To explore students' political education in major universities in China, starting with the actual situation of the allocation of ideological and political education resources in colleges and universities, combined with relevant theories, the allocation evaluation index system of ideological and political education resources in colleges and universities was established based on the in-depth analysis of the purpose, content and principles of the evaluation system for the allocation of resources in ideological and political education in colleges and universities. The evaluation index system covered specific indexes such as the average area of student housing, the number of books owned by students, the number of computers per student, the ratio of teachers and students and so on. The BP neural network was selected as the main evaluation method for constructing the resource allocation evaluation system of ideological and political education in colleges and universities. The analysis and processing of the input layer variables in the concrete construction of the BP network model were set up, and the practice process of the resource allocation evaluation system of the ideological and political education in the BP neural network was studied. Specifically, in the MATLABR2012a software environment, through the concrete data of the allocation of resources in the ideological and political education of colleges and universities and the extended data obtained by using the interpolation method, the BP neural network system was constructed and the simulation training was carried out. The evaluation results are basically consistent with the expected results, which indicate the operability and practicability of the resource allocation evaluation system of ideological and political education in colleges and universities.