Research And Application Of Human Resource Recommender Engine Based On Hybrid Recommender Algorithm

With the rapid development of information technology and internet, The web hasbeen a huge information repository. When faced with explosive information, It will bedifficult for the internet users to find the right message which interest them in a highlyeffective way in a short time. Thus, the problem of information overloading comes up.The traditional methods to handle this problem are classified catalogue and searchengine such as Google. But these methods cannot provide personalized informationwhen users cannot define their requirements clearly. So the recommender systemcomes to rescue. Recommender system can assist users in finding the items orservices that best fit their individual preferences.We researched on human resource field and have found out that it has been atrend for recruiters and job-hunters to hire suitable people and find a job throughinternet. And also information overloading problem happened here. For one way,job-hunters feel flooded when facing lots of jobs published through internet. Foranother way, recruiters feel flooded when facing massive applicants. The relatedresearch in this field is just based on single recommendation method. In this paper, wecame up with a hybrid recommendation method based on PLSA and content attributeand designed the relevant recommender engine. So sums up, we mainly do thefollowing research and work:(1) We studied the related technology of recommender system and the mainrecommender algorithms. Such as collaborative filtering, content-based,knowledge-based and so on. And we give a deep analysis on problems recommendersystem faced now and advantages and disadvantages of the main recommenderalgorithms separately.(2) Based on the above research, we analyzed the recommendation requirementsin recruiting and job-finding scenarios. We came up with a hybrid recommendationmethod based on PLSA and content-based recommender algorithm to implement abilateral recommendation which both consider the recruiters’ preference and thejob-hunters’ preference. We discussed the detailed implementation process and gave atest experiment showing promising results.(3) Finally, we researched on the periphery architecture and the architecture ofrecommender system. And we designed the human resource recommender engine.