Electromyography (EMG) is employed to measure electrical activity in muscles, offering crucial insights for the diagnosis of neuromuscular disorders, including amyotrophic lateral sclerosis (ALS). ALS is a progressive condition that leads to the deterioration and damage of motor neurons. This research employs a machine learning approach to differentiate between ALS patients and healthy subjects by examining surface EMG signals. The study extracts 40 features from raw sEMG signals, combining time and frequency-based characteristics. To enhance classifier performance, these features are then reduced to an optimal number using the sequential feature selection (SFS) technique. The refined feature set is subsequently applied to artificial neural network (ANN) and random forest classifier (RFC) for classification purposes. The results of the experiments show that using the reduced feature set with both ANN and RFC models significantly improves classification accuracy compared to previous studies. Specifically, the ANN model achieved an accuracy of 95.3%, while the random forest classifier attained an accuracy of 92.8%.