SIMILARITY MODEL-BASED DATA PROCESSING METHOD AND SYSTEM

摘要

A similarity model-based data processing method and system, which may effectively improve the conversion rate of customers at reduced costs by using similarity model-based data processing technical means. The method comprises: collecting a plurality of customer data; extracting continuous label data from each piece of customer data, and obtaining multiple groups of discrete label data after binning conversion; calculating the similarity distance for discrete factors in each group of discrete label data, while screening out multiple groups of new discrete label data consisting of discrete factors which contribute significantly; calculating the weight for the discrete factors in the new discrete label data respectively by using the random forest algorithm and the gradient boosting decision tree algorithm, and obtaining weighted results of multiple groups of discrete factors after weighted summation; and calculating the final similarity distance between each piece of customer data and positive sample data respectively by using the Manhattan distance algorithm according to the weighted result of each group of discrete factors and the similarity distance of each discrete factor.