Random Walk-Based Approach

Learn about graph embedding algorithms that use random walk.

We'll cover the following

Random walk

As its name suggests, random walk performs a random "walk" on a given graph to compute the embeddings. Take a graph and select a node at random. Now, select its neighbor at random and move to it. This process is repeated until a fixed number of walks is reached. The number of walks is predetermined by the user or tuned as a hyperparameter.

Let's understand this using an example.

Get hands-on with 1200+ tech skills courses.