A matching model for major orientation of computer science students based on random forest model

Abstract

With the advancement of computer innovation, there are increasingly major computer bearings, such as software development, network security, artificial intelligence, data science, etc. When choosing the direction of computer science, students have to consider their interests, techniques, soft skills and numerous other components to utilize their qualities and accomplish their career objectives. This article centres on how to use machine learning methods to anticipate the direction of computer science students. Theres a solid relationship between students interests, techniques, and soft skills when choosing the direction of computer science majors. By employing a random forest machine learning show to prepare students ability characteristics, it is conceivable to anticipate which direction of specialization students are more appropriate for. Within the explore, we isolated the training, validation, and test sets, agreeing to the proportion of 6:2:2. We utilized the exactness rate to judge the models forecast exactness. After preparing, the expectation precision of this show increments, and the precision rate of the test set comes to 96%. This implies that this machine learning show can reasonably foresee computer science students heading and give them way better career improvement proposals. In the future, we will encourage optimising the demonstration to make strides in the forecast exactness and apply it to a broader run of areas.