Collocation visualization

Visualize distribution of distances between correlated words in the ARANEA Corpora or the Slovak National Corpus.

Enter two words (or lemma, or CQL expressions) (e.g.. pull string):

size of the corpus AranSlov_x is 125002204 tokens
frequency of dať: 130018; ipm=1040.0
frequency of pozor: 10176; ipm=81.0
frequency of tight collocation dať pozor: 1157; ipm=9.3
mean frequency of pozor in our sample (right context of dať): 312.667; ipm=2.5
frequency of pozor in collocation with dať, assuming they are independent: 10.531±0.262; ipm=0.084
mean distance |dať,pozor| in our sample (right context of dať): 1.648±1.047
Fitting... log

Legend:

• x axis: distance of the word pozor from the word dať
• y axis: number of occurrences of pozor at the given position in the corpus
• errorbars: 95 % confidence interval, assuming word occurrence follows the Poisson process
• yellow vertical line: mean distance of the word pozor from the word dať (separately for the left and right context)
• green & blue lines: gamma distribution fit
• green-blue horizontal line (if visible): baseline frequency of pozor, assuming independence from dať (i.e. far from it)