A Survey Convolutional Neural Networks Using SPSS Statistics

Convolutional Neural Networks (CNNs) are a deep learning architecture that has revolutionized computer vision tasks. Their biological inspiration traces back to the seminal work by Hubel and Wiesel in 1959, who unveiled the specialized cells in the visual cortex responsible for light differentiation. This led to Fukushima’s creation of the Neocognitron in 1980, a precursor to modern CNNs. Current CNNs excel in various domains, notably radiology, by utilizing layers such as convolutional, pooling, and fully connected ones. These layers facilitate the adaptive learning of spatial feature hierarchies through backpropagation. In this study, we employ IBM’s SPSS Statistics for robust data analysis, multivariate testing, and advanced predictive modeling to assess CNN architectures. SPSS excels in data management and outcome production, adapting to various data formats. The evaluation metrics include the Loss function, Regularization, Optimization, Fast processing, and practical Applications. The reliability of the CNN model was gauged using Cronbach’s Alpha, yielding a value of .442, indicating a 44% reliability rate. Compared to the literature threshold of 39%, our model is deemed suitable for further analysis.