User interest analysis method based on e-commerce big data

The invention discloses a user interest analysis method based on e-commerce big data, and the method comprises the following steps: S1, constructing a data model of a user based on the type score through the e-commerce big data, determining a type set in which the user is interested according to the type score, and carrying out the scoring of a commodity containing a preference type through the preference type of the user; s2, based on the score vector, recording the user purchase behavior as an event vector occurring according to a time sequence, rearranging scores of each type according to a time sequence direction, and establishing a score matrix according to the scores of each type; and S3, variance fluctuation characteristics generated in different stages are adjusted through a time attenuation factor, user interest data trend characteristics are determined according to convergence degrees of the variance fluctuation characteristics, and prediction accuracy can be effectively improved by combining time s