LIBSVM is an integrated software library for support vector machines, performing support vector classification, (C-SVC, nu-SVC), regression (epsilon-SVR, nu-SVR) and distribution estimation (one-class SVM)

learn more… | top users | synonyms

1
vote
1answer
28 views

Which machine learning method to use for geographic systems prediction?

I am trying to do experiments on geographic systems prediction. We're working on classifying the location where we sell product most. So, we need to analyze the hestorical data and predict the success ...
1
vote
1answer
15 views

How do interpret statistically NULL SVM Output

I am using LibSVM (3.18) as an implementation of SVM. But every time when I'm predicting the result, it's giving zero. I am following these instructions: I have CSV file (+50K lines), Most of data ...
1
vote
2answers
34 views

SVM for unbalanced data

I want to attempt to use Support Vector Machines (SVMs) on my dataset. Before I attempt the problem though, I was warned that SVMs dont perform well on extremely unbalanced data. In my case, I can ...
0
votes
0answers
11 views

libsvm grid search on Parallel Computing Toolbox Matlab

I am using the following code for grid search on libsvm: ...
0
votes
0answers
4 views

Liblinear types of solver

There is many variants of type of solver in liblinear but I don't understand their differences.Which one I must choose? Also why data must be scaled? duo to some numerical issues? ...
0
votes
1answer
16 views

Rank features by predictive power in trained classifier (LibSVM)

I'm training a SVM classifier on a dataset using LibSVM through Weka, but after I have my trained classifier I'd like to know which features are important to my classifier and which features don't ...
0
votes
0answers
29 views

libsvm: Cross Validation with imbalanced classes

I am using libsvm for a 2-class classification problem. For my testing I use the C-SVM with RBF kernel. My main problem seems to be that the classes are highly imbalanced. While I have 35000 datasets ...
0
votes
0answers
18 views

strange behavior for LIBSVM during cross-validation

I am working on a 2-class classification problem and using LIBSVM. I am facing this following problem with cross validation (CV). Any help is highly appreciated. When I perform a 10 fold CV with -v ...
1
vote
1answer
28 views

Normalized Data vs. Non Normalized Data in SVM Training

Im training an SVM on a dataset containing 17000 events with 30 features. as I see it, this amount of features against the dataset size should suffice by itself for avoiding overfitting. Im training ...
0
votes
0answers
12 views

How do I get the support vectors data (the model file) when using cross validation in LibSVM? [migrated]

I have a problem with the output file when I use ./svm-train ... -v k When I use the parameter -v, the output file doesn't get created, but I need the support ...
0
votes
0answers
21 views

Leave-one-out cross-validation in LiblineaR

So I've been encountering this weird issue with LiblineaR where I get different results from just running LiblineaR with cross set to the size of my dataset and actually looping through the dataset, ...
0
votes
0answers
15 views

Why even my training data failed during the prediction of libsvm [duplicate]

Currently I'm using libsvm for my one class classification problem. I have 10 samples in my training set, 5 samples in my testing set, both of my training and testing set is scaled by svm_scale, then ...
0
votes
0answers
40 views

Libsvm includes training data in testing file

Currently I'm using libsvm (Java) for my one class classification problem. I scaled both of my training data and testing data by svm_scale, the problem is that somehow,libsvm includes both training ...
0
votes
2answers
57 views

Multi class classification always have better result than one class classification?

Currently I'm using svm to classify the test samples to two different classes (True and False). When I use multi class classification, I have both true and false samples in my training set, the ...
0
votes
1answer
47 views

LIBSVM Probability estimates in binary classification

I have Training data like following : 1 1:128.319 2:71.4336 3:130.255 4:292.948 5:96.3541 6:71.942 7:136.189 8:71.5032 9:148.21 10:304.011 11:90.3781 12:99.4496 13:164.268 14:138.892 15:220.871 ...
0
votes
1answer
127 views

One class classification with libsvm. Accuracy results in 0%

A quick recap for what I want to do, I want to determine if a text is written by the same author or not. Thus I use one-class classification. In my training set (18 samples), it looks like this (for ...
0
votes
2answers
83 views

C parameter does not affect accuracy (LibSVM)

I'm working on LibSVM and I'm using linear kernel. But i have a problem with C parameter, it does not affect accuracy during test set! Is it possible? I send a picture with the distributions of the ...
0
votes
1answer
40 views

LIBSVM ROC PLOt [closed]

my question is "What is command used to generate ROC PLOT in libsvm as well as for specificity , sensitivity and precision what is the command in libsvm " i Am using ubuntu
0
votes
0answers
60 views

SVM model does not support probabiliy estimates?

I am doing some classification task with Support Vector Machines (SVM). I am using libSVM (with Matlab support) to predict probability estimates matrix. However, the libSVM displays message that; ...
1
vote
0answers
45 views

LibSVM one class classification nu parameter is not a fraction of outliers?

please correct me if I'm wrong, but the one class SVM theory states, that nu parameter is the upper bound (UB) of outliers in the training dataset and lower bound (LB) of number of SVs. Say I'm using ...
2
votes
1answer
55 views

libsvm_linear kernel_increasing C value

I'm using libsvm in C-SVC mode (-s= 0) with linear kernel (-t= 0), and I'm required to train multiple SVMs( I have four classes). My training and test sets have the same number of instances and ...
4
votes
3answers
200 views

Is a lower training accuracy possible in overfitting (one class SVM)

I am using the heart_scale data from LibSVM. The original data includes 13 features, but I only used 2 of them in order to plot the distributions in a figure. Instead of training the binary ...
0
votes
1answer
91 views

How to construct the feature weight vector (or Decision Boundry) from a linear SVM classifier from scikit?

I use the following code to train an svm classifier: clf = svm.SVC(kernel='linear') clf.fit(train_mat, train_labels) that fit the data and save the info in the ...
1
vote
0answers
61 views

Logistic Regression not quite working

I'm playing around with logistic regression (using scikit-learn, which uses liblinear) I created some example data sets, and it often works well. Since I created the data sets, I already know the ...
1
vote
0answers
42 views

Leave one subject out: Evaluating classifier performance

I have a small dataset, and I am trying to train 1-vs-all classifiers (class X or not class X) using a linear SVM. The dataset is simply too small to split up 60/40, 80/20 or anything like that. Also, ...
0
votes
0answers
20 views

what can be inferred from the error plot of my classifier?

I am working on a classification task where I use 12 features. My training set has 400 samples of positive data and 2000+ samples of unlabelled data. While testing has 34 positive samples and 999 ...
0
votes
0answers
38 views

Decompose LibSVM model into binary classifiers

I have a multi-class .model trained with LibSVM. Is there a way to decompose this model into different binary .model files? For example, if I have a .model trained for $n$ classes, I would like to ...
0
votes
1answer
113 views

Using LibSVM for anomaly detection

Im trying to create a one class SVM using libSVM. However whenever I run the svmpredict function it always return an accuracy of 0%. ...
0
votes
0answers
127 views

From one-dimension-SVR to multi-dimension-SVR (?) using MATLAB/LibSVM

this is my first post here. I already checked the Thread-Tags but I did not find any answer... My problem/question is how I can upgrade my SVR from one-dimension- to multi-dimension-regression. Its ...
0
votes
0answers
38 views

Using ksvm in R with differents constraints

I would like to use ksvm from the kernlab package in R to train and test a support vector machine. I'm able to use it but I would like to use different values of C for each observations. To be more ...
1
vote
1answer
120 views

SVM training method (or alg.)

I'm using SVM classification (Matlab) within my research works, and I want to know: The advantages and disadvantages of each training algorithm, i.e., SMO, LS and QP In general case, what is the ...
0
votes
1answer
103 views

Modifying decision function in LibSVM

The decision function for C-support vector classification is $sign(wT\phi(x)+b)=sgn(\sum_{i=1}^{l}y_i\alpha_iK(x_i,x)+b)$ How can I modify it to ...
0
votes
0answers
78 views

LibSVM: A few noisy labels turns good classifier bad. What is going wrong?

I'm using libSVM to train a pair of classifiers, each using an independent "view", or feature set, extracted from the same dataset, i.e. co-training. As part of the co-training method, I utilise a ...
2
votes
1answer
221 views

SVM regression with LIBSVM and python - Expected runtime

Im new to regression, but quit experienced in classification and machine learning. In Classification, the state of the art technique for classification is SVM, so the first solution I would like to ...
1
vote
1answer
344 views

Correct use of cross validation in LibsSVM

I am classifying data points from two different groups using LibSVM. I do the following: Creating the input file for LibSVM. ...
0
votes
0answers
135 views

How to set the labels of One-Class SVM on LibSVM?

I try to perform one-class SVM Schölkopf on LibSVM. But all the predicted labels are "-1 " which is very strange. Below is my code. I use MATLAB and libSVM. % ex5 with iris dataset load iris.mat ...
0
votes
1answer
109 views

boosting with linear svm

I am working on boosting classifier. I am planning to use linear svm as the weak classifier. I am using liblinear for it. My question is how can I weight each instance of liblinear based on the ...
0
votes
1answer
84 views

What is the loss function for C - Support Vector Classification?

In article LIBSVM: A Library for Support Vector Machines there is written, than C-SVC uses loss function: $$ \frac{1}{2}w^Tw+C\sum\limits_{i=1}^l\xi_i$$ OK, I know, what is $w^Tw$. But what is ...
2
votes
1answer
449 views

Training one class SVM using LibSVM

I hope to use one-class SVM of LIBSVM to train a training samples so as to get a model. Then, I will use the model to predict whether the new test data and the training data is same type or not. In ...
1
vote
2answers
100 views

Normalization in SVM

I have applied libsvm with a linear kernel to a set of instances and I have obtained a 68 % success: ...
0
votes
1answer
204 views

Support Vectors Not Falling on Margin Lines for e1071 and kernlab packages in R?

In a previous thread, Computing the Decision Boundary of a linear SVM model(Computing the decision boundary of a linear SVM model), the following R code was given as a way to compute the formula of ...
0
votes
1answer
301 views

Cross Validation folds in LibSVM

how the folds in cross validation of libSVM are done?, I mean, the splits are done aleatory? for example if I have 100 data and it is 4-fold then when it is an aleatory folding then can be splited ...
0
votes
1answer
210 views

LIBSVM “Warning: using -h 0 may be faster”

I am using LIBSVM package (Matlab version) in order to perform wind power forecasting. My training dataset is quite big, approx. 52000 values of 4 parameters. The ...
0
votes
0answers
183 views

Divison of multi class into binary class using libsvm (not classification)

I am working on multi class imbalance data.I want to divide my classes based on one verus one method and then perform sampling techinque How can I do that I just want to divide the classes not ...
0
votes
0answers
72 views

Using a model trained by SvmPerf or LibLinear

I am writing a machine learning program in Node.js, and I want to use libraries such as LibLinear and SvmPerf, that are not available in Node.js. So, for training, I just create a file with the ...
0
votes
0answers
72 views

Prediction of data using regression for time series

I try to prepare the data for regression with time series analysis. I use LIBSVM on Matlab Say I have N days of price and I want to predict the next day's price. The training set X is thus a vector ...
0
votes
1answer
86 views

How to put data in time series regression

I tried to get a result in regression on LIBSVM, and I get the same problem on single feature data or multivariate data. Suppose I try to find the price of ...
2
votes
0answers
208 views

Feature importance scores of SVM multiclass one-vs-one design

Info about dataset: 5 classes, 200 trials, 100 features. (I know about the trial to feature ratio being very low, but can not avoid this here and still got well enough classification results.) ...
0
votes
2answers
162 views

LIBSVM parameter search in time series

I try to predict values for regression in LIBSVM. My data is in time series. I use gridregression.m file in LIBSVM to find optimal parameters c, g and p. Gridregression.m file use cross validation to ...
0
votes
1answer
223 views

Prediction in support vector regression

Has anyone attempted prediction using support vector regression? I'm using LIBSVM, but I'm not sure how to use SVR in either univariate and multivariate time series. Say we have stock prices for $N$ ...