Data Mining - text frequency?

[alvinsmode]

Hi everyone,



I'm handling a data set with over a million of rows(purchase history). I'm thinking what is the simplest way to find the pattern behind it, such as what words or combinations of words are the most popular from the data set? Any advise would be greatly appreciated. Thank you!

[June7]

Pattern of what? Vendor names, product descriptions? You could do a GROUP BY aggregate query on the data.

You have defined set of values you want to search for? Build a table of those values and use DCount() expression in query.

[alvinsmode]

Because the data was entered by many different employees...so very often we'd see something like "Blue Pen 12pc", "Pen 12pc Blue" or even "12piece Pen Blu", but they are the same exact product. So I'm trying to see if there is any easy way I can find the text frequency or the "N-grams?"? so I can find what are the common product descriptions regardless of the cleanliness of the descriptions.

[June7]

You've already asked this question https://www.accessforums.net/showthread.php?t=69205

[orange]

Alvin,
What exactly have you done other than posting the question(s) in the forum? Google/Bing?
Have you identified what you're going to do with the data and the related tables/files when you get it cleansed?