Heterogeneous Information Network Embedding for Meta Path based Proximity
This work tackles the challenge of representing complex heterogeneous networks for improved search and mining, but appears incremental as it extends homogeneous methods to a more specific domain.
The paper addresses the problem of network embedding for heterogeneous information networks, where existing methods only handle homogeneous networks, and proposes a new approach to preserve meta path based proximity.
A network embedding is a representation of a large graph in a low-dimensional space, where vertices are modeled as vectors. The objective of a good embedding is to preserve the proximity between vertices in the original graph. This way, typical search and mining methods can be applied in the embedded space with the help of off-the-shelf multidimensional indexing approaches. Existing network embedding techniques focus on homogeneous networks, where all vertices are considered to belong to a single class.