Tagged Questions

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

0
votes
0answers
5 views

One Class SVM Training in LIBSVM

Could you kindly reply my queries in context of using LIBSVM for One class SVM. Assuming I have samples from one class only, do I need to put a lable for each sample...but a training file without ...
0
votes
0answers
5 views

LibSVM totalSV for multiclass

I am performing classification of K=9 classes using linear SVM with libSVM (MATLAB warp) I am using 400 samples of data to perform the training and I'm getting: totalSV: 203 I know libSVM uses ...
0
votes
0answers
12 views

SVM - RFE for nonlinear case

Could someone give me some reference or idea on how to implement SVM-RFE for nonlinear cases. As we all know that SVM RFE for linear case is well established by Guyon et al. for (gene) selection where ...
0
votes
0answers
17 views

Support Vector Machine with zero bias term

I'm looking for an algorithm to solve SVM with zero bias term. So dual form of such SVM is $max_\alpha \sum_i^n \alpha_i -1/2\sum_i^n \sum_j^ny_iy_jK(x_ix_j)\alpha_i\alpha_j$ subject to: $0 \leq ...
0
votes
0answers
8 views

SVM tends towards one class

I implemented a 1 vs all multi class SVM using libsvm. When I use the trained SVM to predict it will just assign all the samples to the class that is most abundant. Is there a chance that the data I'm ...
0
votes
0answers
12 views

The Shogun Machine Learning toolbox for SVM with precomputed kernel and zero bias

Can I use the Shogun Machine Learning toolbox for SVM with precomputed kernel and zero bias. I should be able to input pre-computed kernel and I also should be able to set bias zero.
0
votes
0answers
11 views

SVM with pre-computed kernel and zero bias

I have an optimization function, where I need to give my own kernel matrix and bias value is zero. The kernel matrix is calculated using the data but there is no specific formula for it. If I have a ...
0
votes
1answer
19 views

libsvm: scaling data results in less features?

I've scaled my training and testing data in BASH like so: ...
0
votes
0answers
3 views

How to account for different ratio of samples during training and detection using a support vector machine (svm)?

Consider the following object recognition case: Detection of objects in an image using a sliding window approach in combination with a svm model. During sliding window search using multiple scale ...
0
votes
0answers
11 views

How to find a good model for an object recognition case using a support vector machine (svm)?

Consider the following example of an object recognition case: I'm trying to detect objects in an image using histograms of oriented gradients (hog) features. The feature vector resulting from hog is ...
0
votes
0answers
9 views

LIBSVM for pre-computed kernel and zero bias (b values is zero)

I want to do binary classification and I'm using LIBSVM library for that. I have a precomputed Kernel and my bias value (b) is zero. Can I do this in LIBSVM or do I have to use some other library? ...
0
votes
0answers
13 views

Usage of libsvm with RBF kernel and no Offset

I'm using libsvm for the binary classification and using a precomputed Kernel. In my particular problem there is no bias term (it's zero). Is there anyway to adjust the bias term in libsvm (and not ...
0
votes
0answers
12 views

Weka SMOreg and LIBSVM with linear kernel problems

I want to test a dataset in weka using either LIBSVM with an e-SVR or SMOreg for regression. I also choose a linear kernel in both (in SMOreg i use an exponent=1 in a non normalized polykernel). ...
0
votes
1answer
16 views

What is the toolkit that implements Cost sensitive Support Vector Machine?

I need implementation of cost sensitive support vector machine. The cost is different for each training example (unlike each class). So problem is to solve $max_\alpha$ $-1/2 \sum_{i,j} ...
0
votes
0answers
11 views

how to run libsvm with histograms?

My train set is array of histograms, each one is 100x5004, as each histogram represent one action, and it contains 100 points. I have train set of array of size 10x(100x5004), and labels set of size ...
0
votes
2answers
59 views

bad results with SVM

I have a dataset of 1157 samples. The number of features are 6 and I have only one output. My training error is very poor although I have tried a wide range of cost values and followed the recommended ...
0
votes
0answers
20 views

usage of libsvm: all the predicted labels are the same

I'm trying to use libsvm to train a classifier on the following dataset: (1) The number of class -1 are the same as the number of class 1 (2) The feature value has been scaled to [-1 1] (3) I've ...
0
votes
1answer
22 views

usage of libsvm-matlab version

Based on the readme file in the libsvm package. The svmpredict matlab version function is used in the following way: ...
0
votes
0answers
37 views

predicting time series with support vector machine using R

I am planning to do time series prediction using support vector Machine. I could not find any materials about time series application of support vector machines using R or Mat-lab. Similar question ...
1
vote
2answers
55 views

Varying LIBSVM predictions based on test series labels

So I have a pretty well testing SVC train series which puts me into the mid 80 percentile without outrageous C/g values. My current C value is 2.0 and gamma is 0.5. Good numbers across the range ...
0
votes
1answer
46 views

How to prepare my data for SVM classifier in matlab

I am new to SVM and Matlab. I would like to have an example how to prepare my data to be as input to the SVM classifer (using libsvm) let us assume that i have a group of words first i have ...
1
vote
1answer
58 views

Selecting most realistic C and g params after gridsearch

I just ran an extended SVC gridsearch in libsvm on about 9000 multi-dimensional vectors representing a time series. Here are the highest scoring results: ...
0
votes
1answer
40 views

About grid search to find the best value of C

I know when I want to find the best values of C and gamma, I should use grid-search. But in my case I want to find just the best value of C. So, this is called a line-search. Is there any function ...
0
votes
0answers
38 views

How to retrieve the prediction equation in R?

I have developed a prediction model prototype in R. The model uses Support Vector Regression to predict. But I need to develop the entire solution in Visual C++ for a real life implementation. I ...
1
vote
2answers
66 views

Linear SVM prediction time is scaling in an unexpected manner based on training data

I'm using LIBSVM to do some training as it was recommended by Andrew Ng and is used under the hood in SciKit Learn. LIBSVM is doing something different than what I expect though: My beliefs are as ...
0
votes
2answers
45 views

Instance weighing in libsvm/liblinear

I often use the instance weights with Libsvm for classification problems. http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/#weights_for_data_instances Does anyone know the details of the algorithm that ...
0
votes
1answer
56 views

How to format multi-row time series data for LIBSVM regression

I would have expected this to be covered in detail by the LIBSVM tutorial but after hours of wasting time googling for answers I've had to throw in the towel. What I am trying to do is rather trivial ...
0
votes
0answers
26 views

Examine SVM result by plotting histogram of decision values of training samples

I'm working for object detection(computer vision) and have some problems in SVM training. My training configuration is as below. Balanced training set (positive 3998/ negative 3998) The dimension of ...
0
votes
0answers
27 views

How should I do a grid search for the gamma in SVM?

I know that you can do a grid search for the C parameter (-c) in libSVM by going through value that go from 10^-5 to 10^5. How should I go about finding the optimal epsilon parameter (-p) ? Is ...
0
votes
0answers
64 views

Find linear SVM feature weights using libsvm

I'm trying to use linear SVM to do some feature selection. I'm using libsvm, but I cannot figure out how to find feature weights. The model file created looks something like this: ...
0
votes
0answers
46 views

Parameter optimization of SVM

Currently I am using SVM to perform some classification task. I use libSVM with Matlab interface. From the practical guide of SVM (Link), we know that there are two parameters need to be tuned, namely ...
3
votes
2answers
72 views

Assigning even partitions for Cross-Validation

This is a very basic question about cross-validation. Say that I have a sample size of 2901(or any difficult to divide number). How do I split this up into equal partitions (other than n=1)? And how ...
1
vote
3answers
51 views

Kernel selection using SVM for keyword frequency classification

I have data in Weka .arff multiple-class training and testing data representing daily word frequencies in RSS feeds as follows: ...
0
votes
1answer
60 views

Interpretation C value in linear SVM

My C value is very low (close to 0). Does this mean that my feature (dimensions) have no real separative (and thus predictive) value? (As the SVM basically chooses to ignore the training data ...
1
vote
1answer
87 views

Simple SVM Question

For a linear SVM, the documentation tells me the formula is: $$ \frac{1}{2}w^Tw+C\sum\limits_{i=1}^l\xi_i$$ Please explain to me in layman's terms what w (and ΞΎ) represent. Is w the distance to the ...
0
votes
0answers
8 views

Different Scoring with libsvm

I have been using libsvm for binary classification, e.g. given data $(x_1,\dots,x_n)$ and labels $(y_1,\dots,y_n)$ where $y_i\in\{0,1\}$. Now I am wondering if the prediction of my model can be tuned ...
0
votes
1answer
47 views

SVM cost parameter

In a SVM with linear kernel, could you explain to me what exactly the C parameter is/represents? An example why it's important to select a good value for C would also be appreciated. Thank you.
0
votes
0answers
57 views

Cross Validated Squared Correlation Coefficient in SVM regression

I am applying SVM regression on a data set on "Karthikeyan Melting Point Dataset" present here. I am using LIBSVM for it. After applying regression LIBSVM returns the cross validated squared ...
0
votes
0answers
46 views

Using LibSVM regression to predict dependent variable

My intention is to use SVR with 1-10 independent variables to best predict a dependent variable. After separating the data into scaled training and testing files containing the independent variables ...
0
votes
1answer
27 views

How to increase a particular terms's weightage?

I am doing Text classification using LibSVM in Rapid Miner. I am using TFIDF values for processing documents. I need to Increase weightage of some terms in the documents(for eg. words in BOLD and ...
1
vote
1answer
89 views

Extracting decision function variable from libsvm

I'm trying to use LIBSVM's single class SVMs for some classification and need to extract the following sum post classification (i.e. the variable that the decision function takes in) $$ ...
0
votes
0answers
44 views

One vs All appraoch to multiclassification in LIBSVM

I'm working on multiclass classification problem (precisely 4 classes). I want to use the most simple approach for this problem: one-vs-all! I have 4 different test set (only labels are different). ...
1
vote
2answers
284 views

Plotting the decision boundary of a kernel SVM (RBF)

Suppose we are given a training set of 2D points that are linearly non-separable. I train a binary SVM with an RBF kernel in order to classify them. What I want to do is to draw the desicion boundary. ...
1
vote
0answers
48 views

STRING KERNELS FOR LIBSVM

I'm working on a protein classification problem and i'm using edit distance kernel defined in libSVM. Now, for instance, the implementation of spectrum kernel is very difficult, but i want try to test ...
0
votes
2answers
37 views

Evolution strategies in libsvm

I'm working on protein multi-classification problem. I'm using libsvm and the edit distance kernel. This kernel depends from a parameter (gamma). I'm able to get the best parameters (gamma and C) ...
0
votes
0answers
31 views

Cross validation and accuracy calculation in lib-linear

I have two questions related to cross validation in LIBLINEAR I have 1000 documents from which i take 300 documents for training and rest 700 for classification . I train 300 documents with ...
0
votes
0answers
45 views

Different results in libsvm and R for regression prediction

i use e1071 package in R for SVM regression prediction and tried the same test in C# libsvm and got different results. All the parameters (gamma,cost,eps) are the same and so is the scaling method. ...
1
vote
1answer
53 views

LIBSVM RBF kernel time complexity

Is it known what is the LIBSVM time complexity when using RBF kernel functions?
0
votes
0answers
25 views

LIBSVM not Perfoming well in weka as opposed to other algorithms

I do have an issue with libSvm Weka which i feel it is very weird However, when i tried it with other algorithms such as naive bayes and j48, it achieved a high accuracy of 78%. Moreover, the same ...
0
votes
0answers
54 views

LIBSVM evaluation functions for probability estimation

I found this tool that is providing some well known evaluation functions like precision and recall. Unfortunately this tool supports only binary-class C-SVM. I was wondering where I can find similar ...