Newsgroups: alt.ufo.reports
From: MrPostingRobot@kymhorsell.com
Subject: watching the watched watch the watchers (2/3)

[uploaded 25 times; last 17/10/2024]

We've prev looked at my diary entries for 2022 correlated against
weather data for each 10x10 deg region across the earth.  The data
series used were all month-based. Many science organisations release
monthly gridded data for land/air/ocean temperatures, earthquakes,
cosmic rates, etc.

When correlating the sightings from the diary month-by-month against
the 1000s of published datasets we found the "usual suspect"
areas of the world lit up like xmas -- key parts of Antarctica,
Africa, S America, and the Atlantic had weather data that showed
non-chance resemblances to the ups and downs in the sightings dataset.
We have to assume those regions, for whatever reason, somehow
"predict" the appearance of unusual objects I see over my property
from time to time.

And now we'll push a bit harder and use another dataset I gather and
process for various purposes -- daily sea surface temperatures.  This
time the gridding is fine on both the time and spatial dimensions.  
It comes from organisations that process satellite data where the
sats in question have IR cameras pointed down at the earth.  Some
groups use the IR info to measure the temperature of the ocean surface
or near-surface at local nighttime, when clouds and the reflection of
the sun don't interfere or blind the camera, respectively.  You can
generally get a daily read on ocean temp for each 1x1 deg area of the
earth between (say) 70S and 70N.

Just for fun I made the problem a bit harder than usual for the AI
programs.  Instead of giving them the data for each 1x1 deg area of
the oceans I used a dataset created some years back to predict ocean
storms and US tornadoes. It's region-based. The model I developed back
then was intended to predict the development of e.g. hurricanes and
typhoons based on changes in SST in key regions. The idea of that
exercise was to find out which regions were the best predictors and
exactly how to process the IR data to get the best predictions into
the future.

So taking data for the 58 regions the programs discovered back then
and feeding it to new programs now, I want them to create a density
plot showing which areas of sea surface seem to warm and cool in exact
sync with the way the number of objects I've seen cross over the sky
or dance around with small aircraft -- as the case may be -- over the
past few years. And does that look like the density map we already
created with the monthly datasets I gave it before?

The programs this time had some small problems trying to translate the
name of a region to specific parts of the ocean and then plotting the
relevant correlation coefficient (I normally use the so-called
"explanation power" of time-series regressions as an indication of how
well the weather data at some location matches up against UFO data
from somewhere else; as usual the TS regr must pass 2 stat tests before
it is accepted; of those stat tests that pass for each region of the
map the best one -- the largest R2 -- is selected as the correlation
value to plot). But it managed a fairly good answer even tho it does
have an interesting couple of errors.

The plot is available here: <kym.massbus.org/UFO-DIARY/MAP2/map.gif>.

It is in reasonable agreement with the prev plot based on totally
different data. As before, the Antarctic and Southern Oceans feature
prominently. Some key regions of some continents also show up.  And
some key regions of a couple seas we did not see in the prev map --
but we have seen before on similar maps using NUFORC data that mostly
relates to objects seen over the US and Canada -- also light up.

There are a couple of "errors" that it will be hard to explain to the
AI's so they get it right next time. :) The first one -- it's assigned
a default value for the world as a whole.  Even tho the data used
were regions of the ocean, the AI didn't twig that land should
probably be marked "unknown" (or black in color on these maps)
rather than giving it some low correlation value based on the average
of the correlation values it found for the 58 regions given.

And a more amusing error we see at the edges of the map. For "some
reason" the mid-Pacific is shown at sig higher correlation than the
average for that ocean. And we also see high corr across the far N and
far S latitudes.  All around the edges of the map we see what looks
like artificially high correlation values.

It turns out *this* is related to how the AI interprets what "regions"
area.  It has a generalized idea what the "Arctic" and "Antarctic" are
based on language use. I.e. very similar to the ideas used in some
high-profile products from Microsoft and Google. But when it comes
to specific locations like "lat 90, lon -180" the AI is not too sure
whether that is INSIDE the Arctic region or outside. And it chooses
outside. Where the correlations we're interested in here are larger
because they correspond with a "world average".

This is part of the fun of using AI programs. Trying to figure out
why they said something obviously strange.

--
[Secret UFO recovery program blown open:]
I hope this revelation serves as an ontological shock sociologically
and provides a generally uniting issue for nations of the world to
re-assess their priorities.
-- David Grusch, 05 Jun 2023
[Talking to Les Kean et al for The Debrief, Grusch called for an end to
nearly a century of global UFO secrecy and warned that humanity needed to
prepare itself for "an unexpected, non-human intelligence contact scenario"].

[David Grusch's] assertion concerning the existence of a terrestrial arms
race occurring sub-rosa over the past eighty years focused on reverse
engineering technologies of unknown origin is fundamentally correct, as
is the indisputable realization that at least some of these technologies
of unknown origin derive from non-human intelligence.
-- Col Karl Nell (ret), 06 Jun 2023

Aliens could be hiding in 'terminator zones' as UK sees surge in new UFO 
sightings
Daily Express, 05 Aug 2023 10:44Z
Aliens could be hiding and thriving in "terminator zones", according to 
astronomers. A study published in April claimed extraterrestrial life could 
exist in so-called terminator zones, the border ...
[There are some indications in the data there could be "bases" way out
in the outer solar system. But the space telescope images show
activity in all directions that correspond with UFO sightings over N
Am.  From my very prelim survey I've found ~6 small areas in the ~12
tiles examined that show "something" moving slowly and blocking out
successive background stars. Some of the things move "horizontally"
left to right or right to left. At least one moves on a diagonal BL to
TR. While I am still to figure out whether these motions are in the
direction of the planets and asteroids, or go "backwards", and whether
they are in the plane of the ecliptic like other bodies or are far
above or below the plane, just the density of them is incredible.
Each tile is about 2.5 deg x 2.5 deg. The stack of images generally is
100 hours of viewing, one frame per hour. (They are 30MB frames so I
haven't downloaded very many of them yet). Given the number of "hits"
it seems we should expect about 1 image in 3000 looking in any
direction to show "something moving". Or any any 1 time around 300
moving objects visible. While some of these things are likely
asteroids (the space telescopes are programmed not to look at known
objects but it's possible some smaller uncharted ones sometimes creep
into frame) we have to remember whatever is going on highly
correlates with people seeing disks, spheres, and lights in the sky
some hours after these images are snapped. It is likely some of them
are "highly unusual". And there appear to be a LOT of them moving
about all the time. This is a very long way to say -- they are not all
out there in the distant solar system. One H*ll of a lot of unusual
things seem to be living right on our doorstep and are going about
their business all the time. And they are not trying to hide at all;
it's just we can't understand what we are looking at yet].