similar to: Statistical significance of a classifier

Hello, I have some trouble in achieving the desired parametrisation for the weka classifier functions, using the package RWeka. The problem is, that the functions result=classifier(formula, data, subset, na.action, control = Weka_control(mycontrol)) do not seem to be manipulated by the mycontrol- arguments Perhaps this should be resepected via the handlers- argument , but the

Hi, I am using some classifiers in RWeka packages and met a couple problems. (1) J48 implements C45 classifier, the C45 should be able to handle missing values in both training set and test set. But I found the J48 classifier can not be evaluated on test set with missing values--it just ignore them. (2) The ensemble classifiers in RWeka such as bagging and boosting: there is a

Le jeudi 09 f?vrier 2012 ? 15:31 +0200, Kari Ruohonen a ?crit : > Hi, > I am trying to replicate a training of AttributeSelectedClassifier with > CFsSubsetEval, BestFirst and NaiveBayes that I have initially done with > Weka. Now, I am trying to use RWeka in R. > > I have a problem of passing arguments to the CfsSubsetEval, BestFirst > and NaiveBayes. I have first created an

Dear All, I'm currently doing a project about unsupervised learning, and I'll be using R to analyse a few network traffic datasets downloaded off Andrew Moore's website (http://www.cl.cam.ac.uk/Research/SRG/netos/nprobe/data/papers/sigmetrics/index.html). Could anyone shed some light on how to build a classifier from the training set A, and then use a validation set B to assess the

Hi, Currently I testing the packets that contain built-in features for classification. Actually I looked packages such as: e1071, Klar, Caret, CORElearn. However, from what I noticed when building a naive Bayesian classifier, that they package use of the finite mixture model to estimate P (x | C) and using a normal distribution. In my research I use binary data and I want modeled P (x | C), eg the

Hello, I just compiled a new kernel.iptable/iproute2 to fresh up my router. There is a little problem... The fw classifier does not work in tc. I read some note found in the source of iproute2 and it says: "* To use "fw" classifier you will need ipfwchains patch." I can''t find this patch. Do you have it? Is there any other way to implement the "fw"

All, I am sure this is just something I am doing wrong, but I am trying to implement a node classifier (shell script) to have my nodes inherit the proper classes. However, when I run it I am not getting any classes except those that are defined in default. Rather than going into detail showing all of my configurations right off the bat, I have a question: I just stood up a

Dear R Helpers I am having difficulty understanding how to use the penalty matrix for the nomROC function in package 'nonbinROC'. The documentation says that the values of the penalty matrix code the penalty function L[i,j] in which 0 <= L[i,j] <= 1 for j > i. It gives an example that if we have an ordered response with 4 categories, then we might wish to penalise larger

Dear R-philes, I am having some trouble averaging across multiple runs of a classifier in an ROC Curve. I am using the ROCR package and the plot() method. First, I initialize a list with two elements where each element is a list of predictions and labels: vowel.ROC <- list(predictions=list(), labels=list()) For every run of the classifier, I append the scores and labels to their

Hi, I understand for 1 d classifiers, you can use ROCR package. Is there a package you can plot ROC curve for 2d classifiers? One of my colleagues asked me about this. I have been quite puzzled, conceptually, how you can do the ROC curve for 2d classifiers. Can someone share his/her knowledge or experience? Thanks in advance. -- Waverley @ Palo Alto

I was wondering, whether there is a way to have statistical significance test for cluster agreement. I know that I can use classAgreement() function to get Rand index, which will give me some indication whether the clusters agree or not, but it would be interesting to have a formal test. Thanks.

Hi everyone, I have trouble to use RWeka, I tried: (w=weather dataset, all preditors are nominal) > m<-J48(play~., data=w) > e<-evaluate_Weka_classifier(m,cost = matrix(c(0,2,1,0), + ncol = 2),numFolds = 10, complexity = TRUE,seed = 123, + class = TRUE) it gives me exactly what I want, but when I tried the same classifier on the other published data: (iris dataset has all numeric

Hi, I have a small doubt regarding naive Bayes. I am able to classify the data's properly but i am just stuck up with how to get the probability values for naive bayes. In the case of SVM we have "attr" function that helps in displaying the probability values. Is there any function similar to "attr" in naive Bayes that can be used for displaying the attribute values. my

Hi everyone, I have a test set with more than 1000 cases, when I use evaluate_Weka_classifier(RWeka)to evaluate my classifier on this test set, the output shows me the result of only 83 cases. I do have missing values in predictors, so I tried na.acton=na.pass, but it dosen't help. Now I confused, why RWeka ignore so many cases in my test set? Is there any setting I didn't notice in

Hi, I don''t really know if this is the correct list for asking this, if not a referral to the correct list would be very much appreciated. Ok here it goes. I would like to create a u32 filter for tcp packets that are larger then 1480 bytes but I am not really sure how of if this can be done. I know about the pattern and mask thing but I am not terribly familiar with the masking

Hi all! On http://lartc.org/howto/lartc.adv-filter.html I read that a classifiers available bases the decision on how the firewall has marked the packet and on http://lartc.org/howto/lartc.qdisc.filters.html the following example: "tc filter add dev eth1 protocol ip parent 1:0 prio 1 handle 6 fw flowid 1:1" "iptables -A PREROUTING -t mangle -i eth0 -j MARK --set-mark 6" My

Hi, Apology for this question being off the topic (OT) of R, though I expect this list might be the best place on the net to ask this question. In brief, the question is: what classification algorithm can one use if the features are histograms? I have a classification problem, and believe that histograms of the distribution of some values may be the best "feature" to use. To

Hello all, As I have promised I am sending my QoS rules. This now works fine with u32 classifier (and parent 1:0 that I could not understand why it did not worked well before). Att, Nataniel Klug ------------------------ #!/bin/sh #------ # Script de QoS Cyber Nett #------ # Nataniel Klug # suporte@cnett.com.br #------ TC="/sbin/tc" IPT="/usr/local/sbin/iptables"

hi, at first; thanks for the help on getting confidence intervals in R. now I have a pure statistical question. I hope you don't mind if I ask ... I have an expectation of how large my beta-weight in a regression should be - so I have an "ideal" or expected regression line. Now the real beta-weight is less then the expected and when I draw the confidence interval lines

Dear R community, Is this sample size large enough to study differences between two groups of the populations? Q1: do the body temperatures differ between the two groups of the overwintering turtles juveniles and adults? One group (adults) has 6 turtles Second group (juveniles) has 1 turtle. There are 3 replications, i.e. the experiment was repeated over the three years, but using