Cognitive load (CL) refers to the mental effort required for a specific task. This paper presents a method for predicting cognitive load solely based on the Galvanic Skin Response (GSR) and includes a feature analysis of GSR signals. GSR, which captures changes in skin conductance as a measure of sympathetic nervous system activation, is directly proportional to cognitive load. This work aims to determine whether an individual is experiencing a cognitive load. The cognitive load, affect, and stress recognition (CLAS) dataset, which is interdisciplinary and multilayered, is used for this purpose. The analysis explores the time domain, frequency domain, and peak-related features, aiming to classify various cognitive load states, whether the load is present or absent. Several supervised learning algorithms were employed for classification, achieving a peak precision of 100% and 99.17% with the SVM and Random Forest, respectively. The findings of this research contribute to the development of noninvasive methods for evaluating cognitive load instantaneously, which can improve interactions between humans and computers as well as learning environments.