A Model Of College Students' Molument Prediction Based On The Classification Algorithm With Natural Neighbor

With the development of Chinese higher education from elite education to mass education,the employment situation of college graduates is becoming more and more serious.One of the most important reasons for the increasingly difficult employment of college students lies in the high employment expectation,which is mainly reflected in the high expectation for the salary.When college students to find a job,most people choose company that will pay a higher salary,but they always ignore company that is better suited to them but will pay a less salary.Therefore,it is very important to make an accurate evaluation of the graduates' employment ability and help them to make a reasonable salary expectation.To solve the problem of employment difficulty of graduates who expect for the impractical emolument,the paper builds a model for emolument prediction on the basis of an classification algorithm with natural neighbor:1.Researching and analyzing the main reasons for the high employment expectations of College students.2.Introducing the definition of the classification process and several common classification algorithms and their advantages and disadvantages.Introducing the mathematical model and the calculation procedure of factor analysis.Then focusing on the concept and the key idea of Natural Neighbor(3N).The advantage of the Natural Neighbor technique is that neighbors of each sample are adaptively formed by the algorithm which does not need to set any parameters.3.Extracting hot words in recruitment requirements,and then setting up 15 variables.After analyzing the employment data of information of technology major graduates in past three years,extracting 4 potential factors that determine the level of College Students' employment compensation,that is,learning ability,practical ability,interpersonal skills and professional ability.4.Proposing the classification algorithm based on Natural Neighbor.Through analyzing,the defect of Naturally Neighbor when used in the classification of high-dimensional data is found,and a new way of weight assignment for training data based on Natural Neighbor search algorithm,related Natural Neighbor definitions and the weight way is proposed to improve the accuracy rate of classification.Then the algorithm classifies the test samples by using Natural Neighbor algorithm and the training data with the weight.According to the experiment,the result shows that the classification algorithm based on natural neighbor is better than KNN algorithm and weighted KNN algorithm.The prediction accuracy of the model for emolument prediction on the basis of the classification algorithm based on natural neighbor is 80.16%.The paper can guide graduates to build a reasonable emolument prodiction or improve employablility.