| 摘要 |
A method and system for classifying users of an e-commerce platform. The method comprises: determining high-level commodities from commodities in the same class based on price and sale volume distribution, determining the levels of other commodities in the same class as non-high levels, and sequentially dividing the non-high-level commodities into x levels from high to low in a high-to-low price order (S101), wherein x is a pre-set natural number greater than or equal to one; calculating a ratio of the commodities at each level to the total commodities purchased by each user, and vectorising the ratio based on each user to obtain a purchasing power vector of each user, the purchasing power vectors being x + 1-dimension vectors, and each dimension corresponding to a level (S102); and clustering the purchasing power vectors to obtain x + 1 point clusters about the purchasing power vectors, each point cluster corresponding to a level, and the level corresponding to the point cluster where the corresponding purchasing power vector of each user is located serving as the purchasing power level of the user (S103). The method more accurately classifies user purchasing power. |
