How to compute
With index relationships computed, choose Compute | concgrams.
My relationship data were sorted first by Word 1, then by the Log ratio score.
Then I selected one term HAND.
and completed the Concgram Statistic window
to show that I wanted a minimum Log Ratio strength of 2.0 for the one term HAND. The result:
The Notes show that 5 terms (the ones in red) were computed as collocates of hand using the same text.
Example (multiple terms)
The data were sorted simply on the CondProbA statistic with highest scores above 140.
This time I completed the window like this:
Results :
This procedure allows a maximum of 10 search-words.
|
What are concgrams?
For years it has been easy to search for or identify consecutive clusters (n-grams) such as AT THE END OF, MERRY CHRISTMAS or TERM TIME. It has also been possible to find non-consecutive linkages such as STRONG within the horizons of TEA by adapting searches to find context words. The concgram procedure takes a whole corpus of text and finds all sorts of combinations like the ones above, whether consecutive or not.
Cheng, Greaves & Warren (2006:414) define a concgram like this
For our purposes, a ‘concgram’ is all of the permutations of constituency variation and positional variation generated by the association of two or more words. This means that the associated words comprising a particular concgram may be the source of a number of ‘collocational patterns’ (Sinclair 2004:xxvii). In fact, the hunt for what we term ‘concgrams’ has a fairly long history dating back to the 1980s (Sinclair 2005, personal communication) when the Cobuild team at the University of Birmingham led by Professor John Sinclair attempted, with limited success, to devise the means to automatically search for non-contiguous sequences of associated words.
Essentially what they were seeking in developing the ConcGram (©) program was "a search-engine, which on top of the capability to handle constituency variation (i.e. AB, ACB), also handles positional variation (i.e. AB, BA), conducts fully automated searches, and searches for word associations of any size." (2006:413)
|
