For graph mining using random walk with restart (RWR), the formula is \(\mathbf{r}_{i}=c \tilde{W} \mathbf{r}_{i}+(1-c) \mathbf{e}_{i}\), where the ranking vector \(\mathbf{r}_{i}\) will start the random walk from node \(i, c\) is the restart probability, \(\tilde{W}\) is the normalized weight matrix, and \(\mathbf{e}_{i}\) is the starting vector. Please explain:

a. Why RWR could capture multiple weighted relationship between nodes?

b. What is the similarity and the difference between random walk based graph kernel and RWR?