Research On The Application Of Data Mining In Differentiated Assessment And Precise Assistance Of Targeted Poverty Alleviation

Targeted poverty alleviation is one of the three major battles of General Secretary Xi in the report of the 19th National Congress of the Communist Party of China,which is a long-term and arduous task.Poverty has long-term and dynamic characteristics.Establishing and improving a long-term mechanism for stable poverty alleviation,and searching for precise poverty alleviation paths in various regions,are of great significance for promoting the economic and social sustainable development of poverty-stricken areas and alleviating relative poverty after 2020.As the work of precision poverty alleviation gradually deepens,the problem of how to implement accurate measures according to the different causes and conditions of poverty in various regions has become increasingly prominent.The thesis aims to explore different poverty alleviation methods adapted to local conditions.The data mining method is used to analyze the survey data of precision poverty alleviation,to achieve accurate management,accurate identification,and improve the effective allocation of poverty alleviation resources and provide supplementary decision-making for assessment and assistance in precision poverty alleviation.The research content of the thesis is as follows:(1)Research on regional differentiation of poverty alleviation effectiveness evaluation indicators based on logistic regression.Effective indicators are screened out from the data of accurate poverty alleviation evaluation and an evaluation index system is established.A logistic regression model was constructed by using this indicator system,and the indicators in the sample area were classified according to the weight of the poverty incidence rate in the sample area.The high-weight indicators will be the focus of the next year's evaluation and poverty alleviation,to achieve regional differential evaluation and assistance.(2)Research on the precise assistance of regional differentiation of poor households based on collaborative filtering.Extract the index data of the poor households from the survey data of the poor areas and combine the assistance data of relevant departments to make user portraits of the poor households and extract their characteristic labels.The feature labels are used to establish the recommendation model of precise assistance projects based on collaborative filtering.Through the similarity of neighbors among poor households,appropriate poverty alleviation projects are recommended for them to achieve precise policy implementation.(3)The above research contents are applied to the specific needs of precision poverty alleviation assessment and assistance work.The differential assessment and assistance analysis system is designed and implement for precise poverty alleviation areas,which provides effective auxiliary decision-making for different regions to carry out differential targeted poverty alleviation work.