Abstract
Aiming at the limitation of existing similarity measure method based on Vague soft sets, the similarity measure formula between Vague soft sets is modified and a novel similarity measure between Vague soft sets in consideration of the difference of interval center of Vague values is introduced, the axiomatic proof is given too. The experimental results of comprehensive evaluation of the network public opinion show that this method is reasonable, effective and practical, which has a good application prospect and effect in the study of internet public opinion.
Highlights
at the limitation of existing similarity measure method based on Vague soft sets
The ex⁃ perimental results of comprehensive evaluation of the network public opinion show that this method is reasonable
分析发现,当聚类数据集规模较小时,2 种算法 的准确率和查全率基本都在 0.85 以上,但当数据样 本逐渐增大时,传统 Vague 软集聚类算法所得到的 准确率和查全率与基于 MapReduce 的并行化聚类 算法有明显差异,这是由于当数据量增大时,数据集 中会 出现很多非球形的不规则的类簇, 而 传 统 Vague 软集聚类算法对于非球形簇并没有很好的聚 类效果。 基于 MapReduce 的 Vague 软集并行化聚 类算法 所得到的准确率和查全率明显优于传统 Vague 软集聚类算法。
Summary
Vague 软集模型描述如下: 定义 1( Vague 软集) 设 U 是一个论域,E 是一 ∀e ∈ A,F(e) 为 U 上的一个 Vague 集,称(F,A) 为 定义 2( Vague 软相等) 设 ( F,A) 、 ( G,B) 为
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
