Daily time series of common single word searches in AltaVista and NorthernLight

Rousseau, Ronald     ronald.rousseau@kh.khbo.be     (KHBO, Department of Industrial Sciences and Technology, Oostende (Belgium))

Cybermetrics 1998/99 (1)
(ReLIS:upv:upvcyb:y:1998-99:v:2-3:i:1:p:2)

Abstract:

Over the past year the AltaVista search engine provided irregular numbers of hits. This is shown by collecting a time series of daily searches. The new AltaVista (on the Web since October 25, 1999) is much more stable, but not all problems have been eliminated. The same searches performed on NorthernLight gave stable results. It is suggested to use filtering techniques, such as median filtering, when dealing with irregular time series. An attempt is made to estimate the growth of the Web (as covered by a search engine), based on neutral single word searches. It turned out that this method systematically underestimates the growth of the Web. Collecting time series should be an essential part of Internet research.

Creation: 1999
Keywords: time series, median filtering, AltaVista, NorthernLight, estimating growth, nova-effect
Year: 1998/99
Issue: 1

 (go top)