融合多模式匹配的网络信息实体关联研究仿真

对网络信息实体进行关联匹配,能够更好的实现网络数据的传递和分析。由于网络数据呈现多源异构,以及非均匀分布等特征,导致难以对其信息实体进行准确快速的关联匹配。由此,提出了融合多模式匹配的网络信息实体关联策略。策略考虑了网络信息实体的复杂性与动态性,首先设计了语法相似性,对大量简单信息实体进行快速匹配;然后基于深度与距离设计了语义相似性,对实体中包含的词干与复合词汇进行准确匹配;再利用数据类型建立类型相似性,对缺失信息的实体进行匹配;最后通过编辑距离与惩戒函数,设计了结构性相似度,对实体之间上下文依赖与约束进行匹配。根据实验结果,验证了融合多模式匹配的网络信息实体关联策略具有灵敏的区分能力,并且在匹配准确度和匹配效率上均取得了显著的性能优化效果,能够有效应对网络信息实体的异构与分布特性。

Research and Simulation of Network Information Entity Association Based on Multi-Pattern Matching

The network information entities are matched by association,and it can better realize the transmission and analysis of network data.Because network data are heterogeneous and multi-source,and the characteristics of non-uniform distribution exist,it is difficult to match the information entities accurately and quickly.Therefore,a strategy of network information entity association based on multi-pattern matching is proposed.This strategy considerd the complexity and dynamics of network information entities.Firstly,the grammatical similarity was designed,and a large number of simple information entities were fast matched.Then semantic similarity was designed based on depth and distance.The stem and compound vocabulary contained in the entity were matched accurately.Then,data types were used to establish type similarity and match entities with missing information.Finally,the structural similarity was designed by editing distance and penalty function,and context dependencies and constraints between entities were matched.According to the experimental results,it is verified that the network information entity association strategy based on multi-pattern matching is sensitive to discrimination.In addition,significant performance optimization results have been achieved in terms of matching accuracy and matching efficiency,which can effectively response to heterogeneity and distribution of network information entities.