Research on Precision Marketing Based on Consumer Portrait from the Perspective of Machine Learning

With the advent of the big data era and the development of artificial intelligence, the traditional marketing model has undergone dramatic changes, and precision marketing has become the focus of current research. Consumer portrait is a labeled consumer model abstracted from the basic attributes, social attributes, behavior characteristics, and psychological characteristics of consumers, which can reflect the real needs of consumers well. Precision marketing based on consumer portraits is a fine division of consumers on the premise of in-depth understanding of consumer-related data. Based on the public dataset of a bank in a certain region, this paper extracts a series of data features such as basic attributes and social attributes of consumers to draw consumer portraits. BP neural network, SVM, and random forest algorithm in machine learning algorithm are used to model and predict whether consumers should take out personal loans, and then, the prediction effect of these algorithms is compared to judge which algorithm has better prediction effect. The results show that the random forest algorithm has the highest prediction accuracy, reaching an average of 94% in 10 calculations, which can help banks to select potential target customers to some extent.