An Advanced Decision Tree-Based Deep Neural Network in Nonlinear Data Classification

Deep neural networks (DNNs), the integration of (NNs) and deep learning (DL), have proven highly efficient in executing numerous complex tasks, such as data image classification. Because multilayer a nonlinearly separable structure is not transparent, it critical to develop specific classification model from new unexpected dataset. In this paper, we propose novel approach using concepts DNN decision tree (DT) for classifying nonlinear data. We first developed tree-based network (DTBNN) model. Next, extend our (DTBDNN), which multiple hidden layers are utilized. Using DNN, DTBDNN achieved higher accuracy compared related relevant approaches. Our proposal achieves optimal trainable weights bias build an by combining benefits DT NN. By conducting in-depth performance evaluations, demonstrate effectiveness feasibility achieving good over different datasets.