Tagged Questions

2
votes
0answers
34 views

SVM classifier (with soft-margin) implementation in R, gamma value and quadprog

I'm trying to implement a Support Vector Machine classifier in R and I have to solve the optimization problem using the quadprog R package which solves problems of the form : $$min_b \frac{1}{2} ...
0
votes
1answer
41 views

regularized logistic regression and support vector machine

L2 regularized logistic regression differs with L2 regularized support vector machine with their loss function. Are there more deep differences for these two models? I tried several data sets, and ...
0
votes
0answers
40 views

How can one setup a linear support vector machine in excel?

Through the last year I have been working with support vector machines for a binary text classification task. Having used software as R and Rapidminer I have not spend much time on understanding what ...
0
votes
0answers
20 views

Pegasos algorithm parameters estimation

For learning pourpose I'm testing my own implementation of the Pegasos algorithm and I'm getting a quite high error rate. My dataset contains 20k examples and I'm using 17,5k of them for the training ...
2
votes
3answers
172 views

Why is svm not so good as decision tree on the same data?

I am new to machine learning and try to use scikit-learn(sklearn) to deal with a classification problem. Both DecisionTree and SVM can train a classifier for this problem. I use ...
1
vote
2answers
55 views

Highly unbalanced test data set and balanced training data in classification

I have a training set with about 3000 positive instances and 3000 negative instances. But my test data set is pretty much un-balanced. The positive set only has 50 instances and negative has 1500 ...
0
votes
1answer
16 views

in nonlinear binary classification problems, which is the optimal dimension for make it lineary separable?

My question pertains to linear separability with hyperplanes in a support vector machine. Is posible to determinate the optimal dimension in which i have to transform a training data set for make it ...
-1
votes
0answers
23 views

kernels distances gram matrix classification

Could you please explain some thing about kernels? As I understand it is technique to map the feature space into a high dimensional feature space where we could separate two classes by a linear ...
1
vote
1answer
55 views

SVM model selection for datasets with sharp corners

I'm working with an artificially generated dataset that is separated by many sharp corners. As an example, imagine an H-shape in a 3D (or higher-dimensional) space. Points within the H are positive, ...
1
vote
1answer
59 views

SVM confidence according to distance from hyperline

For a probabilistic multi-class classifier we can get probabilities of membership of a new point $x$ to each class $y_i$; in case of 3 classes suppose that we get $P(y_a|x) > P(y_b|x) > ...
0
votes
0answers
38 views

CV acc mismatch the prediction

Setting the Context : My project is in C++, I'm using OpenCV svm here I used the function train_auto for the CV, however, I implemented my own cross-validation base on this Matlab example here (I ...
1
vote
0answers
72 views

SVM confusion matrix whose dimensions are more than two

While I am using SVM, I train it with a train data and then I try to predict a sample if its label is -1 or +1. However, I see some confusion matrice for SVM like below. Mine are 2x2 matrice but their ...
0
votes
0answers
81 views

How to compute precision for a multiclass problem?

I have a question about calculating precision on a multiclass problem. If the true positives of some actual class is 0, and its false negatives is also 0, then how to calculate its recall? In this ...
3
votes
2answers
126 views

On combining SVMs

Suppose we have a supervised training set $T=\{ (x_1, y_1),\dots, (x_n,y_n)\}$ where $x_i$ is an example and $y_i \in \{-1,+1\}$ is its label. Further suppose that examples are only observable through ...
4
votes
2answers
522 views

how to calculate precision and recall for multiclass classification using confusion matrix?

all, I wonder how to compute the precision and recall using confusion matrix for multi-class classification problem. In specific, one data can only be assigned with most probable class/label. I like ...
0
votes
0answers
56 views

LIBSVM-based classifier assign very low score to positive validation files

Recently, I have been applying the LIBSVM to build a classifier based on a set of documents. The positive set has about 20000 files and negative set has about 50000 files. The built classifier is then ...
1
vote
1answer
95 views

What data from MATLAB's svmstruct are needed for classification in a different language?

As the title already states Iam wondering what data exactly are needed from the MATLAB svmstruct to be able to classify a new instance outside of MATLAB, e.g. in ...
2
votes
0answers
142 views

Example of classification dataset where SVM with linear kernel performs well

I'm looking for a dataset (preferably with a story, at any rate a real dataset) where a SVM with a linear kernel performs well...in other words i'm looking for a dataset where the class boundary is ...
0
votes
2answers
219 views

Which SVM kernel to use for a binary classification problem?

I'm a beginner when it comes to support vector machines. Are there some guidelines that say which kernel (e.g. linear, polynomial) is best suited for a specific problem? In my case, I have to classify ...
2
votes
1answer
172 views

Training an SVM classifier with non-negative weight constraint

I have a problem, where I need to learn a classifier (such as SVM) such that all the learned weights to be non-negative due a constraint on the classifier function. I found out that "SVM Struct" is ...
0
votes
0answers
52 views

Significance index of unlabeled data prediction

I have a training set used to train an SVM classifier, the model found is used to predict a dataset of several unlabeled examples. I would like to know how to extract an index of the goodness of the ...
2
votes
2answers
107 views

Increasing the sample size does not help the classification performance

I am training a SVM classifier based on a given document collections. I started from using 500 documents for training, then I add another 500 for training, and so on. In other words, I have three ...
1
vote
1answer
66 views

Relationship between number of training set and classification performance

Are there any research/paper on the relationship between the number of documents for training and the classification performance using support vector machine?
0
votes
0answers
95 views

SVM with svm and svmpath function

I am trying to compare the R functions svm (library: e1071) and svmpath (library svmpath). ...
1
vote
1answer
73 views

SVM optimization problem

I think I understand the main idea in support vector machines. Let us assume that we have two linear separable classes and want to apply SVMs. What SVM is doing is that it searches a hyperplane ...
3
votes
2answers
231 views

SVM prediction sensitivity when compared to neural networks and logistic regression

Basically I want to classify a rather rare status (about 2% of the 2000) with some predictors. I have used logistic regression, neural network, and Support Vector Machines to do it. All the ...
3
votes
1answer
84 views

Regarding redundant training data in building SVM-based classifier

To build a SVM-based classifier, I have a training data set consisting of N data points. Some of them are redundant. For instance, there have 50 data points which are exactly the same, and there have ...
2
votes
0answers
71 views

Multiclass classification with SVM a question about the feature vectors

I was told I had to direct my machine learning questions to this site. So here it goes. I'm trying to do Multiclass classification with SVM. I have 7 classes. Now I was wondering if the following is ...
0
votes
1answer
449 views

About SVM cost and gamma parameters tuning

I am using R and e1071 package to tune a C-classification SVM. My question is: regardless of the kernel type (linear, ...
3
votes
2answers
141 views

Transform data to a higher dimensional space

The classes are as follows: $C_1=\{3, 3.5, 4, 4.5, 5, 5.5, 7\} \cup \{15, 16, 17\}$ $C_2=\{0,0.5,1,2\}\cup\{8,9,10,11,12,13\}\cup\{20,25,30\}$ And we wish to classify numbers from the interval ...
0
votes
0answers
69 views

Same (poor) results with linear and non-linear classifiers

I am doing classification of "text quality" using four classes and using 30 features with 1300 samples. I am using the following classifiers: LDAC based on linear discriminant analysis from mlpy. ...
3
votes
2answers
249 views

training approaches for highly-imbalanced data set

I have a highly-imbalanced test data set. The positive set consists of 100 cases while the negative set consists of 1500 cases. On the training side, I have a larger candidate pool: the positive ...
0
votes
0answers
77 views

Best way to classify hierarchical data

First of all, I want to be clear that I am new to AI, although I've learned a bit about classification. Let's say I have a hierarchical sets of words: ...
4
votes
0answers
63 views

Classifiers with post-training constraints on the prediction space

I'm familiar with using tools like SVMs and decision trees for discrete classification problems. But one detail that I have not encountered in that domain is: what do you do if your classifier must ...
2
votes
1answer
238 views

How can I classify images as “normal” or “lesions” using SVM?

I want to classify images as "normal" or "lesions", and then rate the images in the lesions category from mild and severe. This classification would be based on edge, color and texture features. For ...
3
votes
1answer
722 views

Nested Cross-Validation for Classification in Matlab

I am trying to tackle a classification problem with Support Vector Machine in Matlab using SVM. Using sample codes in Bioinformatics Toolbox documentation (SVM Classification with Cross Validation) ...
7
votes
4answers
395 views

Low classification accuracy, what to do next?

So, I'm a newbie in ML field and I try to do some classification. My goal is to predict the outcome of a sport event. I've gathered some historical data and now try to train a classifier. I got around ...
0
votes
0answers
128 views

Feature selection for SVM and Maximum Entropy

In text classification problems where the number of features >> number of documents, is it useful to perform feature selection with filters (e.g. Information Gain) when using Naive Bayes. However, ...
4
votes
1answer
87 views

What to do when every SVM returns “not a member”?

We have trained about 200 Linear Support Vector Machines (hand-coded in C#) belonging to our 200 categories and we use them in text categorization. Due to the shortness of the text in our training ...
1
vote
1answer
173 views

Cross validation accuracy is the same as the fraction of negative labels - what does it mean?

I have a dataset for classification (binary - 1/0) that has around 4000 samples that I use to train the model (I'm using an SVM, if that's relevant). To check whether everything is working fine, I ...
1
vote
1answer
147 views

Recommendations for MRI classification in R of large dataset (n=100, p=20000)

I am working on a magnetic-resonance imaging dataset which includes about 100 observations (= subjects) and 20000 predictors (=voxels). I would like to conduct classification in R using methods like ...
4
votes
3answers
1k views

SVM, Overfitting, curse of dimensionality

My dataset is small (120 samples), however the number of features are large varies from (1000-200,000). Although I'm doing feature selection to pick a subset of features, it might still overfit. My ...
1
vote
1answer
99 views

Possible reason for failing to build a support vector machine

I was trying to build a classifier for a set of documents using a support vector machine. I choose to build the feature space using term occurrence. While experimenting, I found the following ...
6
votes
3answers
977 views

Grid search on k-fold cross validation

I've a dataset of 120 samples in a 10-fold cross validation setting. Currently, I pick the training data of the first holdout and do a 5-fold cross-validation on it to pick the values of gamma and C ...
6
votes
1answer
190 views

The relationship between the number of support vectors and the number of features

I ran an SVM against a given data set, and made the following observation: If I change the number of features for building the classifier, the number of resulting support vectors will also be changed. ...
2
votes
1answer
206 views

How to determine whether a classifier like adaboost is weak?

I run the cross-validation experiment for a given data set, and tried two different approaches: one is based on SVM, another is based on SVM plus Adaboost. But the confusion matrix for two experiments ...
12
votes
3answers
182 views

Is building a multiclass classifier better than several binary ones?

I need to classify URLs into categories. Say I have 15 categories that I'm planning to zero down every URL to. Is a 15-way classifier better? Where I have 15 labels and generate features for each ...
5
votes
3answers
354 views

Suitable number of classes for SVM in text categorization

I'm doing text categorization with R and SVM in the package e1071. I have around 30000 text files for training, and 10000 for test. The goal is to hierarchically categorize these files. For example, I ...
0
votes
0answers
38 views

Range features with SVM

I have some data that I'd like to apply classification to: ...
0
votes
1answer
232 views

Using Adaboost with SVM for classification

I know that Adaboost tries to generate a strong classifier using a linear combination of a set of weak classifiers. However, I've read some papers suggesting Adaboost and SVMs work in harmony (even ...

1 2