這應該是第一篇(?)專業研究八卦版帶風向的文章
9/10才剛刊出 作者是來自 中研院統計科學研究所 的團隊
Research Article
Analysing the Unequal Effects of Positive and Negative Information on the
Behaviour of Users of a Taiwanese On-Line Bulletin Board
原文連結:
http://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0137842
摘要落落長 直接貼結論
Conclusions
Using the PTT gossiping data, we have found some interesting findings for the
effect of social influence on users’ decisions on showing approval or
disapproval towards posted messages. We see that as long as decisions are
made in a sequence and observable, individuals tend to use this information
to help them make their choices.
作者研究了PTT八卦版的data後，發現了風向的重要性。
一篇文章出來後，鄉民們會先看看其他人是推或噓，再來來決定自己的意見。
Our results suggest that such influence occurs at the local level, implying
on-line users tend to adopt such information to conform to the majority of the nearby individuals. Moreover,
when more local information is available, its impact also increases.
簡單來說，就是大家會跟著風向走；風向越強越多人跟著風向走。
Interestingly, even in an anonymous environment where individuals are unknown
to each other, there is still a tendency for everyone to follow the present
popular trend. If previous information is not revealed to users such that
they have little idea on what others have been doing, then whether social
influence still plays a role in users’ behaviour in an on-line community
remains an elusive question and experiments need to be conducted to
investigate this issue.
有趣的是，就算是在匿名，不知道id背後是誰的情況之下也是這樣。
Moreover, the global and local measures in our models assumed independent
effects and each comment has an equal weight. In reality, some users
may be perceived to be more important (or more prestigious) than others
in a community, and their comments may consequently influence others
more strongly.
研究假設每個id的意見的影響力是相同的，但事實上有些id的影響力特別大，
帶風向的能力特強，就是領頭羊這樣。
All these can be interesting questions and give directions to future research
on this particular topic.
更多的研究還在進行中...
心得:
真的不簡單 我看八卦版只能浪費生命而已 竟然有人可以發篇期刊
看來以後研究生在實驗室看八卦板 有新的藉口了