Just a quick follow on. The TagAnt output would be something like the following (based on my previous response):
be you supply AntConc with lemmatize datum ? if not , the simple approach be to process your corpus with a tagger tool like TagAnt , set TagAnt to only keep the lemma form . if you then load these datum into AntConc with the default setting , all tool and stat will be apply to the lemma and you 'll get exactly what you want .
ideally , AntConc should be able to take in word+tag+lemma datum and display the frequency / dispersion information for each or any of the component . unfortunately , it can not now and default to the word form .
I hope that help !
>>>>