[DOC] Collocations

doc/index.rst

@@ -34,6 +34,7 @@ Widgets
    widgets/keywords
    widgets/score-documents
    widgets/semanticviewer
+   widgets/collocations
    widgets/wordlist
    widgets/ontology
 

doc/widgets.json

@@ -233,6 +233,15 @@
      "search"
     ]
    },
+   {
+    "text": "Collocations",
+    "doc": "widgets/collocations.md",
+    "icon": "../orangecontrib/text/widgets/icons/Collocations.svg",
+    "background": "light-blue",
+    "keywords": [
+     "PMI"
+    ]
+   },
    {
     "text": "Statistics",
     "doc": "widgets/statistics.md",

doc/widgets/collocations.md

@@ -0,0 +1,44 @@
+Collocations
+============
+
+Compute significant bigrams and trigrams.
+
+**Inputs**
+
+- Corpus: A collection of documents.
+
+**Outputs**
+
+- Table: A list of bigrams or trigrams.
+
+**Collocations** finds frequently co-occurring words in a corpus. It displays bigrams or trigrams by the score.
+
+![](images/Collocations.png)
+
+1. Settings: observe bigrams (sets of two co-occurring words) or trigrams (sets of three co-occurring words). Set the frequency threshold (remove n-grams with frequency lower than the threshold).
+2. Scoring method:
+   - [Pointwise Mutual Information](https://en.wikipedia.org/wiki/Pointwise_mutual_information) (PMI)
+   - [Chi Square](https://en.wikipedia.org/wiki/Chi-squared_test)
+   - [Dice](https://en.wikipedia.org/wiki/S%C3%B8rensen%E2%80%93Dice_coefficient)
+   - [Fisher](https://en.wikipedia.org/wiki/Fisher%27s_method)
+   - [Jaccard](https://en.wikipedia.org/wiki/Jaccard_index)
+   - [Likelihood ratio](https://en.wikipedia.org/wiki/Likelihood-ratio_test)
+   - Mi Like
+   - [Phi Square](https://en.wikipedia.org/wiki/Phi_coefficient)
+   - Poisson Stirling
+   - Raw Frequency
+   - [Student's T](https://en.wikipedia.org/wiki/Student%27s_t-test)
+
+Example
+-------
+
+**Collocations** is mostly intended for data exploration. Here, we show how to observe bigrams that occur more than five times in the corpus. Bigrams are computed using the Pointwise Mutual Information statistics.
+
+We use the *grimm-tales-selected* data in the [Corpus](corpus-widget.md) and send the data to Collocations.
+
+![](images/Collocations-Example.png)
+
+References
+----------
+
+Manning, Christopher, and Hinrich Schütze. 1999. Collocations. Available at: https://nlp.stanford.edu/fsnlp/promo/colloc.pdf