Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Show HN: Role2Vec: Learning Role-Based Graph Embeddings, Ijcai 2018.
1. Compared to other structural node embedding techniques it is fast.
2. It was easy to build it on high performance implicit factorization packages (e.g. gensim).
3. The structural features are learned from motifs.
4. Structural properties of neighbors are also included.
5. The paper has some theoretical aesthetics.
There are no comments on this comment yet.
1. Compared to other structural node embedding techniques it is fast.
2. It was easy to build it on high performance implicit factorization packages (e.g. gensim).
3. The structural features are learned from motifs.
4. Structural properties of neighbors are also included.
5. The paper has some theoretical aesthetics.