April 16, 2012 at 1:05 pm
Okay. Have fun. Let me know how it goes.
April 16, 2012 at 1:09 pm
we will;-)
"simplicity is the ultimate sophistication."
April 16, 2012 at 1:20 pm
Lynn Pettis (4/16/2012)
Okay. Have fun. Let me know how it goes.
Lynn, do you implement project with data mining method?
---------------------------------------------------
baaaaaaaaaaaaaleh!!! (Ehs_Akb)
April 16, 2012 at 2:01 pm
Lynn Pettis (4/16/2012)
Okay. Have fun. Let me know how it goes.
You can google it , it is possible, search this:Data mining in robo cup simulation
and plz dont let us alone we need your help.
"simplicity is the ultimate sophistication."
April 16, 2012 at 2:25 pm
we can design very simple agent for this problem by using simple AI algorithm
for example you can read more about this simple methods in:
Programming Game AI by Example. Mat Buckland 2005
by using this simple methods and using philosophy of fuzzy logic and map this problem
as a classification and function estimation process we can have more intelligent agents.
also by using Data mining and generating large enough dataset we can find very usefull
information about problem by analyzing through this dataset.
April 16, 2012 at 3:05 pm
you can read this articles to know more about how using DM techniques for this problem:
Amin Milani Fard, Vahid Salmani, Mahmoud Naghibzadeh, Sedigheh Khajouie Nejad, and
Hamed Ahmadi. Game theory-based data mining technique for strategy making of a soccer
simulation coach agent. In Heinrich C. Mayr and Dimitris Karagiannis, editors, Information
Systems Technology and its Applications, 6th International Conference ISTA’2007, May 23-
25, 2007, Kharkiv, Ukraine, volume 107 of LNI, pages 54–65. GI, 2007.
also you can read more about how using neurofuzzy methods to solve this problem in this article:
Reza Zafarani and Mohammad Reza Yazdchi. A novel action selection architecture in soccer
simulation environment using neuro-fuzzy and bidirectional neural networks. International
Journal of Advanced Robotic Systems, 4(1):93–101, March 2007.
this articles are just samples
April 16, 2012 at 3:29 pm
thank you so much .ruhollah!
---------------------------------------------------
baaaaaaaaaaaaaleh!!! (Ehs_Akb)
April 16, 2012 at 4:03 pm
Larry Page(Ehsan.Akbar) (4/16/2012)
PaulB-TheOneAndOnly (4/16/2012)
Larry Page(Ehsan.Akbar) (4/16/2012)
Henrico Bekker (4/15/2012)
Yeah, dont think DM is the right tool for your game....so my problem is which information should be inserted in my database as an input ?
this is my idea: i want to insert all position of my players(x,y) and other team's position so and the sequence of the ball's position .do you think this is going to be my input ?
I don't know, your game, your ideas 🙂
Last but not least, I understand some people has used DM techniques to simulate some board combinational games like a 3x3 chess sort of game but I do not think you can use DM to simulate a soccer game - at least not with the processing power you can get, you are talking about a matrix of millions-x-millions where stuff happens dynamically.
why do you think this method couldn't be the answer?
so if you give the data mining model a lot of method that can be goal you can find a pattern from these data .
well... that's why I asked you on my first post about the dataset you are planning to target for this; no answer yet 😀
_____________________________________
Pablo (Paul) Berzukov
Author of Understanding Database Administration available at Amazon and other bookstores.
Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.April 16, 2012 at 4:05 pm
Larry Page(Ehsan.Akbar) (4/16/2012)
Lynn Pettis (4/16/2012)
Depends, what is the purpose of this exercise? Is this for a class project?no it is a competition in our university.
Can you choose a different subject?
_____________________________________
Pablo (Paul) Berzukov
Author of Understanding Database Administration available at Amazon and other bookstores.
Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.April 17, 2012 at 5:54 am
Although the dynamics of the real time game cannot be "data mined" readily, we could collect a multitude of pieces of data about each player, where they are on the pitch, are they in an offside position, what is their heart beat, have they already committed any fouls, which fouls, time left in the match, cards issued, goals scored, etc. Monitors already exist for both the ball and the players' boots. I can definitely see using all this information in a data mining exercise.
April 17, 2012 at 6:05 am
Scott Murray-240410 (4/17/2012)
Although the dynamics of the real time game cannot be "data mined" readily, we could collect a multitude of pieces of data about each player, where they are on the pitch, are they in an offside position, what is their heart beat, have they already committed any fouls, which fouls, time left in the match, cards issued, goals scored, etc. Monitors already exist for both the ball and the players' boots. I can definitely see using all this information in a data mining exercise.
Perhaps. Is this data publicly available?
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
April 17, 2012 at 6:46 am
I bet the leagues are not publishing anything!!:w00t:
April 19, 2012 at 12:53 am
Phil Parkin (4/17/2012)
Scott Murray-240410 (4/17/2012)
Although the dynamics of the real time game cannot be "data mined" readily, we could collect a multitude of pieces of data about each player, where they are on the pitch, are they in an offside position, what is their heart beat, have they already committed any fouls, which fouls, time left in the match, cards issued, goals scored, etc. Monitors already exist for both the ball and the players' boots. I can definitely see using all this information in a data mining exercise.Perhaps. Is this data publicly available?
no we should generate that.
---------------------------------------------------
baaaaaaaaaaaaaleh!!! (Ehs_Akb)
April 23, 2012 at 1:45 am
we found resources about this method.and we are sure that it 's worked!!!
---------------------------------------------------
baaaaaaaaaaaaaleh!!! (Ehs_Akb)
April 23, 2012 at 6:05 am
my 2 cents worth - not knowing much about data mining..
Given that we would know the current location and vector of each player and the location of the ball (at a players feet presumably) DM should be able to calculate a vector and speed that enables one of our players to intersect the ball's path that does not put them in an off-side position and cannot be intercepted by an opposing player AND moves the ball closer to the goal.
including factors such as dip and swerve of the ball's path through the air due to strike form and gravity, the friction generated by the playing surface to slow down the ball, players ability to alter speed and vector based on the play dynamics and anticipation of expected play formation all add a logarhythmic complexity to the decision tree but the main principal would remain. - I don't fancy writing the calculations though.
Breakout decisions would need to be made whether it was better to shoot at goal on this iteration or make another pass which would increase the probability of scoring. Other decisions on whether it was advantageous to make or induce a foul rather than continue open play would also need to be made - perhaps also the skill of the player to sucessfully execute the preferred option.
Maybe you want to start with a simpler model - like table football (Subutteo) where only one player moves at a time and players take it in turns to move a player
Viewing 15 posts - 31 through 45 (of 48 total)
You must be logged in to reply to this topic. Login to reply