基于语义分析的大规模动态图形相似节点查询算法研究

随着动态图形在网络应用领域的发展与扩展,针对当前相似节点查询算法存在运行响应速度慢、查询占用的存储空间较大、容易受到外界因素干扰而导致查询精确度不佳等问题,提出基于语义分析的大规模动态图形相似节点查询算法。首先,依据语义分析的方法,对动态图形相似节点数据进行预处理,获取节点相似性函数;其次,针对当前节点在不同时刻的相似性函数,设计相应的矩阵以增强计算效率;最后,将求出的节点相似程度值按照升序顺序进行排列,将相似程度值最高的节点看作相似节点,实现相似节点的查询。实验结果表明,所提算法查询速度快,查询精确度较高。

Query algorithm for similar nodes in large scale dynamic graphics based on semantic analysis

With the expansion and development of the dynamic graphics in the field of network application, in view of the current query algorithm has similar node running slow response speed and occupy storage space query is large, vulnerable to the interference of external factors caused problems such as poor query accuracy, proposed large-scale dynamic graphics semantic analysis based on similarity query node algorithm. First of all, on the basis of the method of semantic analysis, preprocess the node data of dynamic graphics similar to obtain node similarity function; secondly, according to the current node similarity function in different time, the design of the corresponding matrix to enhance the calculation efficiency; finally, the node calculated similarity values are arranged in ascending order, will be similar to the highest degree nodes as similar nodes, achieve similar node query. Experimental results show that the proposed algorithm has fast query speed and high query accuracy.