candidate Innovation ability score, Indicates the written test results. and is the weight coefficient of the score and grade, is the adjustment factor, is the comprehensive score; Based on the comprehensive scores, all candidates are sorted from high to low to generate a ranking list to obtain a candidate ranking list.

[0038] Specifically, we collect each candidate's innovation ability score and corresponding written test scores, and ensure the accuracy and completeness of the data obtained through data collation and verification. These data include the candidate's test scores in different fields and the innovative thinking ability demonstrated in practical problem solving. These data will serve as the basic input for calculating the comprehensive score.

[0039] The benefit of the formula is that it combines two different scoring criteria by taking the square root method, so that the score not only takes into account the importance of each score, but also balances the weights between the scores, making the comprehensive score more fair and comprehensive. The parameter is the innovation capability score. The parameters are obtained directly from the candidate's standardized written test scores. and The parameters are set based on the organization's past assessment of the importance of these two dimensions. It is adjusted based on the standard deviation of all candidate scores in historical data to ensure the uniformity of the comprehensive score and the rationality of comparison; Calculation process: If , , , , ,but: ; The result shows that the candidate's overall score is 1.74, a score that reflects the candidate's ability level relative to others.

[0040] Based on the calculated comprehensive scores, all candidates are sorted from high to low to generate a ranking list. This list not only clearly shows the ranking of each candidate based on the comprehensive score, but also reflects the candidate's performance under different tests and evaluation criteria through dynamic adjustment, making the evaluation more comprehensive and fair. The ranking results directly affect the candidate's final hiring decision, ensuring the efficiency and transparency of the selection process, and providing a scientific basis for organizations to find the most suitable talents.

[0041] The present invention provides a search and recruitment system, comprising: The data processing and standardization module is used to collect the interview answer texts of professional and technical personnel, count the word frequency in each candidate's answer, obtain the candidate word frequency matrix, and normalize the candidate word frequency matrix to form a standardized word frequency matrix; The global probability and information entropy calculation module uses the standardized word frequency matrix to calculate the frequency of each word in all candidates' answers to obtain the global probability distribution of the word. Based on the global probability distribution of the word, the information entropy of each word is calculated to form a word information entropy vector. The word information entropy vector and the standardized word frequency matrix are calculated element by element to generate a candidate information entropy matrix. The comprehensive scoring and sorting module is used to sum the information entropy contribution value of each candidate in the candidate information entropy matrix to form the candidate's total entropy value, compare the candidate's total entropy value with the preset threshold, generate each candidate's innovation ability score, combine the written test scores, calculate each candidate's comprehensive score, sort the candidates according to the comprehensive score, and obtain a candidate ranking list.

[0042] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for searching and recruiting professional and technical personnel, characterized in that: The following steps are involved: Based on the interview answer texts of professional and technical personnel, the number of occurrences of words in each candidate's answer is counted to generate a candidate word frequency matrix, and the candidate word frequency matrix is ​​normalized to obtain a standardized word frequency matrix; Based on the standardized word frequency matrix, the overall frequency of occurrence of each word in all candidates' answers is calculated to obtain a global probability distribution of the word; Calculate the information entropy of each word according to the global probability distribution of the word to generate a word information entropy vector; calculate the word information entropy vector and the standardized word frequency matrix element by element to obtain a candidate information entropy matrix; Based on the candidate information entropy matrix, sum up the information entropy contribution value of each candidate, calculate the total information entropy, obtain the candidate total entropy value, compare the candidate total entropy value with a preset threshold, and generate an innovation ability score for each candidate; Based on the innovation ability score and the written test results, a comprehensive score for each candidate is calculated, and the candidates are sorted by the comprehensive score to obtain a candidate ranking list.

2. The method for searching and recruiting professional and technical personnel according to claim 1, characterized in that: The steps for obtaining the standardized word frequency matrix are: Based on the interview answer texts of professional and technical personnel, count the number of occurrences of words in each candidate's answer to obtain the candidate word frequency matrix; Based on the candidate word frequency matrix, the relative frequency of each word in the candidate's answer is calculated using the following formula: ； in, For candidates Pair of words The number of occurrences of is the total number of words, is the relative frequency, For candidates The total number of times all words were used; Based on the relative frequencies, a standardized word frequency matrix is ​​generated through normalization processing.

3. The method for searching and recruiting professional and technical personnel according to claim 1, characterized in that: The steps for obtaining the global probability distribution of the words are: Extract the sum of the frequencies of each word in all candidates' answers from the standardized word frequency matrix to form a cumulative frequency array; Based on the cumulative frequency array, the global occurrence probability of each word is calculated using the following formula: ； in, For candidates For words The standardized frequency, is the total number of candidates, is the total number of words, is the global occurrence probability; Based on the global occurrence probability, a global probability distribution matrix of the word is constructed to obtain the global probability distribution of the word.

4. The method for searching and recruiting professional and technical personnel according to claim 1, characterized in that: The steps for obtaining the candidate information entropy matrix are: Based on the global probability distribution of the words, the information entropy of each word is calculated one by one, and the word information entropy vector is generated according to the information entropy. The calculation formula of the information entropy is: ； in Expressing words The global occurrence probability of For words Information entropy of Based on the word information entropy vector, extract word information entropy values ​​one by one and store them in an information entropy value array, where each element corresponds to the information entropy of a word; The information entropy value array is multiplied element by element by the standardized word frequency matrix to generate a candidate information entropy matrix.

5. The method for searching and recruiting professional and technical personnel according to claim 1, characterized in that: The steps for obtaining the total entropy value of the candidate are: Check each row in the candidate information entropy matrix, and for each candidate, add up all elements in the row one by one to obtain the total information entropy contribution of each candidate; Based on the sum of each candidate's information entropy contribution, verify the correctness of the calculation of each sum, and confirm that there is no calculation error by comparing the original data with the calculation results to obtain the total entropy value of the candidate.

6. The method for searching and recruiting professional and technical personnel according to claim 1, characterized in that: The steps for obtaining the innovation capability score are as follows: Obtaining the total entropy value of the candidate and setting a preset threshold; Compare the total entropy value of each candidate with a preset threshold, check whether the entropy value of each candidate exceeds the preset threshold, and obtain a comparison result; Based on the comparison, an innovation score is assigned to each candidate that exceeds a preset threshold.

7. The method for searching and recruiting professional and technical personnel according to claim 1, characterized in that: The steps for obtaining the candidate ranking list are: Collect each candidate's innovation ability score and corresponding written test score; Based on the innovation ability score and written test score, a comprehensive score is calculated for each candidate using the following formula: ； in, Indicates the candidate The innovation ability score of Indicates the written test results. and is the weight coefficient of the score and grade, is the adjustment factor, is the comprehensive score; Based on the comprehensive scores, all candidates are sorted from high to low to generate a ranking list, thereby obtaining a candidate ranking list.

8. A search and recruitment system for professional and technical personnel according to any one of claims 1 to 7, characterized in that: include: The data processing and standardization module is used to collect the interview answer texts of professional and technical personnel, count the word frequency in each candidate's answer, obtain the candidate word frequency matrix, and normalize the candidate word frequency matrix to form a standardized word frequency matrix; The global probability and information entropy calculation module uses the standardized word frequency matrix to calculate the frequency of each word in all candidates' answers to obtain the global probability distribution of the word. Based on the global probability distribution of the word, the information entropy of each word is calculated to form a word information entropy vector. The word information entropy vector and the standardized word frequency matrix are calculated element by element to generate a candidate information entropy matrix. The comprehensive scoring and sorting module is used to sum the information entropy contribution value of each candidate in the candidate information entropy matrix to form the candidate's total entropy value, compare the candidate's total entropy value with the preset threshold, generate each candidate's innovation ability score, combine the written test scores, calculate each candidate's comprehensive score, sort the candidates according to the comprehensive score, and obtain a candidate ranking list.

Images