my Specificity and Sensitivity against the average values.

Discuss your experiences with and ideas about Stardust@home here.

Moderators: Stardust@home Team, DustMods

Post Reply
zioriga
Posts: 53
Joined: Sat Jun 10, 2006 1:00 pm
Location: Moncalieri, Italy

my Specificity and Sensitivity against the average values.

Post by zioriga »

A suggestion to improve the functionality of this project.

Is it possible to know the average values of Sensitivity and Specificity, just to check how people is doing well (or bad) ??

May be those values published in everyone's search page or in "My event page" or whatever else!.
I think it should be fine whatever type of statistical data about Spec. and Sens.

I think this may be a spurring point for everyone's improvement!!!
fjgiie
DustMod
Posts: 1253
Joined: Sat May 20, 2006 8:47 am
Location: Hampton, SC, US

Specificity and Sensitivity against the average values

Post by fjgiie »

Hi zioriga,

The average scores of dusters who posted in the Post your stats thread are:

Specificity 98.31%

Sensitivity 92.38%

These averages are from only 13 scores that are posted there.
Nikita
DustMod
Posts: 994
Joined: Wed May 17, 2006 8:33 pm
Location: Indiana, USA

Post by Nikita »

Bryan has made some comments about this. In regards to specificity, he says something here. In regards to the scores all together, he says more here with links from that post to other posts he has made.
I hope this helps!
From dust we come
zioriga
Posts: 53
Joined: Sat Jun 10, 2006 1:00 pm
Location: Moncalieri, Italy

Post by zioriga »

OK thanks for the explanations
meatpie
Posts: 9
Joined: Wed Dec 05, 2007 12:56 pm
Location: London ON

Post by meatpie »

I've made some quick calculations (literally using my screen, a ruler and a calculator. This may or may not interest anyone but myself, but:

My monitor shows a movie viewing area of 15.7 cm x 11.6 cm, and the 50 μm scale bar is 3.8 cm. Roughly, this means I'm looking at 31,534 μm^2 of area in each frame.

The total area being searched is 1.0 m2, or 1.0 x 10^12μm^2. This means we have something like 31,711,803 'movies' to search -- ideally, many times.

To put that into perspective, I've looked at 9,600 movies so far and that's only about 0.03% of the available material.

Suddenly, I feel very very small.
"Look in the mirror, and don't be tempted to equate transient domination with either intrinsic superiority or prospects for an extended survival." (SJG)
fjgiie
DustMod
Posts: 1253
Joined: Sat May 20, 2006 8:47 am
Location: Hampton, SC, US

You are ten times bigger than you thought

Post by fjgiie »

meatpie wrote:The total area being searched is 1.0 m2...
More like 0.10 m2 in area, not 1.0 m2 .
The tiles are approximately 2cm X 4cm which is 8 cm2 each.
There are approximately 130 tiles 130 X 8 cm2 equals 1040 cm2 .
A meter2 has 100 cm to a side. 100cm X 100cm equals 10,000 cm2 .
1000cm2 divided by 10,000cm2 is 0.1, a tenth of a meter2

On my screen the movies are 240 microns X 193 microns for 46,000 microns2

Anyhow, around 3 million movies over the whole collector times 400 views is
a billion looks, give or take a few million. Of course what I'm writing could be all wrong in many ways. :)

fjgiie (and you thought nobody would answer you)
meatpie
Posts: 9
Joined: Wed Dec 05, 2007 12:56 pm
Location: London ON

Post by meatpie »

:? What's an order of magnitude between dusters? :?

I still feel pretty small.

And I'm glad you bothered to respond, btw!
"Look in the mirror, and don't be tempted to equate transient domination with either intrinsic superiority or prospects for an extended survival." (SJG)
bmendez
Stardust@home Team
Stardust@home Team
Posts: 530
Joined: Wed Apr 19, 2006 11:28 am
Location: UC Berkeley Space Sciences Lab
Contact:

Post by bmendez »

Hi All,

I'm glad you two decided to work the numbers. A few million movies when all is said and done is about right. There are a few factors which were not mentioned that change your calculations in detail: things like the switch from lower to higer resolution, an overlap between fields of view by about 10% (on the lower resolution movies), and also that the VM displays movies at random which means that you will sometimes see the same real movie more than once.

But we astronomers tend to like to think in factors of 10. So a million movies with a billion searches is the right order of magnitude. When you consider those numbers you are reminded just why we used this distributed thinking approach.

Thanks all,
-Bryan
"I am made from the dust of the stars, and the oceans flow in my veins"
- RUSH
Post Reply