We first build individual graphs for each document and then use GNN to learn the fine-grained word representations based on their local structures, which can also effectively produce embeddings for unseen words in the new document.
We present an automatic mortality prediction scheme based on the unstructured textual content of clinical notes.
While the celebrated Word2Vec technique yields semantically rich representations for individual words, there has been relatively less success in extending to generate unsupervised sentences or documents embeddings.
In document-level sentiment classification, each document must be mapped to a fixed length vector.
Ranked #1 on Sentiment Analysis on IMDb (using extra training data)
Finally, although not trained for embedding sentences and words, it also achieves competitive performance on crosslingual sentence and word retrieval tasks.
Most current approaches to metaphor identification use restricted linguistic contexts, e. g. by considering only a verb's arguments or the sentence containing a phrase.
Visualizing graph embeddings annotated with predictions of potentially suicidal individuals shows the integrated model could classify such individuals even if they are positioned far from the support group.
Our method creates an extractive summary by selecting the sentences with the closest embeddings to the document embedding.