Hi all;
I'm creating a census[1] with all the anti-vandalism bots in the Wikimedia projects history. I want to research the features and techniques used in all these past years. I need your help for compiling all the nicks of those bots. You can help adding info to the page, but if you don't have free time for that, write only the nickname and I will retrieve all the available info about the bot.
With the currently available information, I have found two main categories of anti-vandalism bots: * First generation: simple scoring systems based in regular expressions and heuristics. * Second generation: machine learning, neural networks and bayesian filters.
Have you got suggestions to this classification? What is your opinion about the past and the future of anti-vandalism bots? Can FlaggedRevs and similar approaches make these bots useless?
Thanks, emijrp
[1] http://en.wikipedia.org/wiki/User:Emijrp/Anti-vandalism_bot_census#Census
---------- Forwarded message ---------- From: emijrp emijrp@gmail.com Date: Mon, Nov 29, 2010 at 5:35 AM Subject: [Foundation-l] Anti-vandalism bot census To: Wikimedia Foundation Mailing List foundation-l@lists.wikimedia.org, Research into Wikimedia content and communities < wiki-research-l@lists.wikimedia.org>
Hi all;
I'm creating a census[1] with all the anti-vandalism bots in the Wikimedia projects history. I want to research the features and techniques used in all these past years. I need your help for compiling all the nicks of those bots. You can help adding info to the page, but if you don't have free time for that, write only the nickname and I will retrieve all the available info about the bot.
With the currently available information, I have found two main categories of anti-vandalism bots: * First generation: simple scoring systems based in regular expressions and heuristics. * Second generation: machine learning, neural networks and bayesian filters.
Have you got suggestions to this classification? What is your opinion about the past and the future of anti-vandalism bots? Can FlaggedRevs and similar approaches make these bots useless?
Thanks, emijrp
[1] http://en.wikipedia.org/wiki/User:Emijrp/Anti-vandalism_bot_census#Census _______________________________________________ foundation-l mailing list foundation-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
Thanks to Cobi, Arkanosis and Andrew West for adding some bots and tools to the census. If you have information about anti-vandalism bots in your local language Wikipedia, please, notice me.
At the moment, ClueBot NG is the only bot which uses neural networks. The second generation of bots is coming. : O
Regards, emijrp
2010/11/29 emijrp emijrp@gmail.com
Hi all;
I'm creating a census[1] with all the anti-vandalism bots in the Wikimedia projects history. I want to research the features and techniques used in all these past years. I need your help for compiling all the nicks of those bots. You can help adding info to the page, but if you don't have free time for that, write only the nickname and I will retrieve all the available info about the bot.
With the currently available information, I have found two main categories of anti-vandalism bots:
- First generation: simple scoring systems based in regular expressions and
heuristics.
- Second generation: machine learning, neural networks and bayesian
filters.
Have you got suggestions to this classification? What is your opinion about the past and the future of anti-vandalism bots? Can FlaggedRevs and similar approaches make these bots useless?
Thanks, emijrp
[1] http://en.wikipedia.org/wiki/User:Emijrp/Anti-vandalism_bot_census#Census
emijrp,
thanks for starting this survey. I'd love to see more scholarly work specifically focussed on bots in Wikipedia, how they affect the volume and quality of human contributions, article assessment, vandalism detection or editor recruitment and also how humans and bots interact and whether the design of bots can be informed by empirical results. Apart from the literature on vandalism detection and discussions of bot edits in the context of general research on edit patterns in Wikipedia, I am not familiar with works specifically targeting this topic. Stuart Geiger has done some ethnographic work on the social role of bots, I'd like to have pointers to more quantitative studies on the topic.
Dario
On 30 Nov 2010, at 13:01, emijrp wrote:
Thanks to Cobi, Arkanosis and Andrew West for adding some bots and tools to the census. If you have information about anti-vandalism bots in your local language Wikipedia, please, notice me.
At the moment, ClueBot NG is the only bot which uses neural networks. The second generation of bots is coming. : O
Regards, emijrp
2010/11/29 emijrp emijrp@gmail.com Hi all;
I'm creating a census[1] with all the anti-vandalism bots in the Wikimedia projects history. I want to research the features and techniques used in all these past years. I need your help for compiling all the nicks of those bots. You can help adding info to the page, but if you don't have free time for that, write only the nickname and I will retrieve all the available info about the bot.
With the currently available information, I have found two main categories of anti-vandalism bots:
- First generation: simple scoring systems based in regular expressions and heuristics.
- Second generation: machine learning, neural networks and bayesian filters.
Have you got suggestions to this classification? What is your opinion about the past and the future of anti-vandalism bots? Can FlaggedRevs and similar approaches make these bots useless?
Thanks, emijrp
[1] http://en.wikipedia.org/wiki/User:Emijrp/Anti-vandalism_bot_census#Census
Wiki-research-l mailing list Wiki-research-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wiki-research-l
wiki-research-l@lists.wikimedia.org