Research On Employment Situation In Shandong Province Based On Principal Component Analysis And Clustering Algorith

With the rapid development of China’s economy,the employment environment of citizens needs to be basically guaranteed.The overall idea is to take into account the interests of job seekers and enterprises to achieve a win-win situation.The employment situation of a region not only reflects the development status of the local economy,but also reflects the development level of various industries,which has relatively high research value.The data source of this article is "Shandong Statistical Yearbook-2021".Shandong Statistical Yearbook is an informative statistical yearbook with highly dense statistical information,which reflects the development level of Shandong Province from multiple aspects.This article will statistically analyze the data on the employment situation of 16prefecture-level cities in Shandong Province,using the employment categories of each prefecture-level city in Shandong Province in 2021 as indicators,with a total of 19 employment category indicators.Through principal component analysis and K-means clustering algorithm,it will analyze the employment situation of 16 cities in Shandong Province,mainly completing the following tasks:First,using principal component analysis,index averaging,and principal component analysis after index homotaxis,and improved entropy weight method,the dimensionality of19 employment category indicators in the statistical data was reduced,and four principal components were obtained,with contribution rates exceeding 85%.Secondly,K-means clustering and K-means clustering based on density standard deviation optimization of initial clustering centers are performed on the dimensionality reduced data,and the analysis and evaluation of employment in Shandong Province are completed,providing reference information for employment in Shandong Province.The results show that the improved K-means clustering algorithm based on density standard deviation to optimize the initial clustering center achieves better clustering results than traditional K-means clustering algorithms using fewer iterations,and also has less computational time than traditional K-means clustering algorithms,indicating that the improved K-means clustering algorithm outperforms traditional K-means clustering algorithm,reflecting the effectiveness and accuracy of the improved method.