Word-Word Co-Occurrence PMI Matrix

A Word-Word Co-Occurrence PMI Matrix is a word-word co-occurrence matrix that is a PMI matrix (for word-word co-occurrences).

• AKA: Word-Word Pointwise Mutual Information Matrix.
• Context:
  • It can be a Weighted Word-Word PMI Matrix.
  • It can range from being a Word-Context Window PMI Matrix, to being a Word-Document PMI Matrix, to being ...
• Example(s):
  • [math]\displaystyle{ \begin{array}{c|ccccc} & aardvark & ... & midterm & ... & zoo \\ \hline aardvark & ? & ... & n_{12} & ... & n_{1p} \\ ... & ... & ... & ... & ... & ... \\ midterm & n_{21} & ... & ? & ... & n_{2p} \\ ... & ... & ... & ... & ... & ... \\ zoo & n_{p1} & ... & n_{p2} & ... & ? \end{array} }[/math].
  • …
• Counter-Example(s):
  • tf-idf Matrix.
  • Pixel-Pixel Co-Occurrence PMI Matrix.
• See: PMI Score.

---

---

References

2007

• (Budiu et al., 2007) ⇒ Raluca Budiu, Christiaan Royer, and Peter Pirolli. (2007). “Modeling Information Scent: A Comparison of LSA, PMI and GLSA Similarity Measures on Common Tests and Corpora.” In: Large Scale Semantic Access to Content (Text, Image, Video, and Sound).
  • QUOTE: Unlike LSA, the initial word-by-document co-occurrence matrix is replaced by a word-by-word PMI matrix, in which words are represented as vectors of PMI scores relative to other words in the vocabulary (Niwa & Nitta, 1994).

---