在Jaccard距离和归一化信息距离的模拟之间进行插值
Interpolating between the Jaccard distance and an analogue of the normalized information distance
摘要:
Jiménez,Becerra,and Gelbukh(2013)定义了一类“对称特沃斯基比率模型”$S_{α,β}$,$0\leα\le 1$,$β>0$。每个函数$D_{α,β}=1-S_{α,β}$是给定有限集的幂集上的半度量。我们证明了$D_{α,β}$是一个度量当且仅当$0\leα\le \frac12$和$β\ge 1/(1-α)$。这一结果在精益验证助手中得到了正式验证。这个参数化的度量空间的极值点是$\mathcal V_1=D_{1/2,2}$,Jaccard距离和$\mathcal V_{\infty}=D_{0,1}$,M的归一化信息距离的模拟。Li,Chen,X. Li,Ma,and Vitányi(2004).作为第二种插值,通常我们还证明了$\mathcal V_p$是一个度量,$1\le p\le\infty$,其中$$Δ_p(A,B)=(|B\setminus A|^p+|A\setminus B|^p)^{1/p},$$ $$\mathcal V_p(A,B)=\frac{Δ_p(A,B)}{|A\cap B| + Δ_p(A,B)}.$$
英文摘要:
Jiménez, Becerra, and Gelbukh (2013) defined a family of "symmetric Tversky ratio models" $S_{α,β}$, $0\leα\le 1$, $β>0$. Each function $D_{α,β}=1-S_{α,β}$ is a semimetric on the powerset of a given finite set. We show that $D_{α,β}$ is a metric if and only if $0\leα\le \frac12$ and $β\ge 1/(1-α)$. This result is formally verified in the Lean proof assistant. The extreme points of this parametrized space of metrics are $\mathcal V_1=D_{1/2,2}$, the Jaccard distance, and $\mathcal V_{\infty}=D_{0,1}$, an analogue of the normalized information distance of M. Li, Chen, X. Li, Ma, and Vitányi (2004). As a second interpolation, in general we also show that $\mathcal V_p$ is a metric, $1\le p\le\infty$, where $$Δ_p(A,B)=(|B\setminus A|^p+|A\setminus B|^p)^{1/p},$$ $$\mathcal V_p(A,B)=\frac{Δ_p(A,B)}{|A\cap B| + Δ_p(A,B)}.$$
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。