Big data feature extraction method and system based on deep learning

The invention provides a big data feature extraction method and system based on deep learning, and belongs to the technical field of data processing, and the method comprises the following steps: S1, obtaining original data, and carrying out the preprocessing of the original data through employing a big data preprocessing algorithm; s2, acquiring data features of the original data by using a deep learning model, and performing correlation evaluation analysis on the data features to obtain an evaluation analysis result; s3, processing an evaluation and analysis result to obtain a feature subset; and S4, collecting feature subsets of different sources and different types of data features, and carrying out dimension reduction analysis processing on the feature subsets to generate a dimension reduction analysis result. According to the method, dimension reduction analysis processing is carried out by using a generalized eigenvalue problem, the distinguishability of data after dimension reduction and the capabilit