Author(s)

Wei Yang, Junli Zhang and Xiangxiu Yao

Corresponding Author

Wei Yang

ABSTRACT

The biggest difficulty in targeted poverty alleviation on campus is the difficulty in accurately identifying poor students. This paper makes full use of campus card consumption data, on the basis of building a poverty indicator system, first uses PCA to screen the indicators, and then combines the adaboosting algorithm and the logistic regression algorithm to build a poor student identification model. Finally, the validity of the model is verified by the 2019 student consumption data of a college.

KEYWORDS

Consumption data, Indicator construction, Identification of poor students