<p>Text mining provides a method within the domain of biological and medical research, to facilitate research advancements. This paper utilizes the state-of-the-art text extraction techniques to perform an in-depth analysis of an extensive corpus of research articles focusing on a spectrum of critical medical conditions, including Alzheimer’s, diabetes, cancer, asthma, Fabry, and various syndromes. The presented analysis of this work is mainly focused on abstracts of research papers, affording the capability to extract vital genetic information linked to these diseases. The data set used is meticulously and manually curated, marking the inaugural phase of this pioneering pilot study. Combining tree-based models with neural networks helps the advanced machine learning algorithm create a robust trained architecture, resulting in the new technique being seamlessly optimized with existing models. This unique fusion enables the extraction of profound insights from textual data on complex medical conditions to identify and classify genetic diseases. To evaluate the efficacy of the proposed model, a benchmark is performed against 17 alternative machine learning models. The outcome of the proposed model achieves an accuracy rate of 98%. This represents a substantial improvement compared to the majority of benchmark techniques, which traditionally achieve an accuracy rate less than 96%. This paper underscores the immense potential and significance of the approach presented in the field of text mining for biological and medical research.</p>