基于SSA-BPNN的农村精准扶贫成效评价

Effectiveness evaluation of rural targeted poverty alleviation based on SSA-BPNN

为实现精准扶贫成效评价,建立基于农户满意度的樽海鞘算法(Salp swarm algorithm,SSA)优化反向传播神经网络(Back propagation neural network,BPNN)的农村精准扶贫成效评价模型。首先,从生存环境维度、生活状况维度、精准扶贫政策效果和人文发展与社会保障4个角度建立基于农户满意度的精准扶贫成效评价指标体系。其次,将16个精准扶贫成效评价二级指标的得分数据和精准扶贫成效评价等级作为BPNN的输入向量和输出向量,建立精准扶贫成效评价BPNN模型。最后,运用SSA优化BPNN模型的初始权值和阈值,建立SSA-BPNN的精准扶贫成效评价模型。结果表明,与其他算法相比,SSABPNN具有更高的准确率,为精准扶贫成效评价提供了方法。

In order to achieve the effectiveness evaluation of targeted poverty alleviation,the rural targeted poverty alleviation effectiveness evaluation model based on the salp swarm algorithm(SSA)optimization back propagation neural network(BPNN)based on farmers’satisfaction was established.Firstly,from the perspectives of living environment dimensions,living conditions,the effects of targeted poverty alleviation policies,and human development and social security,the indicator system for evaluating the effectiveness of targeted poverty alleviation based on farmers’satisfaction was established.Secondly,the score data of 16 secondary indicators for the effectiveness evaluation of targeted poverty alleviation and the evaluation grade of the effectiveness of targeted poverty alleviation were used as the input vector and the output vector of BPNN to establish a BPNN model for the effectiveness evaluation of targeted poverty alleviation.Finally,SSA was used to optimize the initial weights and thresholds of the BPNN model,and establish the SSA-BPNN targeted poverty alleviation effectiveness evaluation model.The results showed that compared with other algorithms,SSA-BPNN had a higher accuracy rate and provided a method for evaluating the effectiveness of targeted poverty alleviation.