Energy drinks are widely consumed by teenagers and young adults to enhance focus and energy.

However, excessive consumption may lead to negative health effects, especially when they contain

preservatives like sodium benzoate. In this study, the concentration of sodium benzoate was

determined in four imported energy drink samples available in Asb'ah markets. The analysis was

conducted using a UV spectrophotometric method, which proved to be both simple and accurate. The

concentration of sodium benzoate in the samples ranged from 1361 to 1424.55 mg/L. The results

indicated that all tested samples contained sodium benzoate levels within the upper safety limits

established by the FDA

Abstract: In light of the aspiration to achieve sustainability in the field of education, especially after the development of artificial intelligence technologies in general and artificial neural network technologies in particular, the process of improving e-learning has become one of the most important strategies for achieving sustainability. This study aims to evaluate the impact of using artificial intelligence techniques and artificial neural networks on improving cloud e-learning by extracting and mining data from the electronic cloud and analyzing it to provide a more efficient and effective educational experience. Through several methodologies, including descriptive methodology to describe the factors influencing the improvement of machine learning through the use of artificial intelligence techniques and the mechanism for using these techniques, quantitative methodology was used in collecting data and analytical scientific methodology to analyze the results of the proposed model, which is a hybrid of convolutional artificial neural networks (CNN) and recursive neural networks (RNN). Through the model, it is possible to extract and analyze user data such as interaction with content and duration of study to extract patterns and behaviors that support improving the learning experience, and personalizing individual learning paths. Predicting the academic performance of students using machine learning techniques. The results showed an improvement in the learning experience of 86% and an improvement in platform performance of 85%. Machine learning indicators, such as accuracy (94%), recall (93%) and f1 score (91%), also indicated the success of the model used in