With the rapid development of China universities, especially the expanding enrollment of higher occupation technical school in recent years promotes Chinese higher education into the popularization era. At the same time, it also leads to that college student employment situation is very serious. The employment pressure intensified. In the face of surplus resources situation, comprehensive quality requirements of graduate students are also getting higher in social workplace. So the college students' occupation literacy level has become a key factor in the employment competition, so the research of College Students' occupation quality system is very important. This paper makes full use of data mining and data analysis algorithm, which puts forward a novel multidimensional professionalism evaluation model based on Logistic regression and decision tree algorithm. The model considers common requirements and personalized recommendations of professionalism. This model can be as much as possible used to help students predict their occupation literacy and interest required key professionalism, which can improve and adjust themselves and increase the possibility of being employed.