摘要
对大数据知识库资源信息特征的检测,能够有效提高知识库使用效率。对知识库信息特征的优化检测,需要将样本数据按属性划分为子集,并求出子集的信息熵并对样本进行检测测试。传统方法对知识库资源信息特征数据进行编码,形成初始数据信息群,但忽略了对信息样本进行测试,导致检测精度偏低。提出基于决策树分类的知识库资源信息特征检测方法。对知识库资源信息特征数据建立矩阵并进行矩阵转换处理,将样本数据按属性划分为子集,以子属性的个数为权重系数融入熵值计算,在所有熵值中选择最小的熵值所对应的属性为节点,对其余样本进行检测测试,实现对知识库资源信息特征的检测。实验结果表明,上述方法能够有效减小检测误差,通过熵值大小控制检测中冗余信息含量,对比当前方法有较小的噪声,能够有效地对大数据环境下的知识库资源信息特征进行检测。
Detecting the information features of the big data base resource can effectively improve the efficiency of the knowledge base. To optimize the information characteristics of the knowledge base, it is necessary to divide the sample data into subsets according to the attributes, obtain the information entropy of the subset, and test the sampies. The traditional method encodes the information data of the knowledge base resource to form the initial data information group, but ignores the test of the information samples, resulting in low detection accuracy. A method based on decision tree classification for knowledge resource information feature detection is proposed in the paper. The matrix of knowledge base resources information feature data and matrix conversion processing were established, and the sample data were divided into subsets according to the attribute The number of sub attribute was acted as weight coefficient in the entropy calculation. The nodes corresponding to the minimum entropy of all selected attribute entropies were selected to test the remaining samples and realize the detection of the information characteristics of knowledge base resources. The experimental results show that this method can effectively reduce the detection error and detect the information characteristics of knowledge resource under the big data environment.
作者
贺晓勇
侯冬尽
HE Xiao - yong;HOU Dong - jin(People's Public Security University of China Library, Beijing 100038, China)
出处
《计算机仿真》
北大核心
2018年第6期380-383,455,共5页
Computer Simulation
关键词
知识库资源
信息特征
优化检测
Knowledge base resources
Information characteristics
Optimized detection