Research On Hybrid Recommendation Technology Of Human Resources In The Chemical Industry Based On Deep Learning

Job application has always been a concern of job seekers.There is a huge demand for jobs in the chemical materials industry.The establishment of a proprietary platform for job seekers in the chemical industry can facilitate job recommendation for job seekers in the chemical industry.Based on the analysis of the current application status of online recruitment platforms,the existing recruitment platforms have the following problems:(1)The professional relevance is not strong,the job search information in many fields is mixed,and job seekers are at a loss in the face of huge data.(2)The recommendation method is mainly based on the user’s work intention when registering,and the changes in the user’s subsequent job search needs cannot be considered;(3)The recommendation methods of online recruitment platforms are mostly based on job popularity,which is difficult to achieve today Personalized recommendation requirements.This article studies the above issues,and the specific research content is as follows:(1)According to the subject of the corresponding algorithm for data search and algorithm research.To find the most suitable recommendation method for human resources in chemical industry.At present,there are few recruitment websites for a specific industry in China,such as the Lagou network in the computer industry,which shows its wide application prospects,and job recommendation for the chemical industry can greatly reduce the job-hunting time of job seekers.The data of job hunting information in chemical industry are obtained,and the data are preprocessed and analyzed.(2)Researched the user portrait construction technology for human resources in the chemical industry.In the basic information of job-seeking users,the work experience and education background required in the chemical industry are added,and contextual information such as browsing,delivery,and delivery time has been added to solve the single problem of job-seeking user data.The background information of job-seeking users is clustered and divided into different user subgroups,which reduces the data dimension and the amount of calculation in subsequent algorithms;at the same time,group recommendation can be made according to the characteristics of user subgroups.(3)Optimize the Term Frequency–Inverse Document Frequency(TF-IDF)algorithm.The algorithm introduces job-seeking user behavior type weights and time attenuation coefficients,and converts contextual information such as search and delivery into job-seeking user scoring weights,reducing the calculation error caused by the sparseness of the data in the subsequent calculations.(4)Aiming at the problem of data sparseness in human resource data in the chemical industry,combined with job-seeking user context information,using the idea of autoencoder,a hybrid recommendation algorithm that combines collaborative filtering and AutoSVD++ is proposed--Situation Kmeans++-Collaborative Filtering AutoSVD++(SK-CFA)algorithm.By comparing the experimental results,the SK-CFA algorithm can improve the recommendation quality while alleviating data sparsity.(5)Designed and implemented a human resource recommendation system in the chemical industry,providing job recommendation services for job seekers in the chemical industry,and analyzing the current employment status.