Population estimation of internet forum community by posted article distribution

Authors:
Masao Kubo;Keitaro Naruse;Hiroshi Sato;Takashi Matsubara
Affiliations:
National Defense Academy of Japan, Yokosuka, Kanagawa, Japan;University of Aizu, Fukushima-ken, Japan;National Defense Academy of Japan, Yokosuka, Kanagawa, Japan;National Defense Academy of Japan, Yokosuka, Kanagawa, Japan
Venue:
KES'10 Proceedings of the 14th international conference on Knowledge-based and intelligent information and engineering systems: Part IV
Year:
2010

Citing 3
Cited 1

Lurker demographics: counting the silent

Proceedings of the SIGCHI conference on Human Factors in Computing Systems
Information diffusion through blogspace

Proceedings of the 13th international conference on World Wide Web
Lognormal Distribution of BBS Articles and its Social and Generative Mechanism

WI '06 Proceedings of the 2006 IEEE/WIC/ACM International Conference on Web Intelligence

Word familiarity distributions to understand heaps' law of vocabulary growth of the internet forums

KES'11 Proceedings of the 15th international conference on Knowledge-based and intelligent information and engineering systems - Volume Part III

Quantified Score

Hi-index	0.04

Visualization

Abstract

Traditionally, it seems that there are no typical patterns or regularities on an internet forum and the bulletin board because it is a result of voluntary behaviour of an individual user. However, recent studies show distribution of the number of posted articles per user on them follows as a power law distribution. We think that this regularity of the emergent behaviour is very important for understanding social science and human communication because it is possible to estimate the number of lurkers of each forum thread only using by usually observable data. In this paper, we proposed a set of estimation methods based on preferential attachment which is emerged by reciprocity as a human trait. The effectiveness of our proposed estimation methods is verified by comparing it with indices to be strongly related to the population of its corresponding human behaviour.