CAJ | 학술논문

为解决不均衡林业信息文本分类中少数类分类正确率低问题，提出了一种基于优化 LM 模糊神经网络的不均衡林业信息文本分类算法。在阐述优化 LM 模糊神经网络算法原理的基础上，提取不均衡林业信息文本特征矩阵训练分类器的各项参数，实现对不均衡林业信息文本的精准与快速分类。实验结果表明该算法对少数类辨识准确率高，优于神经网络分类法以及 SVM 算法、模糊神经网络算法，为不均衡林业信息文本的分类提供了新思路。
위해결불균형임업신식문본분류중소수류분류정학솔저문제，제출료일충기우우화 LM 모호신경망락적불균형임업신식문본분류산법。재천술우화 LM 모호신경망락산법원리적기출상，제취불균형임업신식문본특정구진훈련분류기적각항삼수，실현대불균형임업신식문본적정준여쾌속분류。실험결과표명해산법대소수류변식준학솔고，우우신경망락분류법이급 SVM 산법、모호신경망락산법，위불균형임업신식문본적분류제공료신사로。
In order to deal with the problem of low categorization accuracy of minority class of the uneven forestry information text classification algorithm, the uneven forestry information text classification algorithm was puts forward based on optimization LM fuzzy neural network (OLM-FNN). On the basis of expounding the principle of optimization LM fuzzy neural network (FNN), the parameters feature matrix training classifier of uneven forestry information text to of LM fuzzy neural network were extracted, thus realizing accurate and fast classification to uneven forestry information text. The experimental results show that the algorithm had higher classification accuracy of minority class than that of neural network and support vector machine (SVM) and fuzzy neural network. The algorithm provides new ideas for studying on uneven forestry information text classification algorithm.