Students at Monash University used Alveo as part of an introductory course in computational linguistics this year. The students completed an assignment which used basic techniques of corpus linguistics to compare one phenomenon in varieties of English.
The phenomenon of interest was the word used to introduce a term of comparison after the word different: do you use different from, different to or different than? Alveo gave the students access to various collections of Australian English which could be searched as a single corpus for the purposes of the exercise; the results of this search could then be compared with results from the British National Corpus and the Corpus of Contemporary American English (both accessed via corpus.byu.edu).
For those who may be interested, the results show that different from is the most common combination in all three varieties. Different to is almost absent from American usage, but is about equally common in the other two varieties, while different than is absent from British English and more common in American usage than in Australian English. But perhaps the most surprising finding was that the total of all comparative constructions with different was much higher in the Alveo data than in either of the other collections, possibly as a result of the fact that the Australian data is older than the other two sources (and also is biased towards formal written genres). I think that we can infer that this assignment generated a good deal of interest for at least some students from the fact that some of them took multiple samples from the large corpora to provide the most reliable results, even when they were only asked to a sample once.