Statistical Analysis Stack Exchange
Statistical Analysis Stack Exchange

Questions tagged [novelty-detection]

23 questions
30
votes
4 answers

What is one class SVM and how does it work?

I was using one class SVM, implemented in scikit-learn, for my research work. But I have no good understanding of this. Can anyone please give a simple, good explanation of one class SVM?
Nilani Algiriyage
  • 565
  • 2
  • 5
  • 11
8
votes
1 answer

Detecting changes in large number of time-series that share seasonality

I have large number of time-series that are independent of each other, but share some seasonality patterns. I need to detect anomalies/changes (increased volume, change in mean), that appear in the individual time-series. I also have some potential…
Tim
  • 108,699
  • 20
  • 212
  • 390
6
votes
3 answers

Power martingales for change detection: M goes to zero?

I'm trying to apply the power martingale framework by [Vovk et al., 2003] to change detection in unlabeled data streams, just like in [Ho and Wechsler, 2007]. The basic idea involves using a power martingale of the form $$M_n^{(\epsilon)} :=…
snikolenko
  • 61
  • 1
4
votes
2 answers

Parameter optimization in one-class SVM of LibSVM

I am new to machine learning and SVMs. I have a general question regarding the optimization of parameters in one-class SVM in libsvm in R. I found similar posts but yet not conclusive answer. Can someone tell me, do I need to optimize both c (cost)…
Thanos
  • 41
  • 1
  • 2
4
votes
1 answer

Is a one class naive bayes possible?

I have a simple question - I think. I have recently read a…
user3546025
  • 41
  • 3
2
votes
1 answer

novelty detection to prevent prediction outside training dataset

I have a training dataset composed of $d$ independent variables $\bf X$ and a dependent variable $\bf y$ for $n$ observations. I have trained a model with this $n$ observations. What I want to do now is to prevent the user of the model to predict…
AlexC
  • 21
  • 1
2
votes
1 answer

Is it better to do per-class anomaly detection on P(x, y) or P(x | y)?

(Not an expert in anomaly detection.) I'd like to experiment with per-class anomaly detection. That is, we have a feature vector $x$, and a classifier that predicts its class $\hat{y}$. I'd like to see if the combination $(x, \hat{y})$ is an…
kennysong
  • 931
  • 10
  • 18
2
votes
2 answers

How to choose a method for binary classifier based on only positive and unlabelled examples?

I need to build a binary classifier with machine learning, as I fail to manually choose a combination of features to achieve minimal fraction of false positives. What is best practice for choosing a ML method for building a binary classifier,…
wass rubleff
  • 39
  • 2
2
votes
1 answer

Best way to train one-class SVM

Let`s say I have training data which contains 10 classes and have build a classifier using this data. When applying this classifier in real life it may encounter examples not belong to the classes in the training data. I want to build a novelty…
Angry Imbecile
  • 74
  • 1
  • 6
1
vote
2 answers

How to tune parameters for novelty detection with only normal dataset

There exists multiple novelty detection methods. I'll discuss two: One-class SVM LOF Both of them have parameters. For example, the SVM has a $\nu$ parameter and if the SVM uses the RBF kernel, it has a smoothing parameter. The LOF has the…
Tristan
  • 113
  • 4
1
vote
1 answer

Is it OK to have only a single class labels in test data for prediction with one-class-svm?

I have a data which has only a single class, namely, '0'. There is no 'not 0' class. The one-class SVM model was trained on a train dataset containing only a single class '0'. I do not unnecessarily want to find a random 'not 0' class to include in…
Hello World
  • 163
  • 1
  • 2
  • 10
1
vote
0 answers

Weighting loss function using Voronoi-tessellation of response space

Let's say you have some Real-valued features $\mathbf{X}$ and Real-valued univariate responses $\mathbf{y}$. We want to fit a regression model to this data: $$\mathbf{y} = f\left(\mathbf{X};\beta\right) + \mathbf{\varepsilon}$$ Where fitting is done…
user145807
1
vote
2 answers

Novelty and Outlier Detection for Multi-label Data

I met a problem of using novelty and outlier detection for my multi-label data. For example, I have got some training data that is not polluted by outliers. However, the training data are with multi-labels, let's say the data is with 150 cases with…
Samo Jerom
  • 1,439
  • 2
  • 19
  • 31
1
vote
1 answer

Clustering based anomaly detection

I'm trying to implement anomaly detection based on clustering. I'm hopping for confirmation of my approach, and I'm exposing my idea, being aware that I could have miss something in my analysis, so any suggestions would also be very appreciated. I'm…
Kobe-Wan Kenobi
  • 2,437
  • 3
  • 20
  • 33
0
votes
0 answers

What is the best way to evaluate a novelty detection model?

How can I evaluate a novelty detection model (e.g.: Local Outlier Factor) that is trained to identify previously unseen data points that do not belong to the training dataset? My issue is that, unlike in supervised learning (for example), we do not…
Nht_e0
  • 21
  • 3
1
2