Skip to content

Words mentioned in 23-Jun-2010 Canadian Earthquake tweets

words mentioned in earthquake tweets 23-jan-2010

Using twitter gardenhose access, remove stopwords and punctuation sprinkle in a little bit of mapping, some reducing, and voila! The most frequently-occurring words in tweets that mentioned earthquake from June 23, 2010. I left earthquake out of the image itself because being that it was in every tweet, it overwhelmed the rest of the words.  I find it amazing that the most frequently occurring ‘word’ is RT.

Also, wordle seemed to strip out numeric ‘words’ which is a shame because people tweeted the magnitude left-and-right.  See the data below for the top 100 words.

survey:79
hey:79
seriously:80
preliminary:81
info:81
strikes:81
hell:82
4.5:82
gta:82
geological:83
magnitude50:83
check:83
call:84
service:86
globeandmail:86
triggered:87
experience:88
video:89
earthquakes:90
guess:91
caused:92
pm:92
fuck:93
ground:94
bad:94
5.7:94
move:96
minutes:98
2.3:98
damn:99
mini:99
eastern:100
scared:100
will:100
philippec:101
cool:103
northern:104
live:106
struck:107
city:112
work:112
floor:113
epicenter:113
pretty:113
nyc:113
seconds:113
todays:113
afternoon:114
feeling:116
safe:116
haha:117
central:117
tremor:117
whoa:118
downtown:120
rattles:121
warning:121
damage:122
separating:125
guys:126
god:126
tweets:129
heard:129
tremors:130
north:131
rochester:131
earth:134
small:135
california:136
missed:137
cp24:138
finally:139
minor:139
fake:141
scary:141
detroit:143
big:146
breaking:147
coming:151
good:152
weird:156
area:159
experienced:161
lake:161
buffalo:163
happened:163
2010:164
survived:168
hope:172
thing:172
evacuated:173
canadian:176
cleveland:176
tsunami:177
region:178
office:182
shake:184
reported:185
ontarioquebec:187
buildings:193
hits:194
ago:200
house:200
york:200
ohio:205
going:207
border:213
shit:214
usgs:215
quake:216
michigan:217
5.0:222
omg:238
time:240
reports:241
holy:241
southern:242
shook:249
twitter:252
day:253
crazy:255
shakes:270
people:276
wtf:308
building:309
ny:355
thought:362
tornado:375
montreal:375
hit:388
lol:406
quebec:414
wow:418
shaking:423
news:433
g20:490
today:577
magnitude:612
ontario:781
5.5:988
ottawa:1046
canada:1373
feel:1439
toronto:2086
felt:2146
rt:4046
earthquake:14918
survey:79hey:79seriously:80preliminary:81info:81strikes:81hell:824.5:82gta:82geological:83magnitude50:83check:83call:84service:86globeandmail:86triggered:87experience:88video:89earthquakes:90guess:91caused:92pm:92fuck:93ground:94bad:945.7:94move:96minutes:982.3:98damn:99mini:99eastern:100scared:100will:100philippec:101cool:103northern:104live:106struck:107city:112work:112floor:113epicenter:113pretty:113nyc:113seconds:113todays:113afternoon:114feeling:116safe:116haha:117central:117tremor:117whoa:118downtown:120rattles:121warning:121damage:122separating:125guys:126god:126tweets:129heard:129tremors:130north:131rochester:131earth:134small:135california:136missed:137cp24:138finally:139minor:139fake:141scary:141detroit:143big:146breaking:147coming:151good:152weird:156area:159experienced:161lake:161buffalo:163happened:1632010:164survived:168hope:172thing:172evacuated:173canadian:176cleveland:176tsunami:177region:178office:182shake:184reported:185ontarioquebec:187buildings:193hits:194ago:200house:200york:200ohio:205going:207border:213shit:214usgs:215quake:216michigan:2175.0:222omg:238time:240reports:241holy:241southern:242shook:249twitter:252day:253crazy:255shakes:270people:276wtf:308building:309ny:355thought:362tornado:375montreal:375hit:388lol:406quebec:414wow:418shaking:423news:433g20:490today:577magnitude:612ontario:7815.5:988ottawa:1046canada:1373feel:1439toronto:2086felt:2146rt:4046earthquake:14918

3 Comments

  1. Would be interesting to use a word stemmer to reducer verbs to one form. Also removing special Twitter tokens like @s, and RTs helps :) Good job!

    Posted on 24-Jun-10 at 9:39 am | Permalink
  2. You can try Tagxedo, which supports stemming, and allows selective removal of punctuations and words. Also option to support numbers (actually Wordle can do that too). The cloud can actually be made into a shape (e.g. the maple leave)

    http://www.tagxedo.com

    Here’s a thumbnail of the image made with the above text:
    http://www.tagdoku.com/image/3b83f62f5ce9473d

    Here’s the interactive version (requires Silverlight):
    http://www.tagdoku.com/artful/3b83f62f5ce9473d

    Posted on 24-Jun-10 at 10:19 am | Permalink
  3. admin

    very very cool! I’ll have to consider it for future projects. And thanks for representing my data in tagxedo!

    Posted on 24-Jun-10 at 10:23 am | Permalink

Post a Comment

Your email is never published nor shared. Required fields are marked *
*
*