This study explores how style and topic elements of language contribute to identifying online communities and influence the community's reception of content. Utilizing a combination of word and part-of-speech tag n-gram models for style analysis, and Latent Dirichlet Allocation for topic analysis, the research focuses on various Reddit forums. Findings indicate that style is a more significant identifier of community identity than topic, even in forums centered around specific subjects. Additionally, contributions that stylistically align with a community's norms are more positively received, whereas topic similarity does not have a similar correlation with community reception.
-
Tran, Trang, and Mari Ostendorf. "Characterizing the language of online communities and its relation to community reception." arXiv preprint arXiv:1609.04779 (2016).
-
Members
- Chris Anderson
- JoelR
- JoeyM
- envy
- Adriano Faria
- Square Wheels
- Nathan Explosion
- Dilip
- DawPi
- V0RT3X
- ali hagi
- lukash
- TracyIsland
- opentype
- StevenM
- Como
- Marcin Martyniak
- IC Essentials
- Andhrafriends Admin
- adik
- N700
- MissB
- XwReK
- terabyte
- GazzaGarratt
- A Zayed
- PrettyPixels
- Paul
- onlyME
- isvans
- Claudia999
- rainx
- NewVicious
- Daffy
- hyprem
- GuitarGathering
- Tripp
- Kirill Gromov
- Askancy
- MLK
- aXenDev
- Live Games
- Jelly Belly
- eveneme eveneme
- Analog
- Synergy
- burnyourfeelings
- Nomad
- ReyDev
- Morphe
- eivindsimensen
- YourSharona
- lordi
- shahed
- John Horton
- PayMap
- Serval
- Matt
- Nomer3
- Dennis Maidon
- Nicolas PC
- Ioannis D
- bernhara
- Zennuie
- COSMIN
- wulfx01
- Matthew Hawley
- bing11
- Verto
- George Anderssen
- Toby
- Cheryl
- ArashDev
- abobader
- IPS THEME
- SzymonPajacyk
- Bearback
- nosavinggrace
- Aengul
- Labis
- Maxius
- Shawn RR
- Richard Arch
- Marius
- Gary
- Sofia
- Ryan
- JoshB
- John Morris
- Mila
- Montreal
- aLEX49566
- PPlanet
- Ronald
- Fabian Paul Sanabria
- Meddysong
- sulervo
- PasXal
- ozman
- ZLTRGO