Frequency List for Common Phrases?

[why1942]

So I've been playing around with generating frequency lists of vocabulary words in Greek from various texts versions, which works really well using the Analytics output and the search RANGE and wildcard.

 

Then I wondered if it was possible to do a wildcard search in a Greek text (say the range was the NT but it could be any book in the NT) for all phrases that were based on number of words. So, depending on what I set for the number of words in the phrase (2 words, 3 words, 4 words, etc) the search results would in Analytics output a return all phrases in the NT that were x number of words long, sorted by frequency, so I could find out what the most common/most frequently used phrases (or word combinations) were? 

 

Would it be better to do this with an English translation so I can sort out actual intelligent phrases from gibberish, then use the interlinear to find out the Greek translation?

 

Is this even possible?

 

Thanks,

 

w

[docdave]

I found this question really interesting, and unfortunately I don't know whether or not it can be done in Accordance. I am still very much an amateur at using all of the amazing search capabilities of Accordance. At least for an English translation, it does sound like something that could be done relatively easily with something like Python, using regular expressions. The biggest challenge might actually be in creating an operational definition for what a "phrase" consists of.