Quantitative Analysis Technology For Talent Recruitment In Enterprises

Recruitment is of great significance to enterprises. It is an important prerequisite and foundation for ensuring the smooth development of human resources management in enterprises.

How to effectively evaluate the candidates and select the most suitable personnel is a major problem facing the human resources department.

The traditional recruitment interview method, whether conventional written examination, interview or evaluation center, has a greater impact on the result of subjective judgment and choice, which makes the subjective components of the recruitment process very large, sometimes even affecting the effect of recruitment.

The author here introduces several relatively objective and scientific quantitative analysis techniques for recruitment evaluation of enterprises, and discuss with everyone.

The basic idea of analytic hierarchy process (AHP) of 1. is to decompose the comprehensive ability of candidates into several indicators and levels. At the lowest level, the weights of all factors can be obtained through 22 comparison, and finally the final composite index of each candidate is calculated through the analysis from low to high levels. The largest index is the best candidate.

Its basic method is to set up a hierarchical structure model of the candidate evaluation index.

To establish a hierarchy model of evaluation indicators, we should first have a clear understanding of the positions to be recruited, and find out what factors it involves, such as objectives, sub goals, departments, constraints, possible situations, and the relationship between various factors.

Secondly, we divide the evaluation index into several levels.

After setting up the evaluation index level model, we can make 22 comparisons with all the indicators of the recruits, and construct the judgment matrix.

The judgment matrix is an important part of qualitative pition to quantitative analysis. Then, by checking the eigenvectors of the judgement matrix and testing the consistency of the judgement matrix, we check whether the employers have judged the consistency of the thinking when constructing the judgment matrix.

After consistency checking, the normalized feature vector can be used as a ranking weighting value of a certain level to a certain level of evaluation index. Then the ranking weight is calculated from the high level to the low level, and the total ranking of the candidates is obtained.

2. fuzzy decision making method is very vague in real life.

If a tall person is tall, there is no definite definition of it. Different people will have different understandings.

In addition, the concepts of ability, attitude and character are obscure.

The connotation of these concepts is clear, but the extension is vague.

In the reality of enterprise recruitment, the concept of many indicators is vague, so fuzzy decision making method is becoming a practical tool for recruitment decisions.

The fuzzy comprehensive evaluation method is a method of evaluating the merits and faults by fuzzy set theory.

It is characterized by combining qualitative analysis with quantitative analysis, combining subjective analysis with objective analysis.

The basic method of fuzzy decision is to construct the evaluation index set X and the rating domain V.

For example, X={X1 (knowledge), X2 (ability), X3 (personality), X4 (motivation)}, V={V1 (very good), V2 (good), V3 (not too good), V4 (not good)}.

If there is a "knowledge" index for applicants, 30% of employers think "very good", 60% think "good", and 10% think "not good", but no one thinks "bad". For simplicity sake, we can approximately think that the evaluation set of "knowledge" index for applicant a is (0.3,0.6,0.1,0).

Similar analogy, we can get the evaluation matrix of candidates and solve the evaluation matrix of candidates and their corresponding weights. Finally, we will get the comprehensive score of each candidate.

In practice, fuzzy decision making is often used in combination with expert analysis and evaluation and analytic hierarchy process.

The 3. advantage and disadvantage coefficient method is a quantitative analysis method for selecting a better candidate through comparing the indexes of the candidates with the other candidates.

In real life, no applicant is definitely better than the other candidates, and none of them is better than others.

For enterprises, the importance of quality indicators is not the same. Some qualities are relatively important, while others are relatively minor.

Therefore, before calculating the good and bad coefficients, the enterprises need to give different weights to different evaluation indexes. Then, by comparing the various evaluation indexes with standardization, the indexes can be compared, and then the superior and inferior coefficients can be calculated.

The so-called coefficient of merit refers to the ratio between the sum of the weights of a candidate who is superior to that of another applicant and the sum of all weights.

The inferior coefficient is calculated by comparing the difference between the best and the worst of the two schemes.

Because the excellent coefficient only reflects the best candidate, and does not reflect the degree of the applicant's excellence, the bad coefficient only reflects the degree of the applicant's inferiority, but does not reflect the bad candidate. Therefore, the excellent and bad coefficient should be considered comprehensively when making recruitment decisions.

The advantage and disadvantage coefficient method is based on the good and bad coefficient to gradually eliminate the undesirable candidates, which has a wide application value in the recruitment process.

4. artificial neural network, Artificial Neural Network, is a new information processing science. It is a frontier research field for simulating human neural structure and intelligence. Because of its unique structure and information processing methods, it has achieved remarkable results in many practical applications.

In recent years, due to the rapid development of neuroscience, mathematical science, information science and computer science, it is possible to realize the realization of artificial neural network based on working mode of neurons and non procedural information processing.

Artificial neural network does not need to build any mathematical model. It only learns from past experience and expert knowledge, and achieves its output consistent with expected output through network learning.

The self-learning ability of the network makes the traditional knowledge acquisition work which is the most difficult application of the traditional expert system technology into the variable structure adjustment process of the network. It can make a reasonable decision, give a satisfactory solution to the complex problem according to the knowledge that has been learned and the experience of dealing with the problem, give a satisfactory solution, or make an effective preview and estimate for the future process.

As long as we can choose the parameters according to the scientific data to build the network model, it can get expert experience data in the data, judge the quality indicators of the candidates, and give more objective and reasonable results.

At present, the application of artificial neural network in management is still in the stage of research and development.

Huang Yuejun: Master of management, a famous management scientist, former vice president of Hunan University, Mr. Li Shucheng, former president of Xiangtan University, devoted to research and practice in the fields of enterprise diagnosis, human resources, project planning, management training, scientific decision making, artificial neural network and so on.

He has worked as a human resource project manager for a number of foreign and private enterprises, engaged in university management education and theoretical research, management consulting and training, and served as a senior HR consultant of many consulting companies. He has a solid theoretical foundation and rich practical experience in the field of enterprise management and human resource consulting, and has published several articles in media magazines of several economic management fields.