Corpora and collocations

By Stefan Evert. 2007

429 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/747e6c3df3fa89106784cf621fd3d5325df52530

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Formulaic Language in Learner Corpora (2012, methodology): “, Evert, 2008), and learner corpora have until very recently been of limited size. However, two recent studies should be mentioned here. Groom (2009) made use of two association measures, the t score and MI score, to extract significant co-occurrences of the 10 most frequent prepositions from two su”
Lexical Sophistication as a Multidimensional Phenomenon: Relations to Second Language Lexical Proficiency, Development, and Writing Quality (2018, unknown): “…are conceptualized at both word and phrase levels is a phraseological perspective, which suggests that people comprehend, process, and use collocations in a manner similar to their use of single lexical items (Hoey, 2005; Sinclair, 1991; see Evert, 2008, for detailed discussion of collocations).”
A distributional similarity approach to the detection of semantic change in the Google Books Ngram corpus. (2011, methodology): “The scores of the constructed co-occurrence matrix are given by local mutual information (LMI) scores (Evert, 2008) computed on the frequency counts of corresponding 2-grams 2 .”
Distributional Models of Word Meaning (2018, unknown): “…live park ride tell (3) bike c ar dog lion ⎛ ⎜⎜⎜⎝ 0 0.50 0 0 0 0 1.09 1.79 0 0 0.80 1.56 0 0 0 0.18 0 0 0 0 0 2.01 0 1.65 0 0 2.16 2.75 0 0 0 0.26 1.01 0 0 0 ⎞ ⎟⎟⎟⎠ For other types of weighting functions used in DS, see Curran (2003), Evert (2008), Turney & Pantel (2010), and Kiela & Clark (2014).”
Collocations in context:a new perspective on collocation networks (2015, methodology): “As is generally accepted (Evert 2008), the simple MI score emphasizes the exclusivity of the collocational relationship and thus has a propensity to highlight unusual combinations (including even typos and non-standard spellings) that occur only once or twice in the corpus.”
50-something years of work on collocations: What is or should be next … (2013, background): “…nothing unpredictable or interesting about it in terms of both form and function, and many researchers would prefer assigning collocation status to something more functionally useful (even if rarer) such as because of or according to (cf. again Evert 2009 for useful discussion and exemplification).”
Normalized (pointwise) mutual information in collocation extraction (2009, methodology): “For G2 it has been observed that it is equivalent to MI in collocation extraction (e.g., Evert, 2005, Sect.”