Bring Back Most Frequently Used Word in a Column

Hi.

You'd first need to confirm a definitive list of all punctuation that may be present within any of the strings in that column.

This is because, in order to give correct results, any solution must use an appropriate delimiter: we would not, for example, want "human" to be included in a count for "man" simply because it contains the string "man". As such, the usual way to approach this task is to search for the string " man " (i.e. with a space at either end), which would guarantee that such cases as above are ruled out.

Of course, punctuation presents a potential obstacle to this set-up. Searching for " man " would fail to count cases such as "The man's hat", "Is that a man?", "Woman (to the man)" or "Man! What a day!", for example.

If you could provide that definitive list then we can account for such possibilities. Of course, we could first devise a set-up in which all possible punctuation marks are first eliminated, though obviously this would be a touch overkill, and make any potential solution far more complex (not to mention resource-heavy) than necessary.

Regards

There's a workbook at https://app.box.com/s/fkowtna5k76pp804tiqe that will do this:

Row\Col	A	B	C
1	Unique	Output List	Total
2	406		987
3	#	Word	Freq
4	1	to	25
5	2	in	24
6	3	reply	21
7	4	a	20
8	5	this	18
9	6	is	16
10	7	the	15
11	8	Excel	14
12	9	back	13
13	10	forum	13
14	11	bring	12
15	12	column	12
16	13	post	12
17	14	Quote	12
18	15	are	11
19	16	that	11
20	17	thread	11
21	18	by	10
22	19	for	10
23	20	with	9
24	21	word	9
25	22	you	9
26	23	frequently	8
27	24	message	8
28	25	most	8
29	26	Off	8
30	27	Quick	8
31	28	used	8
32	29	and	7
33	30	man	7
34	31	of	7
35	32	replies	7
36	33	i	6
37	34	or	6
38	35	Add	5
39	36	all	5

That's a portion of the histogram of the words on this page (before I posted). I did Ctrl+A, Ctrl+C, and pressed the Input button on the sheet, and that's it.