Hi everyone,
what is the base model of LCM ?
I think LCM is developed based on cellular automata. is this true?
Date
Votes
2 comments
-
Clark Labs Official comment Hi Ali,
LCM offers three techniques for modeling transition potential: Similarity-Weighted Instance-Based Machine Learning (SimWeight), Logistic Regression, and Multi-Layer Perceptron (MLP) neural network.
SimWeight is based on a similarity-weighted K-nearest neighbor procedure. Logistic Regression undertakes binomial logistic regression and prediction using the Maximum Likelihood method. MLP models transition potential through a dynamic learning procedure that uses equal-size samples of pixels that changed and pixels that persisted between the two dates as training data.
You can read more about these three different modeling techniques in the TerrSet Help System, which can be downloaded from the following link: http://clarklabs.org/support/downloads.cfm.
The Clark Labs Team
-
ali kazemzadeh thank you very much,
I'm familiar with these techniques. But i thought these used for extraction of local transition rules on the CA model as well as Markov chain for global transition rules.
according your answer's CA doesn't implemented in the LCM, true?
Please sign in to leave a comment.