Through the prediction of the survivors of the Titanic, a project actual combat was carried out, and in the process of completing the project, through their own thinking and reference to information, they had a clearer understanding of the entire application process of machine learning algorithms, and also made themselves further familiar with the relevant operations of visual libraries such as seaborn and matpliotlib, further deepened their understanding of BP neural network algorithms, enriched their own knowledge system, and also made a good reflection on themselves.
Content:
(1) The first step is to load the training data, overview the data, and analyze the collective details of the data.
(2) Data preprocessing, and deal with the outliers and missing values of the data.
(3) Analyze the relationship between different attributes and survival rate, and determine the characteristics required for prediction
(4) Normalization and numerical conversion of the identified features
(5) Put it into the BP neural network for training prediction
