Construction and Analysis of a Data-Driven T2-SMOTE-GBDT Prediction Model for Vascular Vertigo

Abstract

At present, vascular vertigo has become a high incidence. Still, traditional diagnosis and treatment methods are highly dependent on the clinical experience of medical workers, and are prone to misdiagnosis, missed diagnosis, and excessive diagnosis and treatment, resulting in a waste of medical resources. Based on the popular artificial intelligence algorithm, this paper constructs a clinical prediction model of vascular vertigo based on data-driven and machine learning. In this paper, the missing values and outliers in the data samples are first processed and normalized. Then two different algorithms, Pierre coefficient and random forest, are used for feature selection. Then Smote algorithm is used to generate data to solve the data imbalance problem. Finally, we use the hyperopt algorithm to optimize the parameters of the GBDT model. Compared with other models, the prediction model proposed in this paper is better than other models.