The Amalgamation of Deep Learning Model and Conventional Statistical Model

With the development of current society and the expanding population, deep learning models start to show their advantages over other statistical models--especially the conventional statistical model (CMSs), like the applications of normal distribution. This research mainly discusses the pros and cons of deep learning model compared to conventional statistical models and how to take advantage of both models to obtain a more comprehensive model. In the current society, with the rapid scientific development, research often requires serious, precise, and accurate data and analysis to achieve perfect results. This is what researchers want to see. Ideally, when experimenters amalgamate these two models--CSMs and deep learning model- some data analysis will become more precise than people previously thought. This is the purpose of this research wanted. By observing the pros and cons of these two models and making a comparison between them, finding out the previous cases of applications of the specific model used, as well as the analysis of the data provided by the references, this research finds that, when people take both advantages of these two models, the results offered by these models will be more accurate and less biased, regardless of whether the prediction of the data or the regression line is used to find out the relationship between the x value and y value.