Risk Identification of Animal Husbandry Enterprises
As a basic industry of the national economy, animal husbandry has a very important strategic significance for the entire national economy and people’s livelihood. At this stage, animal husbandry in Jilin Province has become a pillar industry of the rural economy. At the same time, it is also in a critical period of transformation and upgrading. It has a significant “productive and weak nature” and is more dependent on financial capital than before, facing the financial dilemma of insufficient capital supply and shortage of funds. Therefore, it is necessary to correctly understand and understand the current characteristics of the animal husbandry in the process of economic development and the existing constraints, and take corresponding countermeasures to promote its accelerated transformation and upgrading, and complete the transition to modern animal husbandry. This paper applies data mining technology based on rough entropy to apply the method of RFM enterprise segmentation in marketing to the classification of creditworthiness of loan enterprises, and provides decision support for risk control management of bank loans. Decision trees are commonly used classification techniques in data mining, and the generated rules are easy for decision makers to understand and apply. However, when faced with a decision tree generated by a record set with more attributes and redundant and noisy attributes, the redundant attributes cannot be deleted, resulting in a complicated operation process. This paper aims to combine the decision tree method with the rough entropy theory, reduce the attributes, reduce the computational complexity, and generate a relatively simplified rule form.