Sleep apnea, a major sleep disorder, causes serious health risks and lowers the quality of life for millions of individuals worldwide. In this paper, we investigate the deep learning methods for classification of sleep apnea. Our study explores Sleep Health and Lifestyle Dataset which contains 560 patient’s records, each of which is described by 13 variables relating to sleep patterns, lifestyle, and health. We trained and tested effectiveness of various classifiers for sleep disorder prediction considering their performance measures. We tested Multilayer Perceptron (MLP) neural networks, Naive Bayes, Random Forest, K-Nearest Neighbors (K-NN). Notably, the MLP model outperforms other models by achieving accuracy of 92.14% and Kappa coefficient of 0.8387 in classifying sleep apnea. Whereas accuracies achieved by Naive Bayes, k-NN, and Random Forest models are 88.8%, 90.7%, and 91.9% respectively. Surely MLP beats other algorithms due to its ability to grasp complicated correlations within the data. Our findings demonstrate the utility of neural networks, specifically MLPs, in sleep apnea detection.