Microsoft Academic Search

Microsoft have followed Google and their Google Scholar search tool into the academic arena with the Microsoft Academic Search platform. Based on a whole range of experimental technologies being developed by the software giant, the platform promises much for those of us interested in academic research.

The indexing and search facilities are very impressive, not to mention the range of outputs which can be produced with ease. Below are two dynamic graphs generated by the system: The first is a co-author graph for my publications whilst the second is a co-author path from me to the one and only Tim Berners-Lee.

Some really nice touches include BibTex import (great for uploading your existing reference library) and pdf upload to provide access to papers. Links to existing online pdfs are also supported. This is great but the icing on the cake would be to interface with existing online reference managers; I’m really not going to maintain multiple repositories of information and whilst a BibTex export/import process is possible, wouldn’t it be nice if the information was simply harvested and recycled…?

Of course, it is early days yet and there are some fairly obvious issues that will require work. Firstly, the bibliographic structure is very limited and does not currently support the full range of publications, focusing instead on journal articles. Book chapters and more complex references are stripped back to the bare essentials so they will fit into the structure in use. Secondly, there is an issue with author disambiguation. I’m not sure what magic allows Microsoft to work out when two authors with the same name are actually different people, but unless some of my colleagues have been doing some serious moonlighting well outside of the their comfort zones, some author records have been wrongly conflated.

That said, the more ways in to the web of academic publication the better. Up till now, it has been too hard to find relevant papers (using webcat technologies via university libraries for example) and then nigh on impossible in too many cases to access the publication itself (academic establishments not necessarily subscribing to every publication available and for those of us without active academic logins such as Athens authentication, having to purchase individual files at exorbitant costs or spend all our wages on subscriptions!). Better tools to find relevant works become even more useful when combined with the fact that more and more academics are self-publishing their pre-press papers and articles, which can be made available through search tools such as this.

So well done Microsoft! And well done Google! And you others working in this field such as Mendeley and Zotero! Top marks for opening up the ivory towers and innovating ways in which we can all share our knowledge.