search

dselivanov / text2vec

Fast vectorization, topic modeling, distances and GloVe word embeddings in R.
http://text2vec.org
Other
850 stars 135 forks source link

Using the created word embeddings for calculating document similarity #276

Closed FinScience closed 6 years ago

FinScience commented 6 years ago

Hi,

Thanks for the wonderful package. I am working on a use case where based on an input string/ paragraph, a set of documents need to be ranked based on relevance. I have followed the tutorials and have independently been able to create word embedding for the document set as well as replicate the document similarity example. However, since I want my search to be context based(taking presence of neighboring words in a document into account) , I was wondering if there is a way to use the created word embedding in document similarity.

Any help on this would be great.

dselivanov commented 6 years ago

Please post questions on stackoverflow.