Monitoring and Evaluation of Talent Cultivation Quality Based on Big Data

The quality of graduates is directly related to the social prestige and growth of universities. The author uses a random forest algorithm based on big data processing to evaluate the performance of advanced students to create learning skills based on the original data story of some university computer graduates. Before training the classifier, evaluate the importance of the features using the RF ranking method, and select 75 features to reduce the dimensionality to improve the probability of the training model; by training the distribution center and evaluating each employee’s performance, each individual is assessed on their strengths to mitigate the impact of poor performance scores. It is found that the precision and yield of the improved random forest algorithm are not significantly different from those of other classic modified random forest algorithms. However, it has some improvements in the actual design standards. This confirms that the algorithm can improve the accuracy and precision of plant quality assessment and has a practical role in planting in universities.