How trustworthy is that Wikipedia article? This algorithm can help

Published November 1, 2013 6:31pm

With doubts recently being raised on the quality of articles on open-source encyclopedia Wikipedia, two computer scientists have developed a formula that could help assess the credibility of an article.

Xiangju Qin and Pádraig Cunningham of the University College Dublin in Ireland developed the algorithm based on the authoritativeness of the editors involved and the longevity of the edits they made, Mashable reported.

“The hypothesis is that pages with significant contributions from authoritative contributors are likely to be high-quality pages,” Mashable quoted the two as saying.

Qin and Cunningham suggested the new algorithm could help identify new articles of relatively good quality and those of low quality and may need further attention.

They noted "high-quality articles generally involve more communication and interaction between contributors.”

In recent weeks, Wikipedia has come under scrutiny for relatively dubious articles.

Last month, Wikimedia, the foundation behind Wikipedia, shut down at least 250 user accounts that it suspected "may have been paid to write Wikipedia articles to promote organizations or products," according to Wikimedia Foundation executive director Sue Gardner.

Measurements

Mashable said Qin and Cunningham measured the longevity of an edit, checking the size of an edit made by a given author and how long this edit lasts after other revisions.

A "high-quality" edit is one that is more likely to survive future revisions.

Anonymous contributions are ignored in the computations to get around the possibility of vandalism.

The researchers then measured the authority of each editor, with more experienced editors deemed likely to be better connected in the network.

Qin and Cunningham measured authority by checking the number of other editors a given editor is linked to, and by using a Pagerank-type algorithm.

They then combined these metrics of longevity and authority to measure an article's quality.

Testing on 9000 articles

Qin and Cunningham tested the quality of some 9,000 Wikipedia articles using their algorithm.

"They say that the longevity of an edit by itself is already a good indicator of the quality of an article. However, taking into account the authority of the editors generally improves the assessment," Mashable said.

Limitations

But the method has limitations, including addressing reverted edits where an edit is removed to get rid of vandalism.

“At present, we do not have any special treatment to deal with reverted edits that do not introduce new content to a page,” Mashable quoted Qin and Cunningham as saying. — TJD, GMA News

Tags: wikipedia, wikimedia