This study, an extension of the "Smart Computing Models, Sensors, and Early Diagnostic Speech and Language Deficiencies Indicators in Child Communication" project (SmartSpeech), aims to enhance the machine learning model's efficacy in predicting neurodevelopmental disorders (NDs) in children. The primary goal is to assist clinicians in identifying NDs through a serious game that evaluates children's developmental profiles. The study uses a comprehensive machine learning approach, leveraging a recently developed serious game dataset that collects various data on children's speech and linguistic responses. After preprocessing, the dataset was reduced from 520 instances to 473 participants. Cluster analysis revealed distinct patterns, and reliability analysis ensured measurement consistency. A logistic regression model was developed, achieving high accuracy, precision, recall, and F1-score when applied to a subset of 184 participants with an average age of 7 years. The model effectively distinguished between instances with and without NDs, highlighting the effectiveness of machine learning in diagnosing NDs based on cognitive features. This research advances our understanding of NDs and offers new opportunities for decision-making, classification, and clinical assessment, paving the way for early and personalized interventions.This study, an extension of the "Smart Computing Models, Sensors, and Early Diagnostic Speech and Language Deficiencies Indicators in Child Communication" project (SmartSpeech), aims to enhance the machine learning model's efficacy in predicting neurodevelopmental disorders (NDs) in children. The primary goal is to assist clinicians in identifying NDs through a serious game that evaluates children's developmental profiles. The study uses a comprehensive machine learning approach, leveraging a recently developed serious game dataset that collects various data on children's speech and linguistic responses. After preprocessing, the dataset was reduced from 520 instances to 473 participants. Cluster analysis revealed distinct patterns, and reliability analysis ensured measurement consistency. A logistic regression model was developed, achieving high accuracy, precision, recall, and F1-score when applied to a subset of 184 participants with an average age of 7 years. The model effectively distinguished between instances with and without NDs, highlighting the effectiveness of machine learning in diagnosing NDs based on cognitive features. This research advances our understanding of NDs and offers new opportunities for decision-making, classification, and clinical assessment, paving the way for early and personalized interventions.