Extrapyramidal symptoms (EPS), including Drug-induced Parkinsonism, Acute Dystonia, Akathisia, and Tardive Dyskinesia, are prevalent side effects of antipsychotic therapy in schizophrenia patients, presenting significant challenges to patient well-being and care. It is essential to identify demographic and clinical factors associated with EPS to optimize treatment strategies. This study aims to integrate statistical analysis with machine learning techniques to predict the onset and severity of EPS in patients with schizophrenia. A prevalence study was conducted among schizophrenia patients receiving antipsychotic therapy, wherein patient records were examined for EPS and it's associated factors. Statistical tests, factor analysis, and partial correlation networks were utilized to identify contributors to EPS. Additionally, machine learning algorithms were applied to predict EPS based on demographic, clinical, and medication-related variables. The study revealed an estimated EPS prevalence of 38.3% among the patient cohort. Factors influencing EPS included the number of First-Generation Antipsychotics (FGAs), Second-Generation Antipsychotics (SGAs) prescribed, family history, and patient physical characteristics. Partial correlation networks highlighted correlations between body mass index (BMI), Chlorpromazine Equivalent Daily Dose, and duration of schizophrenia. Machine learning model demonstrated promising accuracy in predicting EPS likelihood and severity. These findings provide valuable insights for optimizing antipsychotic therapy and reducing EPS burden on the healthcare system.Distribution of sociodemographic and clinical variables in our study (a) Distribution of key sociodemographics factors such as age, locality, heredity,gender, marital Status, employment status (b) The clinical variables such as smoking habits,appetite, sleep, comorbidity, built, type of antipsychotics among study participants.