Experiencing Machine learning for identifying the immunomodulator character of the medicinal plants

Medicinal plants are well known for their immunomodulatory properties, providing potential therapeutic applications in infectious diseases, autoimmune disorders, and cancer. However, determining and identifying these properties using conventional experimental approaches is often time-consuming, labor-intensive, and costly. This paper aims to characterize the immunomodulator of the medicinal plant by using machine learning ML as a new approach. To achieve this goal more than 200 medicinal plants collected from the literature has been used. Methods such as, support vector machines (SVM), random forests (RF), Neighbor classifier grouped on pipeline have been explored to predict immune-related activities of these plants based on their immuno properties. The different used methods of ML, after being well optimized have successfully identified key immunomodulatory compounds in medicinal plants with 0.80% accuracy. The validation of our result has been carried out based on the experimental process of certain medicinal plants in our laboratory. This result highlights the critical role of ML in immunomodulatory research as a new approach for identifying, the immunomodulatory character of medicinal plants.