A Real-Time Feedback Approach Based on Semantic Clustering for Poverty Alleviation Problem

Abstract

Poverty has always been one of the most acute social problems in the world. In order to eradicate poverty, the Chinese government has invested a lot of manpower and material resources, and promised to lead all poor areas and poor people into a well-off society by 2020. In this process, the county government as the main battlefield of poverty alleviation, the complex work of poverty alleviation has brought considerable pressure to grass-roots cadres and helping cadres. For the sake of improving the efficiency of support work, we propose a real-time feedback approach based on semantic clustering for poverty alleviation problem, through which we can build a bridge between grass-roots cadres and decision makers. In this method, we first use a fast label extraction method to quickly extract important feature words from the complicated help information. Secondly, we use unsupervised text clustering method to identify important poverty alleviation problems from these feature words, so as to provide a reference for the poverty alleviation workers to carry out their work in an orderly and targeted manner. The experimental results for different regions show that the poverty alleviation problem identified by our proposed method can reflect regional characteristics.