A recent article
in NewScientist says researchers at Cornell University, New York, and
Microsoft Research in Washington State have created software that can
identify experts in online communities based on the structure of their
online interactions. It says:
Welser’s group found that the most informative
individuals – dubbed “answer people” – are also relatively taciturn,
rarely participating in discussions heavily. They also tend to shy away
from the “discussion artists” who dominate most threads.
Instead, these people mostly post one or two messages to a lot of
different discussion threads, and tend to respond to users who do not
post a lot. They also tended to avoid long discussions, jumping in when
someone had a specific question, providing a useful answer and then
bowing out from further talk.
Because the findings use quantitative data about posting behaviour,
Welser says they could prove useful for developing automated systems
that assigns high reputation to certain people within a discussion.
Most of the current reputation systems out there including the Karma scale in Slashdot, the Yahoo Answers point system and the Amazon Askville experience points system
are primarily community driven. Automated systems assigning reputation
based on the structure of interactions may add new value to existing
systems. The article goes on to say:
They rated the content of a total of 5,700 messages from
about 450 active users. Then they calculated how often each user
replied to messages or were replied to, how often each person started a
discussion, and how many posts they contributed to an individual
thread. Then, by going through each message and rating their
usefulness, they were able to spot patterns in the behaviour of
This is interesting !! While point systems like the one in Yahoo
Answers are based on a range of activities within the community, it may
make sense to bring in new temporal variables that track “how often”
someone does something of interest in the community.
SezWho – a portable
reputation system that can work across all kinds of social media has a
feature set that seems to be doing bulk of what the article discusses.
The faq section in the site describes how they calculate a Star Power
The SezWho scoring algorithm is a proprietary
page-rank-like recursive algorithm based on reputation score of rater
and commenter, their frequency of participation, time of interaction,
consistency of participation and topic of discussion.
Enterprises need to take a leaf out of the consumer internet space
in understanding reputation mechanisms. This will be crucial in any
Enterprise 2.0 strategy. There are central skills and expertise
management systems in many enterprises that employees don’t update.
Enterprise variations of large scale social platforms like Yahoo
Answers would provide the opportunity for enterprises to leverage the
collective intelligence of their workforce effectively. These are the
kind of social platforms that will allow experts to emerge from the
edges of the organization. And to identify these “Answer people” (as
the article calls these experts) in your enterprise you would need to
- Give them a participatory platform to showcase what they know and
- Devise smart reputation mechanisms that are partly automated and
partly community driven. Reputation needs to be portable across the
enterprise information ecosystem – a SezWho kind of distributed system
for the enterprise.