Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental disorder that affects children and adolescents. It is characterized by symptoms of inattention, hyperactivity, and impulsivity. Currently, the diagnosis of patients with ADHD primarily relies on behavioral observations combined with educational and psychological assessments. This makes early detection and timely diagnosis of ADHD more challenging. However, recent findings have shown that electroencephalography (EEG) signals can provide valuable insights into cognitive abilities and have demonstrated promising potential in early ADHD diagnosis. In this study, EEG signals were utilized in conjunction with several machine learning algorithms to evaluate the effectiveness of these algorithms and their potential for improving the early detection of ADHD. A public EEG dataset of ADHD patients from an open-access database has been used. At first, the preprocessing is carried out through some data wrangling techniques. The next step involves feature selection, where each frequency channel band is considered a feature identified. Finally, the classification will be performed using Support Vector Machine (SVM) models and Linear Discriminant Analysis (LDA) algorithm, to detect ADHD. The results showed that the SVM -based classification model provided more accurate and reliable results (with 96.42% accuracy, 95.77% sensitivity, 96.88% specificity) than the LDA-based model (with 63.92%accuracy, 25.37% sensitivity, 91.33% specificity). These findings will inform future research as the authors seek to develop an effective and accurate tool for early diagnosis of ADHD. Such a tool could help families and healthcare professionals intervene and treat the disorder in a timely manner, leading to better outcomes, especially for children.