RNN and Logistic Regression-Based Ensemble Model for Distinguishing AI-Generated Text

Generative AI models change quickly, highlighting crucial concerns such as the difficulty of distinguishing between human-written and AI-generated texts. This research proposes a unique ensemble model that combines a Recurrent Neural Network and Logistic Regression to efficiently address the differentiation difficulty. The RNN is used to sequence-model complicated patterns in text input, whereas Logistic Regression is used for classification via dense feature representations. The model was trained and evaluated using a publicly available dataset that contained both AI-generated and human-written material. Key performance indicators such as accuracy, precision-recall, and ROC-AUC demonstrate the ensemble model’s ability to accurately classify text. In order to validate the findings, this study also includes intricate visualizations like precision-recall curves and confusion matrices. Future developments in related content authentication and regulatory frameworks will be made possible by this combined version of deep learning with the conventional machine learning end, which strengthens the scalable and reliable process to recognize AI-generated text.