.@marcgalexander
"Deep down, don't all of us want a little more meaning in what we do?"
=> use the HTOED!
#d2e2015
0
0
0
0
#d2e2015
Latest posts tagged with #d2e2015 on Bluesky
Trending
Posts tagged #d2e2015
.@marcgalexander
Big data doesn’t mean we can’t account for everything – accounting for everything is Really Important
#d2e2015
0
0
0
0
0
0
0
0
.@marcgalexander
v interesting – no really New categories in Shakespearean English; & the only truly expanding category is music
#d2e2015
0
0
0
0
.@marcgalexander
"our data is warped" – we know this. But we need to think about whether the lexis or the culture has changed
#d2e2015
0
0
0
0
.@marcgalexander
"Modern English is very weird" – characterised by new fields of words arising
#d2e2015
0
0
0
0
.@marcgalexander
"Work is a bit darker than leisure" – ie older vocabulary
#d2e2015
0
0
0
0
.@marcgalexander
"The Anglo-Saxons had no word for CD-ROM"
#d2e2015
0
0
0
0
.@marcgalexander
showing us the Chaucer Hump and Post-Chaucer Drip
#d2e2015
0
0
0
0
.@marcgalexander
starts by saying corpus linguistics is more significant than the (wonderful) Historical Thesaurus of @OED
#d2e2015
0
0
0
0
Wrapping up #d2e2015 is @marcgalexander with a demo titled "How can we stage 470,000 words?" (featuring less than 93 slides, I hope)
0
0
0
0
Coats:
Q: have you looked at codeswitching?
A: mixed language filtered out from this data :P
#d2e2015
0
0
0
0
Coats:
tweeters shifting from using emoticons to using emojis (..?)
#d2e2015
0
0
0
0
Coats:
Expressive lengthening influenced by phonological considerations..?
#d2e2015
0
0
0
0
Coats:
way more emoticons in FI data than global/US data
#d2e2015
0
0
0
0
Coats:
lengthening of emoticons more common in FI data than US data
E.g. :))) and :DDD
#d2e2015
0
0
0
0
Coats:
Expr. lengthening by geography: FI more in 2013 than 2015; US more in 2015 than 2009
#d2e2015
0
0
0
0
Coats:
Non-standard expressive lengthening of words:
coooooool, yessssss
=> discourse marker of emotional affect
#d2e2015
0
0
0
0
Coats:
Gender in FI 2013 data:
males - nouns, initialisms, punctuation
females - usernames, names, hashtags
#d2e2015
0
0
0
0
Hmm, not sure lexical differences, FI vs anywhere, are all that interesting.
#d2e2015
0
0
0
0
Coats:
Lexical features: Finland 2013 v Global 2009
top 10 words in FI data include: "finland", "helsinki", "finnish", "sweden"
#d2e2015
0
0
0
0
Coats:
usernames filtered for gender - handles like *TweeterHki ignored
#d2e2015
0
0
0
0
Coats:
data tokenized and POS-tagged
- tagger identifies emoticons! :D
#d2e2015
0
0
0
0
Coats:
maps of tweets by language in Finland – Finnish, English all over; Swedish, Russian clearly focused on a couple regions
#d2e2015
0
0
0
0
Coats:
English tweets only - used a python script language identifier
#d2e2015
0
0
0
0
Coats:
2 small datasets, 2013 and 2015, geo-located to Finland
Comparison data: 2009 untagged, 2015 geo-located to US
#d2e2015
0
0
0
0
Ah no, turns out (if I heard correctly) tweets geotagged from smartphones but not from laptops
#d2e2015
0
0
0
0
Final paper at #d2e2015 – Steven Coats on Finland Twitter English
(..can't help feeling I'm about to hear a paper on my tweets!)
0
0
0
0
.@ahonkapo
Conclusions: each scribe has individual inventory of abbreviations for function words – these vary from mss to mss
#d2e2015
0
0
0
0
.@ahonkapo
Scribes copy closely when writing Latin; take more liberties when writing ME
#d2e2015
0
0
0
0