SherbyNet - /page/wikibook/entry.mu

<!DOCTYPE html>

<title>One-class classification</title>

</head>

One-class classification

</h1>

</header>

In <a href="Machine_learning" title="Machine learning">machine learning</a>, one-class classification (OCC), also known as unary classification or class-modelling, tries to identify objects of a specific class amongst all objects, by primarily learning from a <a href="Training_set" class="mw-redirect" title="Training set">training set</a> containing only the objects of that class,<a href="#cite_note-1">[1]</a> although there exist variants of one-class classifiers where counter-examples are used to further refine the classification boundary. This is different from and more difficult than the traditional <a href="Classification_(machine_learning)" class="mw-redirect" title="Classification (machine learning)">classification</a> problem, which tries to distinguish between two or more classes with the training set containing objects from all the classes. Examples include the monitoring of helicopter gearboxes,<a href="#cite_note-2">[2]</a><a href="#cite_note-3">[3]</a><a href="#cite_note-4">[4]</a> motor failure prediction,<a href="#cite_note-5">[5]</a> or the operational status of a nuclear plant as 'normal':<a href="#cite_note-:1-6">[6]</a> In this scenario, there are few, if any, examples of catastrophic system states; only the statistics of normal operation are known.

While many of the above approaches focus on the case of removing a small number of outliers or anomalies, one can also learn the other extreme, where the single class covers a small coherent subset of the data, using an <a href="Information_bottleneck_method" title="Information bottleneck method">information bottleneck</a> approach.<a href="#cite_note-7">[7]</a>

<div class="mw-heading mw-heading2"><h2 id="Overview">Overview</h2></div>

The term one-class classification (OCC) was coined by Moya & Hush (1996)<a href="#cite_note-8">[8]</a> and many applications can be found in scientific literature, for example <a href="Outlier_detection" class="mw-redirect" title="Outlier detection">outlier detection</a>, <a href="Anomaly_detection" title="Anomaly detection">anomaly detection</a>, <a href="Novelty_detection" title="Novelty detection">novelty detection</a>. A feature of OCC is that it uses only sample points from the assigned class, so that a representative sampling is not strictly required for non-target classes.<a href="#cite_note-9">[9]</a>

<div class="mw-heading mw-heading2"><h2 id="Introduction">Introduction</h2></div>

SVM based one-class classification (OCC) relies on identifying the smallest hypersphere (with radius r, and center c) consisting of all the data points.<a href="#cite_note-10">[10]</a> This method is called Support Vector Data Description (SVDD). Formally, the problem can be defined in the following constrained optimization form,

</mrow>

</munder>

<msup>

</mrow>

</msup>

<mtext> subject to, </mtext>

</mrow>

</mrow>

</mrow>

<msub>

</mrow>

</msub>

</mrow>

<msup>

</mrow>

</mrow>

</msup>

<msup>

</mrow>

</msup>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \min _{r,c}r^{2}{\text{ subject to, }}||\Phi (x_{i})-c||^{2}\leq r^{2}\;\;\forall i=1,2,...,n}</annotation>

</semantics>

</math>

However, the above formulation is highly restrictive, and is sensitive to the presence of outliers. Therefore, a flexible formulation, that allow for the presence of outliers is formulated as shown below,

</mrow>

</munder>

<msup>

</mrow>

</msup>

<mfrac>

<mrow>

</mrow>

</mfrac>

</mrow>

</mrow>

</mrow>

</munderover>

<msub>

</mrow>

</msub>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \min _{r,c,\zeta }r^{2}+{\frac {1}{\nu n}}\sum _{i=1}^{n}\zeta _{i}}</annotation>

</semantics>

</math><img src="./a10142ffab766dc5d55a21bd37b7f9d6335bbabd.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -3.005ex; width:18.616ex; height:6.843ex;" alt="{\displaystyle \min _{r,c,\zeta }r^{2}+{\frac {1}{\nu n}}\sum _{i=1}^{n}\zeta _{i}}" loading="lazy">

<mtext>subject to, </mtext>

</mrow>

</mrow>

</mrow>

<msub>

</mrow>

</msub>

</mrow>

<msup>

</mrow>

</mrow>

</msup>

<msup>

</mrow>

</msup>

<msub>

</mrow>

</msub>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle {\text{subject to, }}||\Phi (x_{i})-c||^{2}\leq r^{2}+\zeta _{i}\;\;\forall i=1,2,...,n}</annotation>

</semantics>

</math><img src="./a0a81bd39c0f908b4905e755282a932400b0edfd.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:50.43ex; height:3.343ex;" alt="{\displaystyle {\text{subject to, }}||\Phi (x_{i})-c||^{2}\leq r^{2}+\zeta _{i}\;\;\forall i=1,2,...,n}" loading="lazy">

From the <a href="Karush%E2%80%93Kuhn%E2%80%93Tucker_conditions" title="Karush–Kuhn–Tucker conditions">Karush–Kuhn–Tucker conditions</a> for optimality, we get

</mrow>

</mrow>

</munderover>

<msub>

</mrow>

</msub>

<msub>

</mrow>

</msub>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle c=\sum _{i=1}^{n}\alpha _{i}\Phi (x_{i}),}</annotation>

</semantics>

</math><img src="./c91f44db6e03045cd5d5088b88b93170c29d1f8a.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -3.005ex; width:16.398ex; height:6.843ex;" alt="{\displaystyle c=\sum _{i=1}^{n}\alpha _{i}\Phi (x_{i}),}" loading="lazy">

where the <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \alpha _{i}}">

<msub>

</mrow>

</msub>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \alpha _{i}}</annotation>

</semantics>

</math><img src="./3b1fb627423abe4988b7ed88d4920bf1ec074790.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.671ex; width:2.287ex; height:2.009ex;" alt="{\displaystyle \alpha _{i}}" loading="lazy">'s are the solution to the following optimization problem:

</mrow>

</munder>

</mrow>

</mrow>

</munderover>

<msub>

</mrow>

</msub>

<msub>

</mrow>

</msub>

<msub>

</mrow>

</msub>

</mrow>

</mrow>

</munderover>

<msub>

</mrow>

</msub>

<msub>

</mrow>

</msub>

<msub>

</mrow>

</msub>

<msub>

</mrow>

</msub>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \max _{\alpha }\sum _{i=1}^{n}\alpha _{i}\kappa (x_{i},x_{i})-\sum _{i,j=1}^{n}\alpha _{i}\alpha _{j}\kappa (x_{i},x_{j})}</annotation>

</semantics>

</math><img src="./6ade3eb61c705b7ce0c67ef20c7bfb486888ca3c.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -3.338ex; width:39.449ex; height:7.176ex;" alt="{\displaystyle \max _{\alpha }\sum _{i=1}^{n}\alpha _{i}\kappa (x_{i},x_{i})-\sum _{i,j=1}^{n}\alpha _{i}\alpha _{j}\kappa (x_{i},x_{j})}" loading="lazy">

subject to, <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \sum _{i=1}^{n}\alpha _{i}=1{\text{ and }}0\leq \alpha _{i}\leq {\frac {1}{\nu n}}{\text{for all }}i=1,2,...,n.}">

</mrow>

</mrow>

</munderover>

<msub>

</mrow>

</msub>

</mrow>

<msub>

</mrow>

</msub>

<mfrac>

<mrow>

</mrow>

</mfrac>

</mrow>

</mrow>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \sum _{i=1}^{n}\alpha _{i}=1{\text{ and }}0\leq \alpha _{i}\leq {\frac {1}{\nu n}}{\text{for all }}i=1,2,...,n.}</annotation>

</semantics>

</math><img src="./8e80a2ebd9e31899045870be21c8a2176cd221f9.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -3.005ex; width:49.183ex; height:6.843ex;" alt="{\displaystyle \sum _{i=1}^{n}\alpha _{i}=1{\text{ and }}0\leq \alpha _{i}\leq {\frac {1}{\nu n}}{\text{for all }}i=1,2,...,n.}" loading="lazy">

The introduction of kernel function provide additional flexibility to the One-class <a href="Support-vector_machine" class="mw-redirect" title="Support-vector machine">SVM</a> (OSVM) algorithm.<a href="#cite_note-11">[11]</a>

<div class="mw-heading mw-heading3"><h3 id="PU_(Positive_Unlabeled)_learning">PU (Positive Unlabeled) learning</h3></div>

A similar problem is PU learning, in which a <a href="Binary_classification" title="Binary classification">binary classifier</a> is constructed by <a href="Semi-supervised_learning" class="mw-redirect" title="Semi-supervised learning">semi-supervised learning</a> from only positive and unlabeled sample points.<a href="#cite_note-12">[12]</a>

In PU learning, two sets of examples are assumed to be available for training: the positive set <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle P}">

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle P}</annotation>

</semantics>

</math><img src="./b4dc73bf40314945ff376bd363916a738548d40a.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.745ex; height:2.176ex;" alt="{\displaystyle P}" loading="lazy"> and a mixed set <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle U}">

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle U}</annotation>

</semantics>

</math><img src="./458a728f53b9a0274f059cd695e067c430956025.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.783ex; height:2.176ex;" alt="{\displaystyle U}" loading="lazy">, which is assumed to contain both positive and negative samples, but without these being labeled as such. This contrasts with other forms of semisupervised learning, where it is assumed that a labeled set containing examples of both classes is available in addition to unlabeled samples. A variety of techniques exist to adapt <a href="Supervised_learning" title="Supervised learning">supervised</a> classifiers to the PU learning setting, including variants of the <a href="Expectation%E2%80%93maximization_algorithm" title="Expectation–maximization algorithm">EM algorithm</a>. PU learning has been successfully applied to <a href="Text_classification" class="mw-redirect" title="Text classification">text</a>,<a href="#cite_note-13">[13]</a><a href="#cite_note-14">[14]</a><a href="#cite_note-15">[15]</a> time series,<a href="#cite_note-16">[16]</a> <a href="Bioinformatics" title="Bioinformatics">bioinformatics</a> tasks,<a href="#cite_note-17">[17]</a><a href="#cite_note-18">[18]</a> and <a href="Remote_sensing" title="Remote sensing">remote sensing</a> data.<a href="#cite_note-19">[19]</a>

<div class="mw-heading mw-heading2"><h2 id="Approaches">Approaches</h2></div>

Several approaches have been proposed to solve one-class classification (OCC). The approaches can be distinguished into three main categories, density estimation, boundary methods, and reconstruction methods.<a href="#cite_note-:1-6">[6]</a>

<div class="mw-heading mw-heading3"><h3 id="Density_estimation_methods">Density estimation methods</h3></div>

Density estimation methods rely on estimating the density of the data points, and set the threshold. These methods rely on assuming distributions, such as Gaussian, or a <a href="Poisson_distribution" title="Poisson distribution">Poisson distribution</a>. Following which discordancy tests can be used to test the new objects. These methods are robust to scale variance.

Gaussian model<a href="#cite_note-20">[20]</a> is one of the simplest methods to create one-class classifiers. Due to Central Limit Theorem (CLT),<a href="#cite_note-21">[21]</a> these methods work best when large number of samples are present, and they are perturbed by small independent error values. The probability distribution for a d-dimensional object is given by:

<msub>

</mrow>

</msub>

<mfrac>

<mrow>

<msup>

<mfrac>

</mfrac>

</mrow>

</msup>

</mrow>

<msup>

</mrow>

<mfrac>

</mfrac>

</mrow>

</msup>

</mrow>

</mfrac>

</mrow>

<mfrac>

</mfrac>

</mrow>

<msup>

</mrow>

</msup>

<msup>

</mrow>

</msup>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle p_{\mathcal {N}}(x;\mu ;\Sigma )={\frac {1}{(2\pi )^{\frac {d}{2}}|\Sigma |^{\frac {1}{2}}}}\exp\{-{\frac {1}{2}}(z-\mu )^{T}\Sigma ^{-1}(z-\mu )\}}</annotation>

</semantics>

</math><img src="./e7b027b2557c1809609136b5fdce572fe8ae4dcb.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -3.838ex; margin-left: -0.089ex; width:55.849ex; height:7.176ex;" alt="{\displaystyle p_{\mathcal {N}}(x;\mu ;\Sigma )={\frac {1}{(2\pi )^{\frac {d}{2}}|\Sigma |^{\frac {1}{2}}}}\exp\{-{\frac {1}{2}}(z-\mu )^{T}\Sigma ^{-1}(z-\mu )\}}" loading="lazy">

Where, <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \mu }">

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \mu }</annotation>

</semantics>

</math><img src="./9fd47b2a39f7a7856952afec1f1db72c67af6161.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:1.402ex; height:2.176ex;" alt="{\displaystyle \mu }" loading="lazy"> is the mean and <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \Sigma }">

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \Sigma }</annotation>

</semantics>

</math><img src="./9e1f558f53cda207614abdf90162266c70bc5c1e.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.678ex; height:2.176ex;" alt="{\displaystyle \Sigma }" loading="lazy"> is the covariance matrix. Computing the inverse of covariance matrix (<math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \Sigma ^{-1}}">

<msup>

</mrow>

</msup>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \Sigma ^{-1}}</annotation>

</semantics>

</math><img src="./b58f4501b032d9f46f33aa4667191ab2b2998b50.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:4.011ex; height:2.676ex;" alt="{\displaystyle \Sigma ^{-1}}" loading="lazy">) is the costliest operation, and in the cases where the data is not scaled properly, or data has singular directions pseudo-inverse <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \Sigma ^{+}}">

<msup>

</mrow>

</msup>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \Sigma ^{+}}</annotation>

</semantics>

</math><img src="./c78b913a3f87bba9d7bbba8696a81f05633dee94.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:3.189ex; height:2.509ex;" alt="{\displaystyle \Sigma ^{+}}" loading="lazy"> is used to approximate the inverse, and is calculated as <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \Sigma ^{T}(\Sigma \Sigma ^{T})^{-1}}">

<msup>

</mrow>

</msup>

<msup>

</mrow>

</msup>

<msup>

</mrow>

</msup>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \Sigma ^{T}(\Sigma \Sigma ^{T})^{-1}}</annotation>

</semantics>

</math><img src="./79ddd231fa49d5668fb9632ee4da39d4463a20fd.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:11.955ex; height:3.176ex;" alt="{\displaystyle \Sigma ^{T}(\Sigma \Sigma ^{T})^{-1}}" loading="lazy">.<a href="#cite_note-22">[22]</a>

<div class="mw-heading mw-heading3"><h3 id="Boundary_methods">Boundary methods</h3></div>

Boundary methods focus on setting boundaries around a few set of points, called target points. These methods attempt to optimize the volume. Boundary methods rely on distances, and hence are not robust to scale variance. K-centers method, NN-d, and SVDD are some of the key examples.

K-centers

In K-center algorithm,<a href="#cite_note-23">[23]</a> <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle k}">

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle k}</annotation>

</semantics>

</math><img src="./c3c9a2c7b599b37105512c5d570edc034056dd40.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.211ex; height:2.176ex;" alt="{\displaystyle k}" loading="lazy"> small balls with equal radius are placed to minimize the maximum distance of all minimum distances between training objects and the centers. Formally, the following error is minimized,

<msub>

</mrow>

</msub>

</mrow>

</munder>

</mrow>

</munder>

</mrow>

</mrow>

<msub>

</mrow>

</msub>

<msub>

</mrow>

</msub>

</mrow>

<msup>

</mrow>

</mrow>

</msup>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle \varepsilon _{k-center}=\max _{i}(\min _{k}||x_{i}-\mu _{k}||^{2})}</annotation>

</semantics>

</math><img src="./8ee8465804b3d635dd4a13c6d0b2225594526d08.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -2.005ex; width:32.614ex; height:4.509ex;" alt="{\displaystyle \varepsilon _{k-center}=\max _{i}(\min _{k}||x_{i}-\mu _{k}||^{2})}" loading="lazy">

The algorithm uses forward search method with random initialization, where the radius is determined by the maximum distance of the object, any given ball should capture. After the centers are determined, for any given test object <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle z}">

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle z}</annotation>

</semantics>

</math><img src="./bf368e72c009decd9b6686ee84a375632e11de98.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.088ex; height:1.676ex;" alt="{\displaystyle z}" loading="lazy"> the distance can be calculated as,

<msub>

</mrow>

</msub>

</mrow>

</munder>

</mrow>

</mrow>

<msub>

</mrow>

</msub>

</mrow>

<msup>

</mrow>

</mrow>

</msup>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle d_{k-centr}(z)=\min _{k}||z-\mu _{k}||^{2}}</annotation>

</semantics>

</math><img src="./4620653c71cd878dbcdf198c06bec28800291cc1.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -2.005ex; width:27.694ex; height:4.509ex;" alt="{\displaystyle d_{k-centr}(z)=\min _{k}||z-\mu _{k}||^{2}}" loading="lazy">

<div class="mw-heading mw-heading3"><h3 id="Reconstruction_methods">Reconstruction methods</h3></div>

Reconstruction methods use prior knowledge and generating process to build a generating model that best fits the data. New objects can be described in terms of a state of the generating model. Some examples of reconstruction methods for OCC are, k-means clustering, learning vector quantization, self-organizing maps, etc.

<div class="mw-heading mw-heading2"><h2 id="Applications">Applications</h2></div>

<div class="mw-heading mw-heading3"><h3 id="Document_classification">Document classification</h3></div>

The basic Support Vector Machine (SVM) paradigm is trained using both positive and negative examples, however studies have shown there are many valid reasons for using only positive examples. When the SVM algorithm is modified to only use positive examples, the process is considered one-class classification. One situation where this type of classification might prove useful to the SVM paradigm is in trying to identify a web browser's sites of interest based only off of the user's browsing history.

<div class="mw-heading mw-heading3"><h3 id="Biomedical_studies">Biomedical studies</h3></div>

One-class classification can be particularly useful in biomedical studies where often data from other classes can be difficult or impossible to obtain. In studying biomedical data it can be difficult and/or expensive to obtain the set of labeled data from the second class that would be necessary to perform a two-class classification. A study from The Scientific World Journal found that the typicality approach is the most useful in analysing biomedical data because it can be applied to any type of dataset (continuous, discrete, or nominal).<a href="#cite_note-:0-24">[24]</a> The typicality approach is based on the clustering of data by examining data and placing it into new or existing clusters.<a href="#cite_note-25">[25]</a> To apply typicality to one-class classification for biomedical studies, each new observation, <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle y_{0}}">

<msub>

</mrow>

</msub>

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle y_{0}}</annotation>

</semantics>

</math><img src="./6d943dbbb0b56ca750c4d62c5b54b4ae29a773da.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.671ex; width:2.193ex; height:2.009ex;" alt="{\displaystyle y_{0}}" loading="lazy">, is compared to the target class, <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle C}">

</mstyle>

</mrow>

<annotation encoding="application/x-tex">{\displaystyle C}</annotation>

</semantics>

</math><img src="./4fc55753007cd3c18576f7933f6f089196732029.svg" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.766ex; height:2.176ex;" alt="{\displaystyle C}" loading="lazy">, and identified as an outlier or a member of the target class.<a href="#cite_note-:0-24">[24]</a>

<div class="mw-heading mw-heading3"><h3 id="Unsupervised_Concept_Drift_Detection">Unsupervised Concept Drift Detection</h3></div>

One-class classification has similarities with unsupervised concept drift detection, where both aim to identify whether the unseen data share similar characteristics to the initial data. A concept is referred to as the fixed probability distribution which data is drawn from. In unsupervised concept drift detection, the goal is to detect if the data distribution changes without utilizing class labels. In one-class classification, the flow of data is not important. Unseen data is classified as typical or outlier depending on its characteristics, whether it is from the initial concept or not. However, unsupervised drift detection monitors the flow of data, and signals a drift if there is a significant amount of change or anomalies. Unsupervised concept drift detection can be identified as the continuous form of one-class classification.<a href="#cite_note-26">[26]</a> One-class classifiers are used for detecting concept drifts.<a href="#cite_note-27">[27]</a>

<ul><li><a href="Multiclass_classification" title="Multiclass classification">Multiclass classification</a></li>

<li><a href="Anomaly_detection" title="Anomaly detection">Anomaly detection</a></li>

<li><a href="Supervised_learning" title="Supervised learning">Supervised learning</a></li></ul>

<div class="mw-heading mw-heading2"><h2 id="References">References</h2></div>

/* start https://en.wikipedia.org/ */

.mw-parser-output .reflist{margin-bottom:0.5em;list-style-type:decimal}@media screen{.mw-parser-output .reflist{font-size:90%}}.mw-parser-output .reflist .references{font-size:100%;margin-bottom:0;list-style-type:inherit}.mw-parser-output .reflist-columns-2{column-width:30em}.mw-parser-output .reflist-columns-3{column-width:25em}.mw-parser-output .reflist-columns{margin-top:0.3em}.mw-parser-output .reflist-columns ol{margin-top:0}.mw-parser-output .reflist-columns li{page-break-inside:avoid;break-inside:avoid-column}.mw-parser-output .reflist-upper-alpha{list-style-type:upper-alpha}.mw-parser-output .reflist-upper-roman{list-style-type:upper-roman}.mw-parser-output .reflist-lower-alpha{list-style-type:lower-alpha}.mw-parser-output .reflist-lower-greek{list-style-type:lower-greek}.mw-parser-output .reflist-lower-roman{list-style-type:lower-roman}

/* end https://en.wikipedia.org/ */

</style><div class="reflist reflist-columns references-column-width" style="column-width: 30em;">

/* start https://en.wikipedia.org/ */

.mw-parser-output cite.citation{font-style:inherit;word-wrap:break-word}.mw-parser-output .citation q{quotes:"\"""\"""'""'"}.mw-parser-output .citation:target{background-color:rgba(0,127,255,0.133)}.mw-parser-output .id-lock-free.id-lock-free a{background:url("./mw/Lock-green.svg")right 0.1em center/9px no-repeat}.mw-parser-output .id-lock-limited.id-lock-limited a,.mw-parser-output .id-lock-registration.id-lock-registration a{background:url("./mw/Lock-gray-alt-2.svg")right 0.1em center/9px no-repeat}.mw-parser-output .id-lock-subscription.id-lock-subscription a{background:url("./mw/Lock-red-alt-2.svg")right 0.1em center/9px no-repeat}.mw-parser-output .cs1-ws-icon a{background:url("./mw/Wikisource-logo.svg")right 0.1em center/12px no-repeat}body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .id-lock-free a,body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .id-lock-limited a,body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .id-lock-registration a,body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .id-lock-subscription a,body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .cs1-ws-icon a{background-size:contain;padding:0 1em 0 0}.mw-parser-output .cs1-code{color:inherit;background:inherit;border:none;padding:inherit}.mw-parser-output .cs1-hidden-error{display:none;color:var(--color-error,#d33)}.mw-parser-output .cs1-visible-error{color:var(--color-error,#d33)}.mw-parser-output .cs1-maint{display:none;color:#085;margin-left:0.3em}.mw-parser-output .cs1-kern-left{padding-left:0.2em}.mw-parser-output .cs1-kern-right{padding-right:0.2em}.mw-parser-output .citation .mw-selflink{font-weight:inherit}@media screen{.mw-parser-output .cs1-format{font-size:95%}html.skin-theme-clientpref-night .mw-parser-output .cs1-maint{color:#18911f}}@media screen and (prefers-color-scheme:dark){html.skin-theme-clientpref-os .mw-parser-output .cs1-maint{color:#18911f}}

/* end https://en.wikipedia.org/ */

</style><cite id="CITEREFOliveri2017" class="citation journal cs1">Oliveri P (August 2017). "Class-modelling in food analytical chemistry: Development, sampling, optimisation and validation issues - A tutorial". Analytica Chimica Acta. 982: 9–19. <a href="Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2017AcAC..982....9O">2017AcAC..982....9O</a>. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1016%2Fj.aca.2017.05.013">10.1016/j.aca.2017.05.013</a>. <a href="Hdl_(identifier)" class="mw-redirect" title="Hdl (identifier)">hdl</a>:<a rel="nofollow" class="external text" href="https://hdl.handle.net/11567%2F881059">11567/881059</a>. <a href="PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/28734370">28734370</a>.</cite>

</li>

<li id="cite_note-2"><a href="#cite_ref-2">^</a> <cite id="CITEREFJapkowiczMyersGluck1995" class="citation citeseerx cs1">Japkowicz N, Myers C, Gluck M (1995). "A Novelty Detection Approach to Classification". pp. 518–523. <a href="CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.40.3663">10.1.1.40.3663</a>.</cite>

</li>

<li id="cite_note-3"><a href="#cite_ref-3">^</a> <cite id="CITEREFJapkowicz1999" class="citation thesis cs1">Japkowicz N (1999). <a rel="nofollow" class="external text" href="https://rucore.libraries.rutgers.edu/rutgers-lib/59019/">Concept-Learning in the Absence of Counter-Examples:An Autoassociation-Based Approach to Classification</a> (Thesis). Rutgers University.</cite>

</li>

<li id="cite_note-4"><a href="#cite_ref-4">^</a> <cite id="CITEREFJapkowicz2001" class="citation journal cs1"><a href="Nathalie_Japkowicz" title="Nathalie Japkowicz">Japkowicz N</a> (2001). <a rel="nofollow" class="external text" href="https://link.springer.com/content/pdf/10.1023/A:1007660820062.pdf">"Supervised Versus Unsupervised Binary-Learning by Feedforward Neural Networks"</a> (PDF). Machine Learning. 42: 97–122. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1023%2FA%3A1007660820062">10.1023/A:1007660820062</a>. <a href="S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:7298189">7298189</a>.</cite>

</li>

<li id="cite_note-5"><a href="#cite_ref-5">^</a> <cite id="CITEREFPetscheMarcantonioDarkenHanson1996" class="citation news cs1">Petsche T, Marcantonio A, Darken C, Hanson S, Kuhn G, Santoso I (1996). <a rel="nofollow" class="external text" href="http://papers.nips.cc/paper/1077-a-neural-network-autoassociator-for-induction-motor-failure-prediction.pdf">"A Neural Network Autoassociator for Induction Motor Failure Prediction"</a> (PDF). NIPS.</cite>

</li>

<li id="cite_note-:1-6">^ <a href="#cite_ref-:1_6-0">a</a> <a href="#cite_ref-:1_6-1">b</a> <cite id="CITEREFTax2001" class="citation thesis cs1">Tax D (2001). <a rel="nofollow" class="external text" href="https://rduin.nl/papers/thesis_01_tax.pdf">One-class classification: Concept-learning in the absence of counter-examples</a> (PDF) (Ph.D. thesis). The Netherlands: University of Delft.</cite>

</li>

<li id="cite_note-7"><a href="#cite_ref-7">^</a> <cite id="CITEREFCrammer2004" class="citation book cs1">Crammer, Koby (2004). <a rel="nofollow" class="external text" href="https://dl.acm.org/citation.cfm?id=1015399">"A needle in a haystack"</a>. Twenty-first international conference on Machine learning - ICML '04. p. 26. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1145%2F1015330.1015399">10.1145/1015330.1015399</a>. <a href="ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <bdi>978-1-58113-838-2</bdi>. <a href="S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:8736254">8736254</a>.</cite>

</li>

<li id="cite_note-8"><a href="#cite_ref-8">^</a> <cite id="CITEREFMoyaHush1996" class="citation journal cs1">Moya, M.; Hush, D. (1996). "Network constraints and multi- objective optimization for one-class classification". Neural Networks. 9 (3): 463–474. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1016%2F0893-6080%2895%2900120-4">10.1016/0893-6080(95)00120-4</a>.</cite>

</li>

<li id="cite_note-9"><a href="#cite_ref-9">^</a> <cite id="CITEREFRodionovaOliveriPomerantsev2016" class="citation journal cs1">Rodionova OY, Oliveri P, Pomerantsev AL (2016-12-15). "Rigorous and compliant approaches to one-class classification". Chemometrics and Intelligent Laboratory Systems. 159: 89–96. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1016%2Fj.chemolab.2016.10.002">10.1016/j.chemolab.2016.10.002</a>. <a href="Hdl_(identifier)" class="mw-redirect" title="Hdl (identifier)">hdl</a>:<a rel="nofollow" class="external text" href="https://hdl.handle.net/11567%2F864539">11567/864539</a>.</cite>

</li>

<li id="cite_note-10"><a href="#cite_ref-10">^</a> <cite id="CITEREFZinebHoneineRichard2012" class="citation journal cs1">Zineb, Noumir; Honeine, Paul; Richard, Cedue (2012). "On simple one-class classification methods". IEEE International Symposium on Information Theory Proceedings. IEEE, 2012.</cite>

</li>

<li id="cite_note-11"><a href="#cite_ref-11">^</a> <cite id="CITEREFKhanMadden2010" class="citation book cs1">Khan, Shehroz S.; Madden, Michael G. (2010). "A Survey of Recent Trends in One Class Classification". In Coyle, Lorcan; Freyne, Jill (eds.). Artificial Intelligence and Cognitive Science. Lecture Notes in Computer Science. Vol. 6206. Springer Berlin Heidelberg. pp. 188–197. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2F978-3-642-17080-5_21">10.1007/978-3-642-17080-5_21</a>. <a href="Hdl_(identifier)" class="mw-redirect" title="Hdl (identifier)">hdl</a>:<a rel="nofollow" class="external text" href="https://hdl.handle.net/10379%2F1472">10379/1472</a>. <a href="ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <bdi>978-3-642-17080-5</bdi>. <a href="S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:36784649">36784649</a>.</cite>

</li>

<li id="cite_note-12"><a href="#cite_ref-12">^</a> <cite id="CITEREFLiu2007" class="citation book cs1">Liu, Bing (2007). Web Data Mining. Springer. pp. 165–178.</cite>

</li>

<li id="cite_note-13"><a href="#cite_ref-13">^</a> <cite id="CITEREFBing_LiuWee_Sun_LeePhilip_S._YuXiao-Li_Li2002" class="citation conference cs1">Bing Liu; Wee Sun Lee; <a href="Philip_S._Yu" title="Philip S. Yu">Philip S. Yu</a> & Xiao-Li Li (2002). Partially supervised classification of text documents. ICML. pp. 8–12.</cite>

</li>

<li id="cite_note-14"><a href="#cite_ref-14">^</a> <cite id="CITEREFHwanjo_YuJiawei_HanKevin_Chen-Chuan_Chang2002" class="citation conference cs1">Hwanjo Yu; Jiawei Han; Kevin Chen-Chuan Chang (2002). PEBL: positive example based learning for web page classification using SVM. ACM SIGKDD.</cite>

</li>

<li id="cite_note-15"><a href="#cite_ref-15">^</a> <cite id="CITEREFXiao-Li_LiBing_Liu2003" class="citation conference cs1">Xiao-Li Li & Bing Liu (2003). Learning to classify text using positive and unlabeled data. IJCAI.</cite>

</li>

<li id="cite_note-16"><a href="#cite_ref-16">^</a> <cite id="CITEREFMinh_Nhut_NguyenXiao-Li_LiSee-Kiong_Ng2011" class="citation conference cs1">Minh Nhut Nguyen; Xiao-Li Li & See-Kiong Ng (2011). Positive Unlabeled Learning for Time Series Classification. IJCAI.</cite>

</li>

<li id="cite_note-17"><a href="#cite_ref-17">^</a> <cite id="CITEREFPeng_YangXiao-Li_LiJian-Ping_MeiChee-Keong_Kwoh2012" class="citation conference cs1">Peng Yang; Xiao-Li Li; Jian-Ping Mei; Chee-Keong Kwoh & See-Kiong Ng (2012). Positive-Unlabeled Learning for Disease Gene Identification. Bioinformatics, Vol 28(20).</cite>

</li>

<li id="cite_note-18"><a href="#cite_ref-18">^</a> <cite id="CITEREFBugnon,_L._A.Yones,_C.Milone,_D._H.Stegmayer,_G.2020" class="citation journal cs1">Bugnon, L. A.; Yones, C.; Milone, D. H. & Stegmayer, G. (2020). "Genome-wide discovery of pre-miRNAs: comparison of recent approaches based on machine learning". Oxford Bioinformatics. 22 (3). <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1093%2Fbib%2Fbbaa184">10.1093/bib/bbaa184</a>. <a href="PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/32814347">32814347</a>.</cite>

</li>

<li id="cite_note-19"><a href="#cite_ref-19">^</a> <cite id="CITEREFLiGuoElkan2011" class="citation journal cs1">Li, W.; Guo, Q.; Elkan, C. (February 2011). "A Positive and Unlabeled Learning Algorithm for One-Class Classification of Remote-Sensing Data". IEEE Transactions on Geoscience and Remote Sensing. 49 (2): 717–725. <a href="Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2011ITGRS..49..717L">2011ITGRS..49..717L</a>. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FTGRS.2010.2058578">10.1109/TGRS.2010.2058578</a>. <a href="ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/0196-2892">0196-2892</a>. <a href="S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:267120">267120</a>.</cite>

</li>

<li id="cite_note-20"><a href="#cite_ref-20">^</a> <cite id="CITEREFBishopBishop1995" class="citation book cs1">Bishop, Christopher M.; Bishop, Professor of Neural Computing Christopher M. (1995-11-23). <a rel="nofollow" class="external text" href="https://books.google.com/books?id=T0S0BgAAQBAJ&q=neural+networks+for+pattern+recognition&pg=PP1">Neural Networks for Pattern Recognition</a>. Clarendon Press. <a href="ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <bdi>978-0-19-853864-6</bdi>.</cite>

</li>

<li id="cite_note-21"><a href="#cite_ref-21">^</a> <cite id="CITEREFUllman2017" class="citation book cs1">Ullman, Neil R (2017-01-01). <a rel="nofollow" class="external text" href="http://archives.umc.edu.dz/handle/123456789/116529">Elementary statistics</a>.</cite>

</li>

<li id="cite_note-22"><a href="#cite_ref-22">^</a> <cite class="citation web cs1"><a rel="nofollow" class="external text" href="http://bookstore.siam.org/wc02/">"Introduction to Applied Mathematics"</a>. SIAM Bookstore. Retrieved 2019-04-29.</cite>

</li>

<li id="cite_note-23"><a href="#cite_ref-23">^</a> <cite id="CITEREFYpmaDuin1998" class="citation book cs1">Ypma, Alexander; Duin, Robert P. W. (1998). "Support objects for domain approximation". In Niklasson, Lars; Bodén, Mikael; Ziemke, Tom (eds.). Icann 98. Perspectives in Neural Computing. Springer London. pp. 719–724. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2F978-1-4471-1599-1_110">10.1007/978-1-4471-1599-1_110</a>. <a href="ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <bdi>978-1-4471-1599-1</bdi>.</cite>

</li>

<li id="cite_note-:0-24">^ <a href="#cite_ref-:0_24-0">a</a> <a href="#cite_ref-:0_24-1">b</a> <cite id="CITEREFIrigoienSierraArenas2014" class="citation journal cs1">Irigoien I, Sierra B, Arenas C (2014). <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3980920">"Towards application of one-class classification methods to medical data"</a>. TheScientificWorldJournal. 2014: 730712. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1155%2F2014%2F730712">10.1155/2014/730712</a>. <a href="PMC_(identifier)" class="mw-redirect" title="PMC (identifier)">PMC</a> <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3980920">3980920</a>. <a href="PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/24778600">24778600</a>.</cite>

</li>

<li id="cite_note-25"><a href="#cite_ref-25">^</a> <cite id="CITEREFIrigoienArenas2008" class="citation journal cs1">Irigoien I, Arenas C (July 2008). "INCA: new statistic for estimating the number of clusters and identifying atypical units". Statistics in Medicine. 27 (15): 2948–73. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1002%2Fsim.3143">10.1002/sim.3143</a>. <a href="PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/18050154">18050154</a>. <a href="S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:24791212">24791212</a>.</cite>

</li>

<li id="cite_note-26"><a href="#cite_ref-26">^</a> <cite id="CITEREFGözüaçıkCan2020" class="citation journal cs1">Gözüaçık, Ömer; Can, Fazli (November 2020). "Concept learning using one-class classifiers for implicit drift detection in evolving data streams". Artificial Intelligence Review. 54 (5): 3725–3747. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2Fs10462-020-09939-x">10.1007/s10462-020-09939-x</a>. <a href="Hdl_(identifier)" class="mw-redirect" title="Hdl (identifier)">hdl</a>:<a rel="nofollow" class="external text" href="https://hdl.handle.net/11693%2F77042">11693/77042</a>. <a href="S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:229506136">229506136</a>.</cite>

</li>

<li id="cite_note-27"><a href="#cite_ref-27">^</a> <cite id="CITEREFKrawczykWoźniak2015" class="citation journal cs1">Krawczyk, Bartosz; Woźniak, Michał (2015). <a rel="nofollow" class="external text" href="https://doi.org/10.1007%2Fs00500-014-1492-5">"One-class classifiers with incremental learning and forgetting for data streams with concept drift"</a>. Soft Computing. 19 (12): 3387–3400. <a href="Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2Fs00500-014-1492-5">10.1007/s00500-014-1492-5</a>. <a href="S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:207011971">207011971</a>.</cite>

</li>

</ol></div></div><div><div class="zim-footer">

This article is issued from <a class="external text" title="Last edited on 2025-04-25" href="https://en.wikipedia.org/wiki/?title=One-class_classification&oldid=1287313198">Wikipedia</a>. The text is available under <a class="external text" href="https://creativecommons.org/licenses/by-sa/4.0/deed.en">Creative Commons Attribution-Share Alike 4.0</a> unless otherwise noted. Additional terms may apply for the media files.

</div>

</div></div>

</div>

</main>

</div>

</body></html>