基于BP神经网络的农户生计风险评价模型

An assessment model of farmer livelihood risk based on BP neural network

当今,农户生计风险在农村社会建设的过程中是最为重要和核心的问题之一。作为基本的社会经济单元,农户正面临着来自经济、社会和自然等各方面的风险。石羊河流域农户由于面临着严峻的风险冲击使其生计脆弱性进一步加剧。所以,急需辨明流域农户面临的主要风险类型,并依此找寻提高农户风险应对能力的对策措施。针对流域农户生计风险的非线性关系和复杂性特征,以及BP神经网络的适用性范围,建立了基于BP神经网络的农户生计风险评价指标体系。并基于入户调查数据,运用熵权法计算得出流域农户生计风险评价指标体系中各种生计风险变量。同时,根据石羊河流域农户所面临生计风险的特征,基于BP神经网络结构模型,确定了其评价结果的表征方法。最终对模型的可行性进行了验证。研究表明:基于BP神经网络的生计风险评价模型对定量化测度农户生计风险的适用性较强,这可为决策部门定量评估农户生计风险提供依据。

Farmers in the Shiyang River basin are faced with serious livelihood risk,which has aggravated the farmers＇ livelihood vulnerability. It is extremely urgent now to distinguish the key livelihood risk which the farmers are facing,explore the potential factors which has hindered the farmers to resist the main types of livelihood risk in order to seek effective measures which can improve the abilities of farmers to cope with livelihood risk. In the light of the complexity and nonlinear characteristic of livelihood risk for farmers in the basin,as well as the strong adaptability of BP neural network to nonlinear dynamic system,an assessment model with livelihood risk based on BP neural network was put forward. Meanwhile,the variances among different types of livelihood risk are calculated by entropy method based on data from practical survey. According to the above,combining the characteristics of livelihood risk of farmers in the Shiyang River basin,a evaluation system is put forward which can help to count,evaluate and determine the network structure of BP and characterization of evaluation results.Finally,the feasibility of the model was verified by an example calculation.