Popular words in company blogs
I’ve had a look at the most frequent nouns in my corporate blogging corpus. Here’s the top 50 list:
Rank | Word
1 | time
2 | � (*)
3 | people
4 | way
5 | blog
6 | Microsoft
7 | Windows
8 | year
9 | business
10 | day
11 | today
12 | information
13 | years
14 | week
15 | team
16 | things
17 | system
18 | site
19 | world
20 | something
21 | customers
22 | work
23 | Web
24 | Google
25 | lot
26 | application
27 | New
28 | part
29 | page
30 | data
31 | company
32 | Oracle
33 | Web
34 | number
35 | post
36 | code
37 | Example
38 | product
39 | % (*)
40 | one
41 | market
42 | list
43 | service
44 | user
45 | software
46 | technology
47 | Server
48 | Vista
49 | thing
50 | applications
(*=these are instances where my tagger got confused and erroneously labeled a special character as a noun)
Regarding the high frequency of MICROSOFT, WINDOWS and VISTA it is important to note that posts from blogs.msdn.com make up about 15% of the entire corpus and it can be assumed that these terms are very frequent over there (and note that they don’t just occur by themselves but also as part of product names).
It’s also interesting that you can easily guess where the statistics from the following five sources come from, simply by looking at the frequencies:
Blog #1
1 Southwest NP 208
2 time NN 88
3 Airlines NPS 82
4 year NN 71
5 Dallas NP 63
6 day NN 62
7 flight NN 54
8 way NN 49
9 years NNS 49
10 employees NNS 47
11 customers NNS 46
12 family NN 44
13 people NNS 40
14 home NN 34
15 Flight NP 33
Blog #2
1 Wal-Mart NP 142
2 company NN 65
3 associates NNS 56
4 time NN 54
5 Sam NP 47
6 family NN 44
7 years NNS 43
8 people NNS 41
9 store NN 36
10 day NN 28
11 Ark. NP 27
12 Bentonville NP 23
13 Mr. NP 23
14 way NN 22
15 job NN 22
Blog #3
1 Thomson NP 107
2 information NN 91
3 place NN 74
4 destination NN 74
5 year NN 58
6 event NN 52
7 city NN 47
8 guide NN 47
9 holiday NN 42
10 visit NN 38
11 week NN 37
12 day NN 35
13 Festival NP 34
14 Thomsonfly NP 33
15 Sunday NP 32
Blog #4
1 Boeing NP 147
2 airplane NN 146
3 airplanes NNS 107
4 year NN 95
5 years NNS 60
6 market NN 57
7 program NN 55
8 flight NN 55
9 orders NNS 53
10 time NN 53
11 Airbus NP 49
12 air NN 47
13 Farnborough NP 46
14 customers NNS 45
15 today NN 41
Blog #5
1 McDonald NP 147
2 CSR NP 50
3 food NN 40
4 issues NNS 39
5 business NN 37
6 people NNS 36
7 suppliers NNS 28
8 restaurants NNS 27
9 company NN 25
10 supply NN 25
11 lot NN 24
12 time NN 23
13 Example NN 23
14 chain NN 22
15 world NN 22
Note: interesting that the two sources I would most readily assign an image function - #1 and #2 - both have FAMILY in the top 15.