Table 2 The logit transformation and regularized log-likelihood for both classifiers (WLR and WKLR)
From: Analyzing a co-occurrence gene-interaction network to identify disease-gene association
Model | Logit transformation | Regularized log-likelihood |
|---|---|---|
WLR | \( {ln \left (\frac {p_{i}}{1-p_{i}} \right)=X_{i}\beta } \text {(1)}\) | \( {lnL\left (\beta \right) = \sum \limits _{i=1}^{n} w_{i}ln\left (\frac {e^{y_{i}x_{i}\beta }}{1+e^{x_{i}\beta }} \right)- \frac {\lambda }{2}\left \| \beta \right \|^{2}} \text {(2)}\) |
WKLR | \( {ln\left (\frac {p_{i}}{1-p_{i}} \right) = k_{i}\alpha } \text {(3)}\) | \( { lnL_{W}\left (\alpha \right) = \sum \limits _{i=1}^{n} w_{i}ln\left (\frac {e^{y_{i}k_{i}\alpha }}{1+e^{k_{i}\alpha }} \right)- \frac {\lambda }{2}\alpha ^{T}K\alpha } \text {(4)}\) |