Research on random forest drug classification prediction model based on KMeans-SMOTE

Xiaoyu Song; Yifan Li; Hongyang Wu

doi:10.1117/12.2660089

6 December 2022 Research on random forest drug classification prediction model based on KMeans-SMOTE

Xiaoyu Song, Yifan Li, Hongyang Wu

Proceedings Volume 12458, International Conference on Biomedical and Intelligent Systems (IC-BIS 2022); 124581S (2022) https://doi.org/10.1117/12.2660089
Event: International Conference on Biomedical and Intelligent Systems, 2022, Chengdu, China

Abstract

Scientific medication is of great significance to improve the therapeutic effect of diseases. The etiology of chronic diseases is complex, different patients have different responses to drugs. Analyzing the examination results and medication information of previous patients can make the treatment plan follow a regular basis. First, the existing medical data is used to visualize the examination results of previous patients and the drugs used, to find the corresponding relationship between the two, and to build new attributes based on the attributes of the original data to improve the accuracy. Then, three oversampling algorithms are used to balance the imbalanced properties of the dataset. Finally, five kinds of classifiers were used to build a drug prediction model and conduct comparative experiments. The results showed that the random forest model based on the KMeans-SMOTE showed good results in various evaluations, and the drug prediction accuracy rate reached 95.05%. The model can better predict the drugs needed by patients, and its performance is better than other established models.

Citation Download Citation

Xiaoyu Song, Yifan Li, and Hongyang Wu "Research on random forest drug classification prediction model based on KMeans-SMOTE", Proc. SPIE 12458, International Conference on Biomedical and Intelligent Systems (IC-BIS 2022), 124581S (6 December 2022); https://doi.org/10.1117/12.2660089

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available