data mining

Question

Post reply

data mining

Viewing 4 posts - 46 through 48 (of 48 total)

You must be logged in to reply to this topic. Login to reply

Phil Parkin SSC Guru Points: 246962 More actions · Answer 1

Larry Page(Ehsan.Akbar) (4/23/2012)
we found resources about this method.and we are sure that it 's worked!!!

This reminds me of another tantalising, yet effectively useless, quote:

I have a truly marvellous demonstration of this proposition which this margin is too small to contain.

The absence of evidence is not evidence of absence
- Martin Rees
The absence of consumable DDL, sample data and desired results is, however, evidence of the absence of my response
- Phil Parkin

Ariel Nessi Old Hand Points: 324 More actions · Answer 2

I'm not sure this add's to your demand BUT,

you could get out there a sample data of all the goals made in a season, these reports should contain data about the player, distance of the goal, and other variables of the goal, most important of these, the place from which the goal was scored and the place where the assistance came from.

with that input you could generate hot spots from where assist come to which hot spots goals are most made.

these can be categorized by kick, head, corner.

with that you already got yourself two players in the right zone. the one that assists and the one that scored the goal. from that backwards there's no shortcut for inputs, you would have to watch games, or build a video parser to get the zones from where passes to the assister most comes from and so on.

As for the defense positioning there's again no shortcut, you would have to compute the games where the ones that scored goals on your sample data failed to do so, and build a defense pattern from that on.

As for the keeper you would have to compute all the kicks made in these hot spots and get all the saves vs all the goals and get a goal keeper ratio of sucess

i'm not sure the text was clear sorry for any trouble

Larry Page(Ehsan.Akbar) SSCommitted Points: 1916 More actions · Answer 3

ariel_mlk (4/26/2012)
I'm not sure this add's to your demand BUT,
you could get out there a sample data of all the goals made in a season, these reports should contain data about the player, distance of the goal, and other variables of the goal, most important of these, the place from which the goal was scored and the place where the assistance came from.
with that input you could generate hot spots from where assist come to which hot spots goals are most made.
these can be categorized by kick, head, corner.
with that you already got yourself two players in the right zone. the one that assists and the one that scored the goal. from that backwards there's no shortcut for inputs, you would have to watch games, or build a video parser to get the zones from where passes to the assister most comes from and so on.
As for the defense positioning there's again no shortcut, you would have to compute the games where the ones that scored goals on your sample data failed to do so, and build a defense pattern from that on.
As for the keeper you would have to compute all the kicks made in these hot spots and get all the saves vs all the goals and get a goal keeper ratio of sucess
i'm not sure the text was clear sorry for any trouble

thank you .in fact i generated data .in my simulation i have three operation ,shoot ,run ,and ,pass,

.

so i generated data randomly .for example i generated a sequence of numbers (10 numbers)

1,2,3,2,1,3,,1,2,3,1(1 is shoot ,2 is run ,and 3 is pass)

and i saved the information of player's position and other details.

---------------------------------------------------
baaaaaaaaaaaaaleh!!! (Ehs_Akb)

Ariel Nessi Old Hand Points: 324 More actions · Answer 4

thank you .in fact i generated data .in my simulation i have three operation ,shoot ,run ,and ,pass,
.
so i generated data randomly .for example i generated a sequence of numbers (10 numbers)
1,2,3,2,1,3,,1,2,3,1(1 is shoot ,2 is run ,and 3 is pass)
and i saved the information of player's position and other details.

Perharps i don't quite get you're getting at, but i guess that would assume that a goal is scored in a finite numbers of 10 steps ? whilst that can be computed, i don't really believe it can't be applyed to real world, due to the nature of the game is already hard enough to predict a game outcome with real world data of the games.

I don't see how computer generated data can be of much help here.

You'll get the paths to the goal that way, i'm pretty sure, but it's not clear to me how that could be implemented to flesh and bones soccer afterwards. If that's the purpose.