Research Of Personalized Recommendation Technology In Petroleum Training Assessment System

With the popularization of the training assessment system in the oilfield enterprises,some problems appear.Firstly,the proliferation of training materials makes it more difficult for employees to find the information they need and reduce their enthusiasm for learning.Secondly,the test paper did not take into account the employees' knowledge of the oil field.The training assessment system only provides the basic functions of teaching,not the "staff" as the center,the system's personalization and the lack of intelligence,which leads to the poor learning effect of oilfield employees in the training assessment system.Therefore,this dissertation applies personalized recommendation technology to the oilfield training assessment system to provide personalized training and personalized assessment services for the system.The dissertation studies and compares the individualized recommendation algorithm of major mainstream,In combination with the data characteristics of the less employee behavior data,rich data and test data in the oilfield training assessment system.The dissertation explores the semantic information of text content and provide a more accurate feature model,the LDA theme model is selected to model the recommended objects and employees.The objects of feature modeling--the data,test questions,and the text content of the employee description file are different in length.If there are common words in them,the quality of LDA topic modeling will be affected.So TF-IDF algorithm is bringed in the initialization phase of LDA theme model.The weight of words is calculated by TF-IDF,so as to reduce the influence of common words on text.The dissertation firstly recommends training materials for employees according to the content based recommendation algorithm process,and realizes personalized training based on the obtained model.It also designs the personalized group volume scheme,and realizes the staff's personalized assessment according to the degree of knowledge of the employees' oil field.Finally will use the personalized recommendation technology system and the traditional training evaluation system applied in oilfield enterprises respectively,by comparing the test score the employee training before and after the training test scores,it is concluded that use of the oil field of personalized recommendation technology training appraisal system can better improve the quality of staff training.