Comparison of PCA and LDA Dimensionality Reduction Algorithms based on Wine Dataset

In the construction of machine learning model, when the input data dimension is too large and the data characteristics are particularly complex, the complexity of the model will increase, especially when some sample data is insufficient, which will lead to the poor generalization of the training model. Therefore, it is necessary to remove some redundant features by dimensionality reduction to reduce the dimension of data, so as to facilitate the observation and mining of information. This paper uses the red wine data set in Python to reduce the dimension of PCA and LDA, and on the basis of the existing research, compares the dimension reduction of red wine data set before and after standardization, puts forward the characteristics of PCA dimension reduction and LDA dimension reduction, and the similarities and differences between the two linear dimension reduction methods, so as to provide ideas for the subsequent data processing.