A paper abstract to the list would be helpful, along with a clear statement about how recommenders might be used here. In terms of scholarly impact, there are efforts being made (this week) at the acm altmetrics workshop to imagine new ways of measurement. Group lens at Minnesota published a good deal about recommenders a decade ago. There is now an acm conference dedicated to recommenders. One challenge is that starring systems tend to compress ratings near the top in practice. See the links below. http://chronicle.com/article/As-Scholarship-Goes-Digital/130482/ http://altmetrics.org/altmetrics12/ http://recsys.acm.org/2012/ Thanks! Sean P. Goggins, Ph.D http://www.groupinformatics.org Visit http://www.sociotech.net Phone: (215) 948-2729 "Never doubt that a small group of thoughtful, committed citizens can change the world. Indeed, it is the only thing that ever has." -- Margaret Mead "The most effective way to do it, is to do it." -- Amelia Earhart On Jun 18, 2012, at 4:16, Meelis Ojasild <meelis.ojasild@gmail.com> wrote:
Perhaps it's my subjective angle and isn't necessarily representative, but for me it's rather weird that the interactions take place through an e-mail list. It feels like the '90s.
Why aren't we using better tools like wikis, blogs, collaborative blogs etc? It would solve the tagging and recommendation problem as well.
Meelis
On Sat, Jun 16, 2012 at 6:27 PM, Murray Turoff <murray.turoff@gmail.com>wrote:
not everyone in a professional community can read everything of possible interest, the most common problem we all face is information overload. if you look at hte paper it is based upon a study of another professional community.
what you would be doing is collaborative tagging to create your own evolving index for the group as a whole and then voting on the "importance" of any paper entered by someone, but voting and indexing it by those that have read it. the paper suggest that the members would characterize their interests by using the same index to represent themselves and the voting would be summarized by the keys put on the paper.
the paper suggests using thurstones law of comparative judgement so one can see the strength of the group agreements by distance between the ranked papers. however, a simple five star rating would work to start with. On Sat, Jun 16, 2012 at 12:38 PM, Alexander Furnas <zfurnas@gmail.com
wrote:
do it would be books, papers, reports, drafts, or anything on the general topic including maybe standard changes, etc.
You are the group that should be using the technology you write about. "A seer upon perceiving a flood should be the first to climb a tree"- kalil gibron
A recommender system for what? Academic articles? News of interest? Job
postings? Because I think existing platforms could serve these purposes - we could just create a Air Mendely group or something (does one already exist)? Perhaps an Air subreddit?
That said, I agree that some Air collaborative filtering might be a more useful way to surface things of interest to the community than just email blasts.
On Jun 16, 2012, at 11:22 AM, Murray Turoff wrote:
this group could make a wonderful demonstration by adding a recommender system to your operation. A great phd project
Turoff, M., Hiltz, S.R.: The Future of Professional Communities of Practice. In: Weinhardt, C., Luckner, S., Stößer, J. (eds.) WeB 2008. LNBIP, vol. 22, pp. 144-158. Springer-Verlag, Berlin Heidelberg (2009)
You would be the perfect group to demonstrate the benefits of adding that capability. -- *Distinguished Professor Emeritus Information Systems, NJIT homepage: http://is.njit.edu/turoff * _______________________________________________ The Air-L@listserv.aoir.org mailing list is provided by the Association of Internet Researchers http://aoir.org Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
Join the Association of Internet Researchers: http://www.aoir.org/
-- *Distinguished Professor Emeritus Information Systems, NJIT homepage: http://is.njit.edu/turoff * _______________________________________________ The Air-L@listserv.aoir.org mailing list is provided by the Association of Internet Researchers http://aoir.org Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
Join the Association of Internet Researchers: http://www.aoir.org/
_______________________________________________ The Air-L@listserv.aoir.org mailing list is provided by the Association of Internet Researchers http://aoir.org Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org
Join the Association of Internet Researchers: http://www.aoir.org/