similar to: Features Selection using RWeka

Hi, I'm trying to use Information Gain for feature selection. There is a InfoGain implementation in Weka: *weka.attributeSelection.InfoGainAttributeEval* Is it possible to use this function with RWeka? If yes how? list_Weka_interfaces doesn't show it and there is no make function for AttributeEvaluators. Is there any other implementation of InformationGain in R? Thank you Michael

Hi all, Although I sent the mail to Piotr, the author of FSelector, it should be better to ask here to let others know. Yanwei Begin forwarded message: From: Yanwei Song <yanwei.song@gmail.com> Date: September 7, 2011 4:41:58 PM EDT To: p.romanski@stud.elka.pw.edu.pl Subject: FSelector and RWeka problem Dear Piotr, Thanks for developing the FSelector package for us. I'm a new

Hi All,  After fitting a model with glm function, I would like to do the model selection and select some of the features and I am using the "step function" as follows:  glm.fit <- glm (Y ~ . , data = dat, family = binomial(link=logit)) AIC_fitted = step(glm.fit, direction = "both") I was wondering is there any way to select the features based on BIC rather than AIC? is there

Hi, You should also check my msc.features.select from caMassClass package. It has feature selection algorithm that I found useful in case of mass-spectra data. It performs individual feature selection and/or removes highly correlated neighbor features. Jarek -----Original Message----- From: r-help-bounces at stat.math.ethz.ch [mailto:r-help-bounces at stat.math.ethz.ch] Sent: Friday, January

Hi Rishabh, I had this feeling before. This is a really nice idea BUT we can not go ahead with the project which is still not tested in the experimental settings. Though it may be a wonderful research exercise, I would still vote to go for the state-of-the-art methods which are completely published with full details and experimental results. Hope you get my point. Regards, Parth. On Wed, Apr 4,

Hello, I have a problem in feature selection I would be thankful if you can help me. I have a dataset with limited samples (for example 100) and a lot of features (for example 3000) and i have to do feature selection. if i use cross validation (for example *10 fold*) i rank the features based on 90 samples (using svmrfe method) i achieve ranked feature for example {f2,f4,f1,f3,...} (it means f2

Hi everyone, I have a question about using RWeka package? we know that instruction make_Weka_classifier that can help us to build a model,and evaluate_Weka_classifier instruction can help us to evaluate the performance of the model using on new data. But I have a question about how to using the parameter numFold in evaluate_Weka_classifier.Cross-validation means that using some parts to

Hello! I have a question. I am working with createFolds: folds<- trainControl(method='cv', index=createFolds(data$Score,list = TRUE)) I need to iterate over folds to extract the indexes from each fold. For example, if I do folds$index$Fold01, it contains: 5 11 17 29 44 50 52 64 65 I need to iterate over each $Fold_i to extract the indexes, but I can't do it because I

Hello guys! I am working with some classifiers ( SVM,C4.5,RNA,etc) using 10-C.V. Once I have the model of each one, I make the validation of these models in one dataset. Then,with my model and the dataset, I extract a confusion matrix to know the capacity of prediction from the model. And finally, I extract the accuracy of this prediction based on the diagonal from the confusion matrix. The

Hello! Does anyone know how to apply bagging for SVM? ( for example) I am using adabag package to execute bagging but this method, "bagging", works with classification trees. I would like to apply my bagging to other classifiers as SVM,RNA or KNN. Has anyone do it? Thanks!! [[alternative HTML version deleted]]

Folks, I am looking for a means for calculating the 1-R^2 ratio for variable selection to mimic the values of PROC VARCLUS in SAS. While there may be better methods for variable selection, we are trying to duplicate published results at this time. To date, I have been unable to find a way to obtain this value from Hmisc::varclus , stats::hclust and clustvarsel::clustvarsel. Can any of you give

Running Weka's command line with calls to system(), like this > system("java weka.classifiers.bayes.NaiveBayes -K -t HWlrTrain.arff -o") === Confusion Matrix === a b <-- classified as 3518 597 | a = NoSpray 644 926 | b = Spray === Stratified cross-validation === === Confusion Matrix === a b <-- classified as 3512 603 | a = NoSpray

Hi, I am doing a project on authorship attribution, where my term document matrix has around 10450 features. Can you please suggest me a package where I can find the feature selection function to reduce the dimensions. Regards, Venkata Satish Basva [[alternative HTML version deleted]]

I too have this problem. Everything worked fine last year, but after updating R and packages I can no longer do word stemming. Unfortunately, I didn't save the old binaries, otherwise I would just revert back. Hoping someone finds a solution for R on Windows. Thanks! There is a potential solution for R on Mac OS from Kurt Hornik copied below, but I cannot get this to work on Windows.

Hi All, Sorry if this is a repost (a quick browse didn't give me the answer). I wonder if there are packages that can do the feature selection and classification at the same time. For instance, I am using SVM to classify my samples, but it's easy to get overfitted if using all of the features. Thus, it is necessary to select "good" features to build an optimum hyperplane (?).

Hi, I have about half million binary features, and would like to find a model to estimate the continous response. According to the inference, I can express predictors and response by linear model. (ie. Design matrix: large sparse matrix with 0/1. Response: Continous number) Since it is not a classification problem, someone suggested me to try random forest in R. However, in the randomForest help

See, I have built my own genetic algorithm already and tested it on this problem. I have a solution, but due to the heuristic nature of GA, I cannot guarantee that it is the optimal subset. If I was simply doing this for a company project, you are spot on with the type of algorithm I would use, but I am doing this for a scientific paper. I need to be able to find the optimal subset over my

Good afternoon. I would like to know if there is any function in R to do LOOCV with these classifiers: 1)SVM 2)Neural Networks 3)C4.5 ( J48) 4)KNN Thanks a lot! [[alternative HTML version deleted]]

Dear all, I am trying to construct a wrapper that uses random forest to evaluate the subsets using RWeka and when I do: nombi <- make_Weka_filter("weka/attributeSelection/WrapperSubsetEval") datbin<- nombi(gene ~., data=X1X2X3X4W, control =Weka_control( B = list("weka.classifiers.trees.RandomForest"))) I also have tried with an other induction algorithm:

Hi, I would like to find some methods about feature selection, but I only know the package "randomForest" after searching for a while. Could you recommend some other packages of feature selection? Thank you very much. Sincerely yours, Chad Yang. ==========================================================