Esophageal cancer (EC) is a very aggressive tumor which typically appears in the esophagus, ranks among the most prevalent cancers worldwide. It is also a major contributor to cancer-related mortality. The two predominant histological types are esophageal squamous cell carcinoma (ESCC) and esophageal adenocarcinoma (EAC). The signs of the illness mainly appear at later stages and include trouble swallowing, chest pain, weight loss, and coughing. Early classification significantly improves treatment success and reduces mortality. In the last decade, machine learning has gained tremendous attention and has made a deeper impact in cancer research, offering promising advancements in prognosis prediction, treatment efficacy, and patient stratification. This study applies various Machine Learning techniques including Ensemble models, Neural Networks Trees, SVM to classify esophageal cancer. The dataset includes oncology patient data such as tumor site (ICD-O-3), histology, ICD-10 classification, tissue collection indicators, age, gender, height, and weight. Performance metrics of all models are compared, with Neural Networks and Ensemble Learning achieving about 99.7% validation accuracy in esophageal cancer classification.