Distributional Document Embedding Space

A Distributional Document Embedding Space is a documebt vector space model that is a text-item embedding space composed of distributional document vectors.

• Context:
  • It can (often) have a Distributional Document Vectorizing Function.
  • It can be created by a Distributional Document Embedding Modeling System (that implements a distributional documenbt embedding modeling algorithm).
  • It can range from being a Closed Distributional Document Vector Space Model (that applies only to the words in the training data) to being an Open Distributional Document Vector Space Model.
  • …
• Example(s):
  • a Words Centroid-based Document Embedding Space.
  • a Word Mover's Distance (WMD)-based Document Embedding Space, as in (Kusner et al., 2015).
  • …
• Counter-Example(s):
  • Word Embedding Space.
  • Sentence Embedding Space.
  • Phrase Embedding Space.
  • Paragraph Embedding Space.
• See: Lexical Co-Occurrence Matrix, Distributional Word Vector, Vector Space Model.

---

---

References

2015a

• (Dai et al., 2015) ⇒ Andrew M. Dai, Christopher Olah, and Quoc V. Le. (2015). “Document Embedding with Paragraph Vectors.” In: NIPS Deep Learning Workshop.

2015b

• (Kusner et al., 2015) ⇒ Matt Kusner, Yu Sun, Nicholas Kolkin, and Kilian Weinberger. (2015). “From Word Embeddings to Document Distances.” In: Proceedings of the International Conference on Machine Learning, pp. 957-966.
  • QUOTE: … We present the Word Mover's Distance (WMD), a novel distance function between text documents. Our work is based on recent results in word embeddings that learn semantically meaningful representations for words from local cooccurrences in sentences. …

---