QUALITY ASSESSMENT OF WIKIPEDIA EXTERNAL LINKS

Paraskevi Tzekou

, Sofia Stamou

, Nikos Kirtsis

and Nikos Zotos

Computer Engineering and Informatics Department, Patras University 26500 Patras, Greece

Department of Archives and Library Science, Ionian University, 49100 Ionian, Greece

Keywords: Wikipedia, External links decay, Measurements, Quality assessment.

Abstract: Wikipedia is a unique source of information that has been collectively supplied by thousands of people.

Since its nascence in 2001, Wikipedia is continuously evolving and like most websites it is interconnected

via hyperlinks to other web information sources. Wikipedia articles contain two types of links: internal and

external. Internal links point to other Wikipedia articles, while external links point outside Wikipedia and

normally they are not used in the body of the article. Although there exist specific guidelines about both the

style and the purpose of the article external links, no approach has been recorded that tries to capture in a

systematic manner the quality of Wikipedia external links. In this paper, we study the quality of Wikipedia

external links by assessing the degree to which these conform to their intended purpose; that is to formulate

a comprehensive list of accurate information sources about the article contents. For our study, we estimate

the decay of Wikipedia external links and we investigate their distribution in the Wikipedia articles. Our

measurements give perceptible evidence for the value of external links and may imply their corresponding

articles' quality in a holistic Wikipedia evaluation.

1 INTRODUCTION

Wikipedia is a dominant source of online informa-

tion for millions of people. To date, the English ver-

sion of Wikipedia hosts over 3 million articles

and

it is constantly evolving as new material is been pro-

vided by numerous editors who work on a volunteer,

collaborative basis. The most fascinating thing about

Wikipedia is its open nature, which enables people

to create, modify and/or extend an article and which

has turned Wikipedia into an unprecedented source

of collectively-supplied information.

However, Wikipedia's open nature has raised

scepticism about the quality of its hosting articles

and has paved the ground for studies that investigate

the automatic assessment of Wikipedia's quality. In

this respect, numerous quality assessment methods

have been proposed, most of which rely on the

cross-examination of Wikipedia internal characteris-

tics, e.g. the articles' length (Blumenstock, 2008a),

their contextual elements (Stvilia et al., 2005a), the

number of edits (Buriol et al., 2006), their linkage

within the Wikipedia graph (Kamps and Koolen,

2009), the formality of their language (Emigh and

Herring, 2005), their factual accuracy (Giles, 2005),

http://en.wikipedia.org/wiki/Special:Statistics

etc. But Wikipedia, like almost any other website,

provides links from its articles to external (i.e. out-

side Wikipedia) web pages. The scope of external

links is to provide supplementary information about

the article's content that is accurate and which could

or should not be added to the article for reasons such

as copyright or amount of detail

. Although external

links are intended to provide complementary infor-

mation about the articles' contents, they have not

been systematically accounted in the evaluation of

Wikipedia's quality.

In this paper, we investigate the role of Wikipe-

dia external links in conveying accurate information

about the article contents. Our investigation is not

confined to external links appearing under a ho-

monymous section at the very end of articles, but

rather it addresses the role of all external links

within Wikipedia articles. The motive for our study

is to quantify the external links' usefulness and merit

towards improving the articles’ comprehensiveness

and explore them as implicit signals of the articles'

quality. Our investigation relies on the intuition that

the quality of an article is determined based not only

on the accuracy of the information that the article

per se brings, but also on the usefulness and the ac-

curacy of its pointing external resources.

http://en.wikipedia.org/wiki/Wikipedia:External_links

248

Tzekou P., Stamou S., Kirtsis N. and Zotos N..

QUALITY ASSESSMENT OF WIKIPEDIA EXTERNAL LINKS.

DOI: 10.5220/0003299502480254

In Proceedings of the 7th International Conference on Web Information Systems and Technologies (WEBIST-2011), pages 248-254

ISBN: 978-989-8425-51-5

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

In the course of our study, we built upon earlier

work (Kirtsis et al., 2010) in which we assessed the

amount to which Wikipedia external links add new

content to their corresponding articles’ body and we

make the below-listed contributions:

 We examine the statistical distribution of exter-

nal links in Wikipedia articles, we identify popular

domain names in the articles' external resources and

we capture the correlation between the articles'

length and the fraction of their external links. Our

examination helps assess the appropriateness of the

articles’ external links, since according to Wikipedia

editing guidelines, external links should point to

information that is not yet part of the article.

 We quantify decay in Wikipedia external links in

an attempt to assist editors and administrators de-

termine the amount of repair that the articles' linked

sources should undergo. To measure decay of an

article's external pages, we use the proportion of

dead links to which the article points. A dead link

points to a page that either is no longer available or

redirects to a page that has nothing to do with the

original page. Measuring decay of Wikipedia linked-

to resources indicates the appropriateness of the arti-

cles' linking pages in providing readers with reliable

yet valid information about the subject of the article.

The remainder of the paper is organized as follows.

We start our discussion with an overview on related

work. In Section 3, we describe how external links

can serve as implicit indicators of Wikipedia’s qual-

ity. Specifically, we discuss how to capture and in-

terpret the distribution of external links in Wikipedia

articles and we address the problem of identifying

external links' decay. In Section 4, we present an

experiment we carried out in which we assessed the

usefulness of external links for communicating valid

and complete information about the Wikipedia arti-

cle contents. The discussion of experimental results

sheds light on the article external features to be con-

sidered in future Wikipedia quality assessment ef-

forts. We conclude the paper in Section 5 where we

outline our plans for future work.

2 RELATED WORK

Related work falls in two main categories, which we

discuss in turn. First, there is related work in study-

ing the nature and the quality of Wikipedia. In this

respect, researchers have suggested a number of

article features that signify quality, e.g. their articles’

survival period (Cross, 2006), the number and fre-

quency of their edits (Wilkinson and Huberman,

2007), their revision history (Adler and de Alfaro,

2007), the amount of outbound citations to scientific

publications (Nielsen, 2007), the dedication and ex-

pertise of their editors (Riehle, 2005), etc. Our work

extends prior studies that infer the Wikipedia arti-

cles’ quality based on their contents investigation in

that we also examine the impact of external links to

their corresponding articles’ quality. In our work, we

significantly enrich our earlier study (Kirtsis et al.,

2010) by examining additional features in assessing

the article links’ and by running large-scale evalua-

tion experiments on the Wikipedia outlinks’ distri-

bution and properties.

In a similar direction, (Buriol et al., 2006) study

the evolution of the Wikipedia link graph over time

and observe an increasing link density in Wikipedia

as time goes by. This is also attested in the work of

(Kamps and Koolen, 2009), who contrast the link

structure of the Web and Wikipedia in an attempt to

assess the impact of global and local link topology in

web retrieval effectiveness. Although our work re-

lates to the above studies that examine the Wikipe-

dia link structure, it is different in that we investigate

the contribution of external resources in comple-

menting the article contents. Most importantly, we

estimate not only external links' distribution but also

their decay. Although, the encyclopaedic organiza-

tion of Wikipedia is different from the Web data

organization in that it is steered by specific guide-

lines about what and how to link

still Wikipedia

like all large websites suffers from the link rot phe-

nomenon. As of November 2006 dump, nearly 10%

of external links were broken and although repairing

and maintenance actions are been taken ever since,

the problem of rotten external links persists.

Thus, our study also relates to existing works

that focus on the identification of dead links. The

majority of existing works concentrate on the web

information sources' decay and examine millions of

web pages for deriving statistics about the fraction

of dead web links (Fetterly, et al., 2003; Morishima

et al., 2008). Another signal of web decay, apart

from dead links, is soft-404 server errors, which

imply that a URL redirects to a page that returns an

OK HTTP code, but contains a totally different con-

tent from the one requested (Bar-Yossef et al., 2004;

Lee et al., 2009). Moreover, researchers attested that

decayed pages are characterized by outdated or else

abandoned content. A number of studies have been

reported that try to capture abandoned pages based

on the estimation of their links and content age (Ja-

towt et al., 2007; Popitsch and Haslhofer, 2010). Our

http://en.wikipedia.org/wiki/Wikipedia:Linking.

QUALITY ASSESSMENT OF WIKIPEDIA EXTERNAL LINKS

249

study builds upon existing works on links' decay but

instead of investigating the entire web's link rot, we

concentrate on the links pointed-by the Wikipedia

articles. This is because we seek to understand the

value of Wikipedia external links for their corre-

sponding articles and assess the effectiveness of

Wikipedia maintenance policies in delivering valid,

reliable and longevous content.

3 WIKIPEDIA LINKS’ QUALITY

One thing about Wikipedia that has not been thor-

oughly examined in the course of its quality assess-

ment is the impact of Wikipedia external links in

determining their corresponding articles' value. The

lack of such examination may be because Wikipedia

has set specific linking guidelines, which, if re-

spected, might assure the additive value of external

links to the article contents. In brief, Wikipedia link-

ing instructions request that external links are placed

in identifiable sections under a primary heading (i.e.

external links, citations, etc.) at the end of the arti-

cle’s body to identify sources outside Wikipedia that

are directly relevant to the article that points them.

Moreover, Wikipedia requests that external links to

an article are helpful to the reader, are (and likely to

remain) functional, provide consistent information,

express a neutral stance for their contents and are

kept minimal. The quest in establishing linking stan-

dards is to ensure that Wikipedia refers the reader to

credible and accurate supplemental material, which

would distinguish Wikipedia as an encyclopaedia

from any other online resource.

However, the phenomenal growth of Wikipedia

and the absence of a formal peer review process for

its contents (Denning et al., 2005) make it extremely

difficult for both editors and administrators to vali-

date external links on a regular and coordinated ba-

sis. This, coupled with the insufficiency of existing

link templates and policies to ensure proper external

linking in all situations, impose the imperative need

to automatically validate the contribution of external

links to their respective articles' comprehensiveness

and merit. In the following paragraphs, we examine

a number of features in Wikipedia external links in

order to capture how these adhere to the foreseen

linking guidelines. In Section 3.1 we investigate the

statistical distribution of external links in Wikipedia

collection; we examine how link distribution corre-

lates to the articles' length and we identify popular

domains in Wikipedia' external resources. In Section

3.2, we address the problem of quantifying decay in

Wikipedia external links, in an attempt to assist ad-

ministrators determine effective maintenance and

update Wikipedia policies.

3.1 External Links’ Distribution

Identifying Wikipedia external links is fairly easy as

these are placed in distinct sections at the end of the

articles’ body, released in the Wikipedia dump.

Therefore, we explore the Wikipedia dump to detect

and quantify the links of every article to non-wiki

sources. Nevertheless, assessing the value of exter-

nal links to their article contents is less straightfor-

ward as we would need to capture their appropriate-

ness and their correlation to their linked-to articles.

To address that, we firstly need to identify the

non-wiki sources to which every article points. Such

identification is somewhat complex because there

are different ways of linking to external resources.

To tackle difficulties, we firstly merge all identified

external links into a common file, we remove the

space and/or bullet characters that appear at the be-

ginning of every non-wiki link and then we organize

them in three main clusters according to their syntax.

The emerging clusters group together links provided

in the following forms: (i) URLs placed in square

brackets followed by a hyperlink with a serial num-

ber as its label, (ii) hyperlinked URL names and (iii)

URLs placed in square brackets followed by a space

and some text used to label the hyperlink.

Having extracted and clustered external links, we

process the contents of every cluster in order to get

the URL of each non-wiki linked source. We then

parse the URLs of the external resources in order to

identify their domain names and estimate their dis-

tribution in the Wikipedia collection. For computing

the distribution of URL domain names in Wikipedia,

we simply count the number of articles containing

an external link to some URL listed under each of

the identified domains. The last thing we examine

with respect to Wikipedia external resources is the

correlation between the articles' length and the

amount of their external links. For estimating each

article's length, we parse the XML file of the

Wikipedia dump in order to extract for every file

entry (i.e. article) its pure text. Then, based on the

pure text of every article, we quantify the article's

length by counting the number of characters it con-

tains. Thereafter, we represent every article as a vec-

tor of paired numerical values identifying the arti-

cle's length and the amount of its external resources.

The computed vectors, when plotted, help us visual-

ize the correlation between the article's textual size

and its non-wiki linked resources.

In section 4.1, we report the statistics of our es-

WEBIST 2011 - 7th International Conference on Web Information Systems and Technologies

250

timations, which serve as preliminary evidence

about the degree to which Wikipedia external links

comply with their editing guidelines and help us

infer the quality of Wikipedia external links.

3.2 External Links’ Decay

To estimate decay in Wikipedia external links, we

measure the fraction of dead pages among the non-

wiki linked sources. A dead page is a page that is not

publicly available over the web. Determining if a

page is dead is easy to detect as it will fail either the

URL parsing or the resolution of its host address and

the attempt to fetch it will return an error HTTP

code response

. On the other hand, a page is alive if

it can be successfully fetched and returns an HTTP

code in the 2xx series or 4xx series, except for 403,

404 and 410. But, there exist cases where discrimi-

nating existent (alive) from non-existent (dead)

pages is tedious. Such cases concern redirect pages

that cannot be identified as such since their HTTP

return code is not in the 3xx series. Instead their

landing pages return either an error 404 HTTP code

or an OK code. According to (Bar-Yossef et al.,

2004) in the former case we can safely consider that

the page is dead, but in the latter case there are two

possibilities. The page is dead if one of the follow-

ing applies: (i) the page redirects to its host's home

page and a non-existing randomly generated URL in

the same host directory also redirects to the host's

home page, (ii) the page redirects to a URL, which

in turn redirects and its fetching enters in a loop of

redirects, or (iii) the page and a non-existing ran-

domly generated URL redirect to pages of nearly-

identical content, determined via shingling (Broder

et al., 1997). On the other hand, the page is alive if:

(i) it is a root of a website, or (ii) it redirects to a

URL and a non-existing randomly generated URL in

the same host directory returns a failure HTTP code.

The above situations and their interpretation have

been encapsulated into an elegant algorithm for

identifying dead pages, introduced in (Bar-Yossef et

al., 2004). In our study, we implemented the above

algorithm and run it against the extracted Wikipedia

external links. The results delivered by the algo-

rithm, presented and discussed in Section 4.2, might

assist Wikipedia editors take remedial actions with

respect to out-link updates.

For a list of error codes see: http://en.wikipedia.org/wiki/

Wikipedia:Dead_ external_links

4 EXPERIMENTS AND RESULTS

To capture the contribution of Wikipedia external

links to their respective articles' contents, we relied

on the October 2009 dump of the English-language

Wikipedia, a 24.5 GB XML document, which we

processed in order to extract the pure text of every

article and the URLs of its external resources. Table

1 summarizes the statistics of our dataset. As the

Table shows, out of the 3,290,179 articles in the

English Wikipedia 786,857 (i.e. 23.91%) provide no

links to external resources, whereas the remaining

2,503,322 articles contain 13,355,687 external links

in total; with the ratio of article/ outbound link mean

distribution reaching to 4.06 (median 2). Based on

the articles' identified external links, we downloaded

their pages via a simple crawler module we built and

we processed them as previously described. Then,

we applied our measurements to the article external

links in order to quantify their distribution and decay

in the Wikipedia collection.

Table 1: Statistics of Wikipedia collection.

Collection ID October 2009 Wikipedia dump

Number of articles 3,290,179

Number of external links 13,355,687

External links/ article 4.06

Articles without external links 786,857

4.1 Distribution of External Links

Figure1 plots the distribution of external links to

Wikipedia articles.

300000

600000

900000

1200000

1500000

1800000

2100000

2400000

0 10 30 100 500 1000 1500 1800

Number of external links

Number of articles

Figure 1: External links’ distribution in Wikipedia articles.

As the figure shows this is a normal distribution,

indicating that there are some Wikipedia articles

with very many external links and some articles with

very few links but the amount of external links that

the majority of articles contain is close to the mean.

In particular, results indicate that 7.30% of the

articles point to more than 100 non-wiki sources,

23.91% contain no out-links at all and 68.79% of the

articles contain between 1 and 10 external links.

QUALITY ASSESSMENT OF WIKIPEDIA EXTERNAL LINKS

251

Based on the Wikipedia linking guidelines, which

advise editors to give links to non-wiki sources

when these contain useful material for the article

subjects that is not yet added to the articles' body, we

may interpret our finding as follows.

A small fraction of Wikipedia articles (i.e. those

with many external links) provide incomplete infor-

mation about their subjects, thus they point the

reader to many external resources for additional in-

formation. Moreover, a small fraction of articles (i.e.

those without or with very few external links) pro-

vide complete information about their subjects,

therefore they either need no additional information

or the non-wiki information that complements them

is minimal. Finally, most of the articles contain ade-

quate information about their subjects, which can be

enriched by the contents of some external resources.

But to check the validity of this interpretation, we

need to investigate the correlation between the arti-

cles' length and the amount of their external links.

This in order to attest whether a large amount of

external links signifies incomplete article contents or

not. Figure 2 plots the correlation between the arti-

cles' length and the average number of their external

linked-to sources.

100

120

140

160

180

200

1000 10000 20000 50000 100000 200000 800000

Article length (number of characters)

Avg. number of external links

Figure 2: Correlation between article length and amount of

external links.

According to the figure, there is an upward trend

of external links over long articles, which peaks at

200 links on average for length articles, i.e. those

containing nearly 800K characters and generally

takes values between 3 and 8 links for articles con-

taining on average 10K characters (note that 70.6%

of the articles are 10K characters long). The peak in

the quantity of external links is implicitly imposed

by Wikipedia, which upon the detection of long link

lists invokes a warning to editors and suggests alter-

native ways of linking. But still, the fact that long

articles contain more links than short ones connotes

a contradiction between the instructions given in

Wikipedia linking guidelines and the way in which

these are adopted in practice. In particular, Wikipe-

dia editors are instructed to point to external re-

sources if their content is proper in the article's con-

text and is not yet part of the article. Thus, one

would expect lengthy articles to contain fewer links

than short ones in the sense that the more text the

article contains the decreased the need to link it with

supplemental non-wiki material. But, this is not ob-

served in our results. Next, we report the distribution

of link domain names across the Wikipedia collec-

tion. Such distribution will help us infer the appro-

priateness of Wikipedia external links as additional

information sources, since according to linking

guidelines editors should avoid separate links to

multiple pages in the same website and link to the

site directly.

Figure 3 depicts how domain names span to

Wikipedia articles. Specifically, the x-axis represents

the 50 most popular domain names within the

Wikipedia external resources, where popularity is

determined based on the amount of Wikipedia

pointed URLs listed under each domain. The y-axis

illustrates the amount of Wikipedia pointed URLs

listed each of the examined domain names.

250000

500000

750000

1000000

1250000

1500000

1750000

2000000

d10

d13

d16

d19

d22

d25

d28

d31

d34

d37

d40

d43

d46

d49

URL domain names (sorted by popularity)

Number of external links

Figure 3: Correlation between article length and external

links.

As the figure shows, this is a power-law distribu-

tion, indicating that only a few domains host a large

fraction of the Wikipedia external links, while the

majority of domains host only a small amount of the

Wikipedia external links. In particular, we observe

that the most popular domain alone hosts 14.29% of

the Wikipedia external links, while 44.87% of all

out-links are listed in only 10 different domain

names. Considering that in-linked Wikipedia articles

exhibit strong topical correlation (Koolen and

Kamps, 2009) our finding make sense since URLs in

a domain pointed by many articles of related con-

tents give rise to that domain's popularity inside

Wikipedia.

4.2 Decay of External Links

To compute the fraction of dead links in Wikipedia

external resources, we run the dead link detection

WEBIST 2011 - 7th International Conference on Web Information Systems and Technologies

252

algorithm against the identified non-wiki links of

nearly 2,000,000 articles and estimated the amount

of links that point to dead pages. For running the

algorithm, we specified the following parameters: (i)

we set a page fetching timeout to 10 seconds; so that

if there is no server response within that time frame

the page is declared dead, and (ii) we set the maxi-

mum number of redirects to 5; so that if more than 5

redirects are encountered the page is declared dead.

Moreover, for generating random URLs in the host

directory of the page the algorithm is trying to fetch,

we adopted the method proposed in (Bar-Yossef et

al., 2004), which appends to the host's URL a se-

quence of 25 randomly combined lower case Latin

letters. Based on the above parameters, we run the

algorithm against 4,575,154 linked external URLs

that span to 1,939,445 Wikipedia articles and com-

puted the fraction of external links that point to ei-

ther dead (non-existing) or decayed (redirect to ir-

relevant content) pages. In total, we found 839,245

(roughly 18.34%) of the examined links be dead, i.e.

returned hard or soft-404 codes. Figure 4 plots the

distribution of decayed links in the examined

Wikipedia articles. The x-axis reports the amount of

articles that contain as many external links as their

corresponding y-axis values.

100

150

200

250

300

350

0 0.02K 0.06K 0.13K 0.24K 0.89K 2.9K 0.43M 1.5M

Number of articles

Number of dead links

Figure 4: Distribution of dead links in Wikipedia articles.

According to the figure, the distribution of dead

links follows a power-law with very few articles

containing a considerable amount of dead links and

the striking majority of the articles (i.e., 77.31%)

containing no dead links at all. Results suggest that

the majority of the web pages pointed by Wikipedia

are reachable, thus implying that Wikipedia adminis-

trators have quite effective mechanisms for filtering

rotten links and that Wikipedia editors carefully pick

resources to link their articles.

5 CONCLUDING REMARKS

In this paper, we have examined the quality of Wi-

kipedia external links in an attempt to infer how

these contribute to the improvement of their linked

article contents. For our examination, we quantified

the distribution of external links in Wikipedia collec-

tion and we estimated the amount of external links’

decay. Our estimations suggest that if Wikipedia is

equipped with improved link control and filtering

mechanisms, it will combat link rot and improve the

overall quality of its external links and implicitly of

its article contents.

REFERENCES

Adler, N. T., de Alfaro, L. 2007. A content-driven reputa-

tion system for the Wikipedia. In Proceedings of the

International World Wide Web Conference, pp.

261-270.

Bar-Yossef, Z., Broder, A. Z., Kumar, R., Tomkins, A.

2004. Sic transit gloria telae: towards an understand-

ing of the web's decay. In Proceedings of the 13

In-

ternational World Wide Web Conference, pp. 328-337.

Blumenstock, J. E. 2008(a). Automatically Assessing the

Quality of Wikipedia Articles. UCBiSchool Report

021.

Blumenstock, J. E. 2008(b). Size matters: Word count as a

measure of quality on Wikipedia. In Proceedings of

the 17

International World Wide Web Conference,

pp. 1095-1096.

Broder, A. Z., Glassman, S. C., Manasse, M. S., Zweig, G.

1997. Syntactic clustering of the web. In Proceedings

of the 6

International World Wide Web Conference,

pp. 391-404.

Buriol, J., Castillo, C., Donato, D., Leonardi, S., Millozzi,

S. 2006. Temporal evolution of the wikigraph. In Pro-

ceedings of the IEEE Web Intelligence Conference, pp.

45-51.

Cross, T. 2006. Puppy smoothies: improving the reliability

of open, collaborative wikis. First Monday 11(9).

Denning, P., Horning, J., Parnas, D., Weinstein, L. 2005.

Wikipedia risks. Communications of the ACM, vol.48,

no.12.

Emigh, W., Herring, S. 2005. Collaborative authoring on

the Web. In Proceedings of the 38

Hawaii Intl. Con-

ference on System Sciences.

Fetterly, D., Manasee, M., Najork, M., Wiener, J.L. 2003.

A large-scale study of the evolution of web pages. In

Proceedings of the 12

International World Wide Web

Conference, pp. 669-678.

Giles, J. 2005. Internet encyclopedias go head to head.

Nature, 438, pp. 900-901.

Jatowt, A., Kawai, Y., Tanaka, K. 2007. Detecting age of

page content. In Proceedings of the 9

Annual ACM

International Workshop on Web Information and Data

Management, pp. 137-144.

Kamps, J., Koolen M. 2009. Is Wikipedia link structure

different? In Proceedings of the 2

International Con-

ference on Web Search and Data Mining, pp. 232-241.

QUALITY ASSESSMENT OF WIKIPEDIA EXTERNAL LINKS

253

Kirtsis N., Stamou S., Tzekou P., Zotos N. 2010. Informa-

tion Uniqueness in Wikipedia. In Proceedings of the

International Conference on Web Information Sys-

tems and Technologies (WebIST), Valencia, Spain.

Koolen, M., Kamps, J. 2009. What's in a link? Form

document importance to topical relevance. In Proceed-

ings of the 2

International Conference on the Theory

of Information Retrieval, pp. 313-321.

Lee, T., Kim, J., Kim, J. W., Kim, R. S., Park, K. 2009.

Detecting soft errors by redirection classification. In

Proceedings of the 18

International Web Conference,

pp. 1119-1120.

Mrishima, A., Nakamizo, A., Iida, T., Sugimoto, S., Kita-

gawa, H. 2008. PageCasher: A tool for the automatic

correction of broken web links. In Proceedings of the

International IEEE Conference on Data Engi-

neering, pp. 1486-1488.

Nielsen, F. A. 2007. Scientific citations in Wikipedia.

Computing Research Repository.

Popitsch, N. P., Haslhofer, B. 2010. DSNotify: Handing

Broken Links in the Web of Data. In Proceedings of

the World Wide Web Conference.

Riehle, D. 2005. How and why Wikipedia works: an inter-

view. In Proceedings of the Intl. Symposium on Wikis,

pp. 3-8.

Stvilia, B., Twidale, M. B., Smith, L. C., Gasser, L.

2005(a). Assessing information quality of a commu-

nity-based encyclopedia. In Proceedings of the Inter-

national Conference on Information Quality, pp. 442-

454.

Stvilia, B., Twidale, M. B., Gasser, L., Smith, L. C.

2005.(b) Information quality discussions in Wikipedia.

In Proceedings of the International Conference on

Knowledge Management.

Wilkinson, D. M., Huberman, B. A. 2007. Assessing the

value of cooperation in Wikipedia. First Monday

12(4).

WEBIST 2011 - 7th International Conference on Web Information Systems and Technologies

254