This paper presents an improved Fuzzy-weighted Convolutional Neural Network (F-CNN) for taxonomic classification of bacterial DNA sequences, specifically focusing on the 500-bp segments. The proposed model aims to overcome the limitations of traditional classification methods by leveraging the power of deep learning and fuzzy logic processing. The improved fuzzy deep learning model is proposed to handle the problem of classifying samples with similar probabilities in the classification layer. It incorporates a feature selection stage using various techniques and a fuzzy weighting system to handle the uncertainty associated with similar classes in the classification layer and optimize parameters using fuzzy weights. The experimental results on the Ribosomal Database Project Release 11 (RDP 11) sequences dataset show the superiority of the proposed model, especially at the 500-bp region.

Source: CNN December 23, 2025 18:00 UTC