A Kernel Based Support Vector Machine Approach: Predicting Psychiatric Disorders Using Classified Psychiatric Features

Psychiatry, a medical discipline is a current public health concern for worldwide. One's mental health is influenced by several elements that are simple to measure. The physical and financial well-being of a person may be impacted by a variety of variables. The risk of acquiring a mental illness rises with psychiatric treatment. The severe mental illness schizophrenia affects both men and women equally, though more commonly. Mentally ill people are more likely to act antisocially and influence the behavior of others. Social problems that were already noticeable have gotten much worse as a result. International surveys show that the major mental traits of anxiety, substance abuse, hazardous conduct, arrogance, suicidal thoughts, sadness, disorientation, and awareness are quite prevalent among individuals between the ages of 20 and 30. A presented questionnaire survey is utilized to collect data, which was a challenging undertaking. Based on the application of nonlinear kernel Support Vector Machine (SVM) classifiers to the constructed dataset of mental patients, we have predicted a machine learning classifier that will classify the relevant extracted characteristics. The primary conclusion of this work is the recommendation of polynomial kernel SVM as a classifier to predict the psychiatric disease of each psychiatric patient, taking into account all performance criteria.