基于多目标混合推荐算法的就业创业平台个性化推荐研究

Research on personalized recommendation of employment and entrepreneurship platform based on multi-objective hybrid recommendation algorithm

随着大学生就业创业需求的日益增加,传统的多目标算法无法根据各学生群体特征给出个性化推荐。因此,为给大学生提供个性化的就业和创业推荐,研究基于遗传算法的改进OR-树算法(Genetic Algorithm-based MORA, GA-MORA),设计出面向学生群体的就业创业个性化推荐模型。该模型通过模拟生物进化过程来寻找最优解,最终生成个性化的推荐结果。结果可知,通过对GA-MORA算法在就业创业平台推荐中的性能评估,发现该算法在多样性等指标上表现出色。此外,研究还发现不同学生群体对职业偏好的程度受个人兴趣、专业属性、区域熟悉度和经济因素等多种因素影响。女性学生群体的区域熟悉度指标为0.8,比男性更为集中,可知女性群体在就业时更易选择在更熟悉的地方就业。综上可知,此次研究的算法模型优越,有利于为大学生就业创业提供一个可靠的方案。

With the increasing demand for employment and entrepreneurship of college students,the traditional multi-objective algorithm cannot give personalized recommendation according to the characteristics of each student group.Therefore,in order to provide personalized employment and entrepreneurship recommendations for college students,the improved OR-tree Algorithm(GAMORA)based on Genetic Algorithm is studied,and a personalized employment and entrepreneurship recommendation model for students is designed.The model simulates the process of biological evolution to find the optimal solution,and finally generates personalized recommendation results.The results show that through the performance evaluation of GA-MORA algorithm in the recommendation of employment and entrepreneurship platform,it is found that the algorithm has excellent performance in diversity and other indicators.In addition,the study also found that the degree of career preference of different student groups is affected by various factors such as personal interest,professional attributes,regional familiarity and economic factors.The regional familiarity index of female students is 0.8,which is more concentrated than that of male students.It can be seen that female students are more likely to choose more familiar places for employment.To sum up,the algorithm model of this study is superior,which is conducive to providing a reliable scheme for college students’employment and entrepreneurship.