There is either something seriously flawed with the G-corr tool or something flawed with the underlying data. . For instance if "gateway bible" has the highest correlation with "ibuprofen" then we could expect that the reverse would be true.
I suspect the tool is OK but the data is off (or requires more filtering). IMHO a classic instance of GIGO. On Tue, Oct 4, 2011 at 10:03 PM, Steve Smith <[email protected]> wrote: > Shock but not surprise: > > Google Correlates with "Gppg;e" or off-home left hand? > > https://www.google.com/trends/**correlate/search?e=google&e=** > gppg+e&t=weekly#<https://www.google.com/trends/correlate/search?e=google&e=gppg+e&t=weekly#> > > Most of my Correlate results show a strong annual periodicity... I'm > curious if others see this and if so what their speculations might be about > it.. > https://www.google.com/trends/**correlate/search?e=egypt&e=** > about+egypt&t=weekly#<https://www.google.com/trends/correlate/search?e=egypt&e=about+egypt&t=weekly#> > What might make egypt less interesting to global searchers during the > summer and winter solstices? > > This observation lead me to try to use Trends opposite Corrolate: > > http://www.google.com/trends?**q=egypt%2C+solstice&ctab=0&** > geo=all&date=all<http://www.google.com/trends?q=egypt%2C+solstice&ctab=0&geo=all&date=all> > > And to work with a term (solstice) which is clearly going to have a > (bi)annual periodicity. > > I'm waiting for Google Speculate, I wonder what that will look like? > > > ==============================**============================== > FRIAM Applied Complexity Group listserv > Meets Fridays 9a-11:30 at cafe at St. John's College > lectures, archives, unsubscribe, maps at http://www.friam.org >
============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
