HTC Posted January 31, 2013 Report Share Posted January 31, 2013 Hello every body! I study topic: Estimation forest base on K nearest neighbour and Regression Kriging method. I search in internet and literature about K nearest neighbour and Regression Kriging. They use R software to process. I know little about it. I want use another software to process. Example: Arcgis. I dont know Arcgis can process it? If someone know how to process K nearest neighbour and Regression Kriging in Arcgis. Please help me (stuff, tutorial or guide....). Regard. Quote Link to comment Share on other sites More sharing options...
yousef2233 Posted January 31, 2013 Report Share Posted January 31, 2013 Hi HTC, I found this one http://resources.esr...lationships.htm for k nearest neighbors, and also this http://webapps.fundp...terpolating.pdf sorry i didnt know that you are in hurry 1 Quote Link to comment Share on other sites More sharing options...
yousef2233 Posted February 7, 2013 Report Share Posted February 7, 2013 The k-NN algorithm can also be adapted for use in estimating continuous variables. One such implementation uses an inverse distance weighted average of the k-nearest multivariate neighbors. This algorithm functions as follows: Compute Euclidean or Mahalanobis distance from target plot to those that were sampled. Order samples taking for account calculated distances. Choose heuristically optimal k nearest neighbor based on RMSE done by cross validation technique. Calculate an inverse distance weighted average with the k-nearest multivariate neighbors. Using a weighted k-NN also significantly improves the results: the class (or value, in regression problems) of each of the k nearest points is multiplied by a weight proportional to the inverse of the distance between that point and the point for which the class is to be predicted. http://en.wikipedia....ghbor_algorithm if you looking for idw tools in arcgis, it's under interpolation toolset of spatial analyst tools 2 Quote Link to comment Share on other sites More sharing options...
yousef2233 Posted February 7, 2013 Report Share Posted February 7, 2013 Natural neighbor interpolation is a method of spatial interpolation, developed by Robin Sibson. The method is based on Voronoi tessellation of a discrete set of spatial points. This has advantages over simpler methods of interpolation, such as nearest neighbor, in that it provides a more smooth approximation to the underlying "true" function. http://en.wikipedia.org/wiki/Natural_neighbor it's under interpolation toolset of spatial analyst tools 1 Quote Link to comment Share on other sites More sharing options...
HTC Posted February 7, 2013 Author Report Share Posted February 7, 2013 I dont find tools in Arcgis. I only see: IDW, Kriging..... Not seen K nearest neighbour and Regression Krigging. Quote Link to comment Share on other sites More sharing options...
yousef2233 Posted February 7, 2013 Report Share Posted February 7, 2013 k nearest is inside the idw, its the number of neighbors getting used in interpolation, instead of krigging regression use the krigging iterpolation 1 Quote Link to comment Share on other sites More sharing options...
HTC Posted February 7, 2013 Author Report Share Posted February 7, 2013 Thank my friend. Now i was understood the problem. Many thanks. Quote Link to comment Share on other sites More sharing options...
yousef2233 Posted February 7, 2013 Report Share Posted February 7, 2013 your r welcome dude Quote Link to comment Share on other sites More sharing options...
rp314 Posted May 29, 2013 Report Share Posted May 29, 2013 Can someone recommend any software to do k-NN multiple regression? Quote Link to comment Share on other sites More sharing options...
yousef2233 Posted May 30, 2013 Report Share Posted May 30, 2013 I recommend you using GWR (Geographically Weighted Regression) which is so much better http://www.gisarea.com/index.php?/topic/2749-geographical-weighted-regression/ Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.