The paper introduces a Structured Dependency Tree-based Graph Convolutional Network (SDTGCN) for aspect-level sentiment analysis, aiming to enhance the precision of predicting sentiment polarities of specific aspects in sentences. The SDTGCN model addresses the limitations of existing graph neural network models by incorporating positional information, sentiment commonsense knowledge, part-of-speech tags, and syntactic dependency distances to construct a structured dependency tree. This structured tree enriches the generic dependency tree, allowing for more accurate extraction of relationships between aspects and opinion words. The model uses a weighted graph convolutional network to aggregate node information, assigning higher weights to pivotal nodes and lower weights to irrelevant nodes. Experimental results on five public datasets demonstrate the effectiveness of the proposed model, showing significant improvements in accuracy and F1-score compared to state-of-the-art approaches. The study also includes ablation experiments and a case study to validate the contributions of different components of the SDTGCN model.The paper introduces a Structured Dependency Tree-based Graph Convolutional Network (SDTGCN) for aspect-level sentiment analysis, aiming to enhance the precision of predicting sentiment polarities of specific aspects in sentences. The SDTGCN model addresses the limitations of existing graph neural network models by incorporating positional information, sentiment commonsense knowledge, part-of-speech tags, and syntactic dependency distances to construct a structured dependency tree. This structured tree enriches the generic dependency tree, allowing for more accurate extraction of relationships between aspects and opinion words. The model uses a weighted graph convolutional network to aggregate node information, assigning higher weights to pivotal nodes and lower weights to irrelevant nodes. Experimental results on five public datasets demonstrate the effectiveness of the proposed model, showing significant improvements in accuracy and F1-score compared to state-of-the-art approaches. The study also includes ablation experiments and a case study to validate the contributions of different components of the SDTGCN model.