Reputation
15,108
Next tag badge:
86/100 score
21/20 answers
Badges
6 42 88
Newest
 Good Answer
Impact
~1.2m people reached

2d
revised How compare aggregate data before and after policy?
edited title
May
18
comment Any easy way to cluster GPS trajectories?
Possiby gdal_rasterize, but you should ask on gis.stackexchange.com
May
17
answered Any easy way to cluster GPS trajectories?
May
14
answered Data splitting and cross validation
May
8
awarded  Good Answer
Apr
25
revised Proving conditional probability sequence using the sum rule
edited tags; edited title
Apr
21
revised Do CART trees capture interactions among predictors?
edited title
Apr
21
comment Do CART trees capture interactions among predictors?
The flaw in the argument is that splits are made on subsets of covariates defined by splits done previously.
Apr
15
comment How to do dimensionality reduction in R
I'm afraid its lost forever... Generally it was about implementations of clustering methods which does not explicitly build each-to-each similarity matrix but investigate objects on demand.
Apr
13
answered Different probability values when using DecisionTreeClassifier and RandomForestClassifier
Apr
12
revised How to superimpose two MATLAB images rigidly transformed to perform a metrics
added 4 characters in body; edited title
Apr
10
awarded  Enlightened
Apr
10
awarded  Nice Answer
Apr
1
awarded  r
Mar
23
awarded  Constituent
Mar
21
awarded  Caucus
Mar
15
revised Estimating parameters of Dirichlet distribution
edited title
Feb
25
comment Support Vector Machines and the curse of dimensionality
Well, obviously with a finite sample you can never be sure that some association is not spurious, but IMHO the fact that a feature keeps appearing in such a procedure is pretty much the best confirmation you can get. And still there is an issue whether something is missing (; You can also check out my paper where something like this is used to benchmark few RF-based feature selectors.
Feb
24
comment Support Vector Machines and the curse of dimensionality
Sure. Still this "stochastic agreement" may also fail; if only spurious features are selected, they will be likely different each time and there will be no consensus. This is why FS should be all relevant as defined in this Nilsson et al paper.
Feb
24
answered Support Vector Machines and the curse of dimensionality