Classification and Prediction of Multi-Class DNA Sequences Utilizing Artificial Intelligence Technology

Multi-class DNA sequence classification is of critical importance in bioinformatics, as it facilitates the accurate identification of species, gene functions, and disease-related variations. Conventional laboratories and alignment-based methodologies frequently require extended periods and are computationally onerous when applied to large-scale genomic datasets. This study proposes a set of propositions regarding the employment of artificial intelligence techniques for the classification and prediction of multi-class DNA sequences, with a view to achieving both efficiency and accuracy. This research includes an experimental component in which the techniques are applied to a dataset consisting of 10 classes of DNA. The DNA sequences obtained from random genomic files were subjected to preprocessing and transformed into numerical representations. This was achieved by employing feature engineering techniques, such as k-mer frequency extraction and one-hot encoding. Therefore, in this study, classification models, including the Random Forest algorithm, Support Vector Machine (SVM), and Convolutional Neural Network (CNN), were implemented and evaluated. Furthermore, this study proposed and experimented with CNN and SVM algorithms as hybrid models. The performance of each model was evaluated using standard evaluation metrics, including accuracy, precision, recall, and F1-score. The findings emphasize the efficacy of AI-driven methodologies in genomic data analysis, offering a scalable technique for the classification and prediction of multi-class DNA. This study contributes to advancements in bioinformatics, supporting applications in disease diagnosis, evolutionary studies, and personalized medicine.