CBOW context words to vector for creating embedding

Course Q&A Natural Language Processing NLP with Probabilistic Models
1

In the lecture, when creating word embedding, the CBOW methods described here is: effectively the context words vector is created by averaging the vectors of context words. I want to ask the difference between this approach vs. convert context words to multiple observations. Anyone has insights please help!

2

Hi Xixi_NXCR,

You can find some discussion about this on stackoverflow, for example here. You can search stackoverflow for additional discussions.

3

Thank you for your suggestion and the link. I sure will read.