11 May 2006

Google Scores in Scientific Journals

It has been reported on PhysicsWeb that Google's searching algorithm - PageRank - is capable of unearthing worthy scientific literature, and is in some ways superior than the traditional citation method. Although the experiment is based on physics articles, I see no reason why it shouldn't work on papers on other sciences. I suppose this gives me even more reason to use Google Scholar when searching for articles.

Of course, PageRank has its pitfalls, as highlighted in the article. In that case, would a combination of PageRank result and number of times cited produce the best results?

Google or Google Scholar, however, only gives the abstract and other basic information of the journal articles. Most of the time, if anyone wishes to view the complete article itself, there will usually be a request to sign up as a member or pay for the article for something like that. This is natural, given that these journals have to operate at least with a profit. Fortunately, there are two ways around this. Most libraries subscribe to many online services. For example, most journal articles are available if you log in to the NUS library website. Another method is to visit ArXiv (pronounced "archive", since the "X" is actually the Greek letter "chi"), which is where many authors, before sending their scripts to the publishers, put their draft for peer reviewing. Of course, this means that the article may not be accurate/corrected/edited, but I never had any problems.

No comments: