二进制特征的相似性和非相似性度量(the Similarity Measure and Dissimilarity Measure for Binary Features)

假设我们使用两个二进制下标来计算两个对象的特征。$n_{00}$和$n_{11}$分别代表两个对象特征同时不存在和存在，$n_{01}$和$n_{10}$代表对象特征只存在在一个对象。对于两个数据点$x_i$和$x_j$两种常见类型的相似性度量如下
$$S_{ij}=\frac{n_{11}+n_{00}}{n_{11}+n_{00}+w(n_{10}+n_{01})}$$
$w=1,$ simple matching coefficient; $w=2,$ Rogers and Tanimoto measure; $w=1/2,$ Gower and Legendre measure.
这些度量直接计算两个对象之间的匹配。未匹配的对根据它们对相似度的贡献进行加权。
$$S_{ij}=\frac{n_{11}}{n_{11}+w(n_{10}+n_{01})}$$
$w=1,$ Jaccard coefficient; $w=2,$ Sokal and Sneath measure; $w=1/2,$ Gower and Legendre measure.
These measures focus on the co-occurrence features while
ignoring the effect of co-absence.