Having asked questions last week about how to read these articles properly, I have discovered that the full text behind the abstract is a lot richer in information. As a result this week I read about how the average search on search engines such as Google is 2.3 words long. Furthermore the articles explained how web crawlers gather the information from trillions of web pages each second and how each crawler only does a specific word it is assigned to. Moreover a simple crawler could take days to complete a search, however over time they have improved. For example they only look for relevant information, they do not convey sites with lots of spam and tend to leave out the web pages with the least relevance to the search. I also read how to narrow down searches with quotation marks. Also interestingly when comparing search engines due to the different way in which web crawlers are programmed, they bring up around 85% of unique results and only 3% of the same result was actually on all three websites, they had searched on. It could be because different systems such as Lycos, store the information differently into a database, due to the crawlers programming being slightly different.
Citing for the posts this week:
Hawking, D. 2006. Web search engines. Part 1. Computer, 39 (6), pp. 86-88. Available from: doi: 10.1109/MC.2006.213 [Accessed: 15 Oct 2013].
Hawking, D. 2006. Web Search Engines: Part 2. Computer, 39 (8), pp. 88-90. Available from: doi: 10.1109/MC.2006.286 [Accessed: 15 Oct 2013].
Mauldin, M. 1997. Lycos: design choices in an Internet search service. IEEE Expert, 12 (1), pp. 8-11. Available from: doi: 10.1109/64.577466 [Accessed: 15 Oct 2013].
Spink, A., Jansen, B., Blakely, C. and Koshman, S. 2006. Overlap Among Major Web Search Engines.Information Technology: New Generations, 2006. ITNG 2006. Third International Conference on, pp. 370-374. Available from: doi: 10.1109/ITNG.2006.105 [Accessed: 15 Oct 2013].