Waste mobile phone dynamic pricing method based on M-WU concept drift detection

Abstract

The invention discloses a waste mobile phone dynamic pricing method based on M-WU concept drift detection, and belongs to the field of machine learning and data mining. According to the method, an i-cat decision tree is adopted as a dynamic pricing model, and concept drift detection is carried out by adopting an M-WU method. Firstly, key features influencing pricing are selected from the model by adopting a principal component analysis method; secondly, an incremental learning model based on a decision tree is established, and dynamic pricing of the waste mobile phone is achieved; and finally, designing a concept drift detection algorithm based on M-WU test to realize drift detection of sample data. According to the invention, the pricing accuracy of the waste mobile phone can be well improved.

Description

technical field [0001] The invention belongs to the field of machine learning and data mining, in particular to a dynamic pricing method based on M-WU concept drift detection. Background technique [0002] In recent years, as consumers replace mobile phones more and more frequently, how to dispose of used mobile phones has become a topic of increasing concern. The recycling of used mobile phones can promote a virtuous circle of sustainable economic and environmental development. However, more and more used mobile phones on the market make manual pricing one by one unrealistic. Fortunately, with the development of the Internet and big data technology, data mining technology-based The intelligent pricing mechanism has become a new business model, which is conducive to promoting the formation of a win-win situation for consumers and enterprises. [0003] The current mainstream pricing methods are mainly static pricing methods. The data used in the training set is often all his...

AI Technical Summary

Problems solved by technology

However, the dynamic pricing process for new data will face a problem, that is, in real market conditions, the mapping relationship between commodity prices and their influencing factors will fluctuate with market changes, when the target variable changes over time The model built to analyze such data will become outdated over time. This phenomenon is called concept drift, and this phenomenon is especially reflected in mobile phones, which are updated quickly. Obviously, it is a difficult problem in the dynamic pricing problem of commodities

Images