<p>In the contemporary era, the use of computing-based tools and automated systems has become indispensable across various domains. These advancements have enabled humans to leverage the capabilities of such systems to enhance quality and efficiency in numerous fields. Artificial intelligence (AI), recognized as a dynamic and integrated system, has expanded its influence across diverse scientific disciplines. This widespread adoption has led to a significant increase in demand from both users and organizations. Among the many branches of AI, natural language processing (NLP)—which focuses on the analysis of textual and auditory data—has seen remarkable growth over the past two decades. This growth is driven by the explosive generation of text data, coupled with advancements in processing power, memory resources, and the development of novel concepts and architectures such as transformers, attention mechanisms, and large linguistic models. Consequently, NLP has garnered significant attention within the research and development communities. At the intersection of AI &amp; law, this study introduces a novel framework for aspect-based sentiment analysis (ABSA) in legal contexts. Specifically, we propose a framework based on the T5 language model, trained to analyze the sentiments of parties involved in legal cases using the SigmaLaw-ABSA dataset. Our framework classifies the sentiments of the parties into three categories: positive, neutral, and negative. This enables the distinction between constructive and inefficient sentiments within legal texts, which can facilitate a more effective and efficient legal hearing process. To assess the performance of our proposed model, we evaluate it using accuracy and F1-Score metrics after fine-tuning it on a processed training dataset. The results demonstrate promising performance, with an accuracy of 82.83% and an F1-Score of 81.45%. These outcomes suggest that our model achieves a high level of accuracy, positioning it as a leading approach in this area of research.</p>