Human Resource Management Recommendation Based on Machine Learning Clustering in Distributed Computing
Due to the influence of feature dimension and data scale,it is more difficult to recommend human resources accurately.In order to improve the self-adaptive recommendation accuracy of human resources management platform,machine learning clustering recommendation strategy was adopted,and AP clustering algorithm was applied to large-scale human resources recommendation,and the recommendation efficiency was improved with the help of Storm compu-ting framework.Firstly,feature extraction and initialization were carried out according to the user behavior records of the human resources platform,and feature extraction was carried out with the help of PCA algorithm,then the scoring func-tion of job seekers and positions was constructed,and then the characteristics of job seekers and positions were clustered by AP clustering algorithm,and the candidate position recommendation sequence was obtained according to the cluste-ring results.All processes of AP clustering were completed in Storm distributed nodes.Through the simulation analysis of four large-scale online recruitment platforms,the performance of human resource recommendation was analyzed from the perspectives of different recommendation sequences,different cluster numbers and different feature numbers.The empirical study shows that AP clustering achieves higher Recall and F1 performance in human resource recommendation of four platforms,while distributed computing effectively improves the recommendation acceleration ratio.
Human resource recommendationMachine learningDistributed computingAP clusteringStorm compu-ting framework
NETL
NSTL
万方数据
