This paper (first reference) is the result of a class project I was part of
almost two years ago for CSCI 5417 Information Retrieval Systems. It builds
on a class project I did in CSCI 5832 Natural Language Processing and which
I presented at Wikimania '07. The project was very late as we didn't send
the final paper in until the day before new years. This technical report was
never really announced that I recall so I thought it would be interesting to
look briefly at the results. The goal of this paper was to break articles
down into surface features and latent features and then use those to study
the rating system being used, predict article quality and rank results in a
search engine. We used the [[random forests]] classifier which allowed us to
analyze the contribution of each feature to performance by looking directly
at the weights that were assigned. While the surface analysis was performed
on the whole english wikipedia, the latent analysis was performed on the
simple english wikipedia (it is more expensive to compute). = Surface
features = * Readability measures are the single best predictor of quality
that I have found, as defined by the Wikipedia Editorial Team (WET). The
[[Automated Readability Index]], [[Gunning Fog Index]] and [[Flesch-Kincaid
Grade Level]] were the strongest predictors, followed by length of article
html, number of paragraphs, [[Flesh Reading Ease]], [[Smog Grading]], number
of internal links, [[Laesbarhedsindex Readability Formula]], number of words
and number of references. Weakly predictive were number of to be's, number
of sentences, [[Coleman-Liau Index]], number of templates, PageRank, number
of external links, number of relative links. Not predictive (overall - see
the end of section 2 for the per-rating score breakdown): Number of h2 or
h3's, number of conjunctions, number of images*, average word length, number
of h4's, number of prepositions, number of pronouns, number of interlanguage
links, average syllables per word, number of nominalizations, article age
(based on page id), proportion of questions, average sentence length. :*
Number of images was actually by far the single strongest predictor of any
class, but only for Featured articles. Because it was so good at picking out
featured articles and somewhat good at picking out A and G articles the
classifier was confused in so many cases that the overall contribution of
this feature to classification performance is zero. :* Number of external
links is strongly predictive of Featured articles. :* The B class is highly
distinctive. It has a strong "signature," with high predictive value
assigned to many features. The Featured class is also very distinctive. F, B
and S (Stop/Stub) contain the most information.
:* A is the least distinct class, not being very different from F or G. =
Latent features = The algorithm used for latent analysis, which is an
analysis of the occurence of words in every document with respect to the
link structure of the encyclopedia ("concepts"), is [[Latent Dirichlet
Allocation]]. This part of the analysis was done by CS PhD student Praful
Mangalath. An example of what can be done with the result of this analysis
is that you provide a word (a search query) such as "hippie". You can then
look at the weight of every article for the word hippie. You can pick the
article with the largest weight, and then look at its link network. You can
pick out the articles that this article links to and/or which link to this
article that are also weighted strongly for the word hippie, while also
contributing maximally to this articles "hippieness". We tried this query in
our system (LDA), Google (site:en.wikipedia.org hippie), and the Simple
English Wikipedia's Lucene search engine. The breakdown of articles occuring
in the top ten search results for this word for those engines is: * LDA
only: [[Acid rock]], [[Aldeburgh Festival]], [[Anne Murray]], [[Carl
Radle]], [[Harry Nilsson]], [[Jack Kerouac]], [[Phil Spector]], [[Plastic
Ono Band]], [[Rock and Roll]], [[Salvador Allende]], [[Smothers brothers]],
[[Stanley Kubrick]]. * Google only: [[Glam Rock]], [[South Park]]. * Simple
only: [[African Americans]], [[Charles Manson]], [[Counterculture]], [[Drug
use]], [[Flower Power]], [[Nuclear weapons]], [[Phish]], [[Sexual
liberation]], [[Summer of Love]] * LDA & Google & Simple: [[Hippie]],
[[Human Be-in]], [[Students for a democratic society]], [[Woodstock
festival]] * LDA & Google: [[Psychedelic Pop]] * Google & Simple: [[Lysergic
acid diethylamide]], [[Summer of Love]] ( See the paper for the articles
produced for the keywords philosophy and economics ) = Discussion /
Conclusion = * The results of the latent analysis are totally up to your
perception. But what is interesting is that the LDA features predict the WET
ratings of quality just as well as the surface level features. Both feature
sets (surface and latent) both pull out all almost of the information that
the rating system bears. * The rating system devised by the WET is not
distinctive. You can best tell the difference between, grouped together,
Featured, A and Good articles vs B articles. Featured, A and Good articles
are also quite distinctive (Figure 1). Note that in this study we didn't
look at Start's and Stubs, but in earlier paper we did. :* This is
interesting when compared to this recent entry on the YouTube blog. "Five
Stars Dominate Ratings"
I think a sane, well researched (with actual subjects) rating system
well within the purview of the Usability Initiative. Helping people find and
create good content is what Wikipedia is all about. Having a solid rating
system allows you to reorganized the user interface, the Wikipedia
namespace, and the main namespace around good content and bad content as
needed. If you don't have a solid, information bearing rating system you
don't know what good content really is (really bad content is easy to spot).
:* My Wikimania talk was all about gathering data from people about articles
and using that to train machines to automatically pick out good content. You
ask people questions along dimensions that make sense to people, and give
the machine access to other surface features (such as a statistical measure
of readability, or length) and latent features (such as can be derived from
document word occurence and encyclopedia link structure). I referenced page
262 of Zen and the Art of Motorcycle Maintenance to give an example of the
kind of qualitative features I would ask people. It really depends on what
features end up bearing information, to be tested in "the lab". Each word is
an example dimension of quality: We have "*unity, vividness, authority,
economy, sensitivity, clarity, emphasis, flow, suspense, brilliance,
precision, proportion, depth and so on.*" You then use surface and latent
features to predict these values for all articles. You can also say, when a
person rates this article as high on the x scale, they also mean that it has
has this much of these surface and these latent features.
= References =
- DeHoust, C., Mangalath, P., Mingus., B. (2008). *Improving search in
Wikipedia through quality and concept discovery*. Technical Report.
- Rassbach, L., Mingus., B, Blackford, T. (2007). *Exploring the
feasibility of automatically rating online article quality*. Technical
I have asked and received permission to forward to you all this most
excellent bit of news.
The linguist list, is a most excellent resource for people interested in the
field of linguistics. As I mentioned some time ago they have had a funding
drive and in that funding drive they asked for a certain amount of money in
a given amount of days and they would then have a project on Wikipedia to
learn what needs doing to get better coverage for the field of linguistics.
What you will read in this mail that the total community of linguists are
asked to cooperate. I am really thrilled as it will also get us more
linguists interested in what we do. My hope is that a fraction will be
interested in the languages that they care for and help it become more
relevant. As a member of the "language prevention committee", I love to get
more knowledgeable people involved in our smaller projects. If it means that
we get more requests for more projects we will really feel embarrassed with
all the new projects we will have to approve because of the quality of the
Incubator content and the quality of the linguistic arguments why we should
approve yet another language :)
NB Is this not a really clever way of raising money; give us this much in
this time frame and we will then do this as a bonus...
---------- Forwarded message ----------
From: LINGUIST Network <linguist(a)linguistlist.org>
Date: Jun 18, 2007 6:53 PM
Subject: 18.1831, All: Call for Participation: Wikipedia Volunteers
LINGUIST List: Vol-18-1831. Mon Jun 18 2007. ISSN: 1068 - 4875.
Subject: 18.1831, All: Call for Participation: Wikipedia Volunteers
Moderators: Anthony Aristar, Eastern Michigan U <aristar(a)linguistlist.org>
Helen Aristar-Dry, Eastern Michigan U <hdry(a)linguistlist.org>
Reviews: Laura Welcher, Rosetta Project
The LINGUIST List is funded by Eastern Michigan University,
and donations from subscribers and publishers.
Editor for this issue: Ann Sawyer <sawyer(a)linguistlist.org>
To post to LINGUIST, use our convenient web form at
From: Hannah Morales < hannah(a)linguistlist.org >
Subject: Wikipedia Volunteers
-------------------------Message 1 ----------------------------------
Date: Mon, 18 Jun 2007 12:49:35
From: Hannah Morales < hannah(a)linguistlist.org >
Subject: Wikipedia Volunteers
As you may recall, one of our Fund Drive 2007 campaigns was called the
"Wikipedia Update Vote." We asked our viewers to consider earmarking their
donations to organize an update project on linguistics entries in the
English-language Wikipedia. You can find more background information on this
The speed with which we met our goal, thanks to the interest and generosity
our readers, was a sure sign that the linguistics community was enthusiastic
about the idea. Now that summer is upon us, and some of you may have a bit
leisure time, we are hoping that you will be able to help us get started on
Wikipedia project. The LINGUIST List's role in this project is a purely
organizational one. We will:
*Help, with your input, to identify major gaps in the Wikipedia materials or
pages that need improvement;
*Compile a list of linguistics pages that Wikipedia editors have identified
"in need of attention from an expert on the subject" or " does not cite any
references or sources," etc;
*Send out periodical calls for volunteer contributors on specific topics or
*Provide simple instructions on how to upload your entries into Wikipedia;
*Keep track of our project Wikipedians;
*Keep track of revisions and new entries;
*Work with Wikimedia Foundation to publicize the linguistics community's
We hope you are as enthusiastic about this effort as we are. Just to help us
get started looking at Wikipedia more critically, and to easily identify an
needing improvement, we suggest that you take a look at the List of
Many people are not listed there; others need to have more facts and
added. If you would like to participate in this exciting update effort,
respond by sending an email to LINGUIST Editor Hannah Morales at
hannah(a)linguistlist.org, suggesting what your role might be or which
entries you feel should be updated or added. Some linguists who saw our
on the Internet have already written us with specific suggestions, which we
share with you soon.
This update project will take major time and effort on all our parts. The
result will be a much richer internet resource of information on the breadth
depth of the field of linguistics. Our efforts should also stimulate
students to consider studying linguistics and to educate a wider public on
we do. Please consider participating.
Editor, Wikipedia Update Project
Linguistic Field(s): Not Applicable
LINGUIST List: Vol-18-1831
There is a request for a Wikipedia in Ancient Greek. This request has so far
been denied. A lot of words have been used about it. Many people maintain
their positions and do not for whatever reason consider the arguments of
In my opinion their are a few roadblocks.
- Ancient Greek is an ancient language - the policy does not allow for
- Text in ancient Greek written today about contemporary subjects
require the reconstruction of Ancient Greek.
- it requires the use of existing words for concepts that did
not exist at the time when the language was alive
- neologisms will be needed to describe things that did not
exist at the time when the language was alive
- modern texts will not represent the language as it used to be
- Constructed and by inference reconstructed languages are effectively
We can change the policy if there are sufficient arguments, when we agree on
When a text is written in reconstructed ancient Greek, and when it is
clearly stated that it is NOT the ancient Greek of bygone days, it can be
obvious that it is a great tool to learn skills to read and write ancient
Greek but that it is in itself not Ancient Greek. Ancient Greek as a
language is ancient. I have had a word with people who are involved in the
working group that deals with the ISO-639, I have had a word with someone
from SIL and it is clear that a proposal for a code for "Ancient Greek
reconstructed" will be considered for the ISO-639-3. For the ISO-639-6 a
code is likely to be given because a clear use for this code can be given.
We can apply for a code and as it has a use bigger then Wikipedia alone it
clearly has merit.
With modern texts clearly labelled as distinct from the original language,
it will be obvious that innovations a writers needs for his writing are
This leaves the fact that constructed and reconstructed languages are not
permitted because of the notion that mother tongue users are required. In my
opinion, this has always been only a gesture to those people who are dead
set against any and all constructed languages. In the policies there is
something vague "*it must have a reasonable degree of recognition as
determined by discussion (this requirement is being discussed by the language
subcommittee <http://meta.wikimedia.org/wiki/Language_subcommittee>)."* It
is vague because even though the policy talks about a discussion, it is
killed off immediately by stating "The proposal has a sufficient number of
living native speakers to form a viable community and audience." In my
opinion, this discussion for criteria for the acceptance of constructed or
reconstructed languages has not happened. Proposals for objective criteria
have been ignored.
In essence, to be clear about it:
- We can get a code for reconstructed languages.
- We need to change the policy to allow for reconstructed and
We need to do both in order to move forward.
The proposal for objective criteria for constructed and reconstructed
languages is in a nutshell:
- The language must have an ISO-639-3 code
- We need full WMF localisation from the start
- The language must be sufficiently expressive for writing a modern
- The Incubator project must have sufficiently large articles that
demonstrate both the language and its ability to write about a wide range of
- A sufficiently large group of editors must be part of the Incubator
-----BEGIN PGP SIGNED MESSAGE-----
Next Thursday's office hours will feature Véronique Kessler, the
Foundation's Chief Financial Officer. If you don't know
Naoko, you can get to know her at
Office hours on Thursday are from 2100 to 2200 UTC (3:00 PM - 4:00 PM PDT).
If you do not have an IRC client, there are two ways you can come chat
using a web browser: First is using the Wikizine chat gateway at
<http://chatwikizine.memebot.com/cgi-bin/cgiirc/irc.cgi>. Type a
nickname, select irc.freenode.net from the top menu and
#wikimedia-office from the following menu, then login to join.
Also, you can access Freenode by going to http://webchat.freenode.net/,
typing in the nickname of your choice and choosing wikimedia-office as
the channel. You may be prompted to click through a security warning.
It should be all right.
Please feel free to forward (and translate!) this email to any other
relevant email lists you happen to be on. Also note, this is
Veronique's first foray into IRC, so lets show her how welcoming we can
Volunteer Coordinator, Wikimedia Foundation
Support Free Knowledge: http://wikimediafoundation.org/wiki/Donate
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
-----END PGP SIGNATURE-----
This is a follow up to my proposal that Fan History Wiki join the wMF
family, based on my experiences via e-mail, on the list and on strategy
wiki. This isn't as coherent as I would like.
To give some back story that might not have been as obvious in our initial
proposal, we were interested in joining the WMF for several primary reasons:
1) Wikimedia Foundation is a non-profit organization where we fundamentally
believe in the objectives of the organization.
2) Wikimedia Foundation has credibility that could be extended to our
project, helping us accomplish our mission.
3) Fan History and I feel like we have positive relationships with Wikimedia
Foundation staff, based on our interactions at RecentChangesCamp and in
various chat rooms like #wiki and #mediawiki on irc.freenode.net.
We have certain things that we want to accomplish that have been detailed
elsewhere. We're in the process of looking for and determining if we should
partner with some one to accomplish these goals, what we're looking for in
partnering or being acquired by some one. Our general criteria have been:
1) Non-profit, no-profit or for profit business where the emphasis would be
on helping us to succeed with our mission. Monetization of the project is
fine so long as major content focus and creation is focused around
monetizing. We see our project as fundamentally for a greater good, to
preserve and document the history of fan communities, and we don't want that
made secondary to commercial interests.
2) Financial issues. In an ideal world, we would want one or two or three
of our staffers to get some form of compensation for helping to maintain the
content, enforce policies and helping work towards the mission. We also
want to make sure that the project has the funding to continue indefinetely.
3) Fix and improve our back end. Thankfully, it feels like a fair amount of
this has been addressed in the past two weeks so we're much less stressed
about this than we were.
4) Increase the visibility and credibility of our project. Get more people
That out of the way, time to discuss the process of trying to get acquired
by the Wikimedia Foundation. Simply put, there were three basic steps that
1) Contacted members of the Wikimedia Foundation to ask them if they would
be interested in bringing Fan History into the Wikimedia Foundation family.
Got directed to other people, told not sure who in the organization this
would be best proposed to, got told that the Foundation itself probably
wouldn't be interest, finally suggested I post this on the list because if
community consensus is yes than we can go ahead.
2) Posted the proposal on the mailing list. Good feedback. Suggested I
post it to the Strategy Wiki.
3) Posted to strategy wiki.
Step one is fine. The only problem I might have had with step one was not
getting out right rejected.
The problem is really when it comes to steps two and three. To my
knowledge, all of the projects that are currently part of the Wikimedia
Foundation are home grown; they did not join as part of any aquisition
process. In this regards, our proposal was unique. Steps two and three
are kind of where we got hung up: What is the timeline? What are the next
steps to take after these?
The timeline issue is a big one. For us, this is not that big of a deal
necessarily. We're finacially in a place where we can probably chug along
for a while in that regards. We're not facing issues of possibly being shut
down because of legal problems or scripting problems. We do not have issues
that say this is a last resort option for us to keep us open. In the
future, others may contact Wikipedia where this may be an issue or where the
founders may see this as the only solution.
I tried to ask various people to get a feel for the timeline that we were
looking at to, well, know if WMF was interested in acquiring us or setting
up some sort of official relationship. What I got told by people in the
know on #wikimedia-strategy was that we were looking at three to eighteen
months before we got some sort of official response back regarding whether
this was something that the Wikimedia Foundation community was interested
in. I was left with the impression that unless I was basically agitating
and trying to get support on an almost daily basis, the timeline was
actually closer to eighteen months. EIGHTEEN MONTHS.
And that eighteen months didn't even factor in under what conditions WMF
would want us, what we would have to give up, and if we'd even be willing to
accept WMF's offer if they made that decision. We couldn't get the
information that we needed to know if we even wanted to work with Wikimedia
Foundation for that long.
That puts us in a bind. There are other places we would like to approach.
(And if you have ideas for who would be a good fit, please get in touch with
me.) Is it ethical for us to approach other people and organizations while
we have this on the table with the WMF? If we approach other people in the
mean time, does that signal that our interest in the WMF is dead? We just
don't know. Is it fair to the WMF to basically keep us in limbo for three
to eighteen months? We kind of don't think so.
Outside of the timeline issue, there is the whole procedural issue.
Proposal is made. It is posted to the appropriate places that WMF employees
and volunteers have guided us to. What are the next steps? There really
hasn't been any clarity for us on this. When we've asked in the chat, the
answer is persistence. That's not a step and it isn't particularly
helpful. Persistence how? Post repeatedly to the mailing list? Troll talk
pages of contributors to Strategy Wiki asking for them to vote yes, Fan
History should be part of the WMF?
Looking around at other new content proposals, Strategy Wiki looks like a
place where proposals go to die. There are proposals that have been there a
year, that have no votes, with no comments on them. There doesn't appear to
be any follow up by 1) the person who posted the proposal, 2) volunteers for
the WMF, 3) WMF employees who are working on strategy wiki. This is not
encouraging for several reasons. If you are really excited about your
content proposal and it would be a really good fit, you have the time to
basically put full force into launching the project in the next three to six
months, you've followed the guidelines that I have gotten and the procedure
on the wiki, time and procedure become a huge issue that are potentially
By deterents, I mean that these projects could end up off WMF and you could
lose contributors who can't get past the bureaucracy to accomplish their
goals, are not insider enough to push to get things done, are going to deter
future proposals and may deter future contributions to Wikipedia and related
As some one who has proposed a new project for the WMF (which would really
probably be an acquisition if it happened), some changes need to be made:
1) Clear procedure for what happens step by step in making such a proposal.
Post proposal. Contact people who support your position to vote in favor of
it using talk pages on Strategy wiki. After one hundred votes vast in favor
with no more than half that total in opposition, project moves to
development stages where WMF staff will be in contact with the person making
the proposal. Something like that.
2) Clear timeline of what happens and when so that people can plan
3) Expectations regarding exclusivity of proposal to the WMF during the
proposal process. Can people propose it elsewhere or seek acquisition by
others while there is an open proposal on Strategy Wiki?
As for where Fan History's proposal to join WMF stands now, we're not sure.
The mailing list conversation died. Strategy wiki's only commentary has
been regarding getting us off the blacklist for Foundation projects.
With our last board meeting falling a little later in the year than
usual, and coming close to holidays, I'm a little late in giving this
brief report on what happened. As you know, the board approved the
audited financial statements for the 2008-2009 fiscal year, and those
were posted on the Wikimedia Foundation website a couple weeks ago.
As the organization has gotten settled in San Francisco, this has become
a smoother process than in years past. It's not just that we're on solid
financial ground (that's good, though even in the past we haven't been
in immediate danger financially), but also as we've established the
necessary infrastructure, we're better at tracking our finances and
addressing issues that may need to be resolved in the course of an
audit. The stress level around the audit is low enough now that although
I had to miss the audit committee conference call (where I normally
participate as an observer), I still had confidence that Veronique and
Stu and everyone else would bring it to a successful conclusion. (That's
my personal stress level I speak of, I won't claim there's no stress
involved for Veronique.) Thanks to Veronique and her team, as well as
the auditors, for their work. Thanks also to the volunteers on the audit
committee for their service. One other thing the board reviewed out of
the committee's work was a risks analysis that is being put together of
the top risks and mitigation strategies for addressing them, which will
become part of the strategic planning process.
Speaking of which, strategic planning was the biggest single item on our
agenda, and tended to be a thread running throughout the rest of the
meeting as well. With help from the Bridgespan team, we worked through
some preliminary strategy questions to discuss priorities and setting
goals. We don't have concrete results from that to share, in part
because the overall process is still in an early stage, but it was a
good exercise for us in thinking on a strategic level and preparing for
the more challenging decisions ahead. One thing that did come out of it
is that we reviewed the guidance that the nominating committee was given
in the search to fill the one remaining vacancy on the board, and
provided some additional input for them as a result.
One note I would add is the valuable contributions of the newer members
of the board in bringing a more complete set of views to the group. SJ
and Arne many of you know, and their emphasis on the health and
potential of the community and chapters was important. I also really
appreciated what Matt Halprin had to offer, especially in two areas:
One, his capacity for strategic thinking and helping to keep us focused
on strategy issues instead of drifting off-track, and two, his
experience with other nonprofit boards and their practices was a good
perspective to add. Their strengths help compensate for the fact that
Jimmy wasn't able to attend the meeting, and I anticipate they will make
us a good working group through the strategic planning process.
The board also conducted a sort of general review and evaluation, not
just of the financial situation but the state of the organization and
the work being done by Sue and the staff. We think Sue is great as our
Executive Director and has done a terrific job getting the organization
to this point, where we're healthy and have the space to develop our
capacity for achievement in the years ahead. And I can't say enough
about the work the staff are doing, so I'll just share this one
anecdote. Since the board meeting overlapped with the first week of the
fundraiser, it's fortunate that we have the new office where there's
more than one meeting room, otherwise I don't know where the board would
have met. That's because the fundraising team would spend all day
working together in one of the larger meeting rooms. This would involve
as many as 6-8 people, and you wouldn't know that they eat or take
breaks from casual observation, although we did make sure they at least
got the leftover pizza from our lunch at one point. They've worked hard
to keep the fundraiser operational, not to mention considering a lot of
feedback and making improvements. It's easier when you see what things
are like in the office, but I hope we appreciate the dedicated staff and
their support for the work of the Wikimedia movement.
Per the new posting limits <
each user is limited to 30 posts per month, after which they are put on
moderation. Anthony has reached 30 posts. He has been placed on moderation
for about the next 19 hours or so (until about Midnight UTC, or whenever one
of us happens to be at a computer around that time).
Continued input on these policies, either publicly or privately, is always