基于可拓知识库的短文本敏感信息查询扩展方法

A Query Expansion Method of Sensitive Information for Short Text Based on Extension Knowledge Base

为了解决一般检测算法在短文本查询上效率较低的问题,基于可拓学的方法构建特定领域可拓知识库并定义一种新的菱形推理模式,提出了基于可拓知识库的概念查询扩展算法,通过概念间的可拓关联性及可拓推理来处理短文本中敏感信息检测问题,并通过实例验证了算法的可行性。实验证明该算法克服了短文本自身长度较短、描述概念能力弱的问题,可减少相关信息的遗漏,该算法提高了文档敏感信息检测的准确率与召回率。

In order to solve the problem of the low efficiency caused by the traditional query expansion retrieval methods,a novel approach based on extension knowledge was issued with a novel rhomb reasoning method by using extension theory. By introducing the extension relationship between concepts with extension reasoning,the proposed method can relationally solve the detection problem of sensitive information. For illustration,an example was utilized to show the feasibility of the method in solving detection problem with concept query expansion method. Empirical results showed that the proposed method has a good performance on detection of short text which has a low degree of description,and could reduce the omission of relative information. The proposed method can improve the detection precision and recall of the document sensitive information.