Feasibility of designing a cloud-based platform for entrepreneurship training of college students

This paper aims to improve college students’ entrepreneurial abilities as its main goal. A platform for college students’ entrepreneurship development is constructed using cloud technology, and the entrepreneurship data on the platform is classified and divided by combining it with the KNN algorithm. The algorithm is optimized for distance calculation and distance sorting stages to reduce inter-program dependencies and shorten the program execution time. The distance between matrices is calculated by calling the functions in the cublas function library. The experimental data are used to analyze the feasibility of entrepreneurship training for college students. The results show that the KNN algorithm can be applied to data processing in cloud computing. When the data dimension is 128 and the number of sample points is 215, the algorithm can get a maximum speedup of 1.52 times. The server configuration of the platform needs to be in a 4-way dual-core or 4-core to ensure the platform’s normal operation. This study is beneficial in promoting the improvement and development of entrepreneurial skills for college students.