Application Of Principal Component Analysis And Clustering In Liaoning Province Worker Pay Analysis

With the rapid development of China’s economy, the national economic level accordingly has been greatly improved, in order to ensure the harmonious development of the country and society, national income was corresponded to basic needs and ensure a reasonable increase. Wages by enterprises in accordance with their operations or management needs to determine, the state stipulated the minimum wage taking into account the current situation in the real economy, in order to ensure the stability of operation of enterprises, and it must also take into account workers basic needs of life, prompting companies to assume their social responsibility, and it strive to improve their profitability, the overall idea is to balance the interests of workers and enterprises to achieve win-win. Thus, the wage level of a region is not only reflected in the standard of living, but also reflects the current local economic development, has high research value.The paper’s data sources the "Liaoning Provincial Statistical Yearbook-2015", it is an annual informative, information is intensive, comprehensive statistical data covered the economy, science and technology, social and other aspects of Liaoning Province, from many aspects reflects the economic development of Liaoning province.The Liaoning Province’s 14 prefecture-level city data on wages were statisticed,including data and average wages in the post, a total of 52 indicators of wages from2011 to 2014 wage workers in the post 2014 in various industries.The main content of this paper is to analyze the Liaoning Province urban workers wages. Through data mining algorithms to analyze the city in Liaoning Province in terms of wages, it mainly completed the following two tasks.First, the using of the main component score in Liaoning Province Cities’ workers wages rankings.Principal component analysis is a comprehensive statistical method, it will be converted a plurality of composite indicators into few main components, and high-dimensional data information was changed into low-dimensional principal component information, and a few low-dimensional information can reflect the original main ingredient the vast majority of the data attribute data variables.By principal component analysis, the statistics of 52 wage index dimensionality were reduced, and it gave three main components, by calculating the main component score, it proceeded for ranking results analysis Liaoning Province 14 cities’ Ranking in staff wages,.Second, the main component analysis and cluster application using in Liaoning Province urban workers wages clustering.Cluster analysis is more important in data mining algorithm. K-means clustering algorithm is commonly used algorithm, K-means algorithm is a broad partitioning algorithm, it has the advantage that it has the ability to handle large data sets, it is capable of high-dimensional data clustering and effect was better. Features K-means clustering algorithm is to try to gather the same or similar objects together, but the different objects were divided to different classes.Principal component clustering algorithm synthesized the principal component analysis algorithm and K-means clustering algorithm, and the first principal is component analysis and the second step is clustering. The three main components of principal component analysis results were used to classifying the 14 prefecture-level city by analyzing the 14 prefecture-level cities’ staff wages, and the cities were divided into three categories, and it analyzed the clustering results to provide reference information for the wages of workers in Liaoning Province.