Commodity purchase prediction modeling method

Abstract

The invention discloses a commodity purchase prediction modeling method. The method comprises the steps that a purchase record marking training sample is used to predict whether to purchase or not; a sliding window commodity purchase sample is constructed; commodity purchase features are designed based on a time preference; a gradient improvement decision tree algorithm is used for training prediction; after the sample and the features are constructed, feature processing and selection need to be performed, and then the features are input into the gradient improvement decision tree algorithm for training prediction; and feature selection indicators include feature value distribution and relevancy, feature information gains, feature calling frequency, influences of feature knockout, etc. Ordering is performed on feature importance by integrating the indicators, and redundant features with low importance are eliminated. According to the method, a sliding window sample construction method and a feature system based on the time preference are proposed, the accuracy of a commodity purchase prediction model is effectively improved, and the method is used for realizing commodity personalized recommendation in a big data background to precisely recommend proper commodities to a user at a proper time and a proper place.

Description

technical field [0001] The invention relates to a machine learning modeling method, in particular to a commodity purchase forecasting modeling method, which belongs to the technical field of artificial intelligence. Background technique [0002] Forecasting is to mine past behavioral data and predict the development of future events based on reliable calculation and deduction. The general process of time series forecasting is to first determine the destination and target of the forecast, then select the appropriate forecast period and method, then collect the data that may be used in the forecasting process, and finally make the forecast through rigorous analysis. [0003] According to the distance from the predicted target, it can be simply divided into the following categories. Short-term forecasts are generally within hours or days, short-term forecasts are weeks or months, medium-term forecasts are generally no more than 5 years, and long-term forecasts are generally mo...

AI Technical Summary

Problems solved by technology

Similarly, this method will also lose some periodic laws, and it is impossible to grasp the local changes in detail

Images