Multivariate data prospecting prediction system based on machine learning

The invention discloses a multivariate data prospecting prediction system based on machine learning, and aims to solve the problem of prospecting difficulty. Comprising a first mineral deposit model construction module, a second spatial database construction module, a third data standardization module, a fourth prediction model training data distribution module, a fifth mineral prospecting prediction model construction module based on machine learning, a sixth algorithm optimization and parameter adjustment module and a seventh potential area analysis and target area delineation module. The method can be used for sequentially optimizing a favorable metallogenic layer, comprehensively processing multivariate information, establishing a multivariate spatial database, standardizing and gridding database data, and carrying out multi-model multi-parameter metallogenic potential analysis based on an artificial neural network and a support vector machine of machine learning. The ROC subject characteristic curve is u