Naive Bayes classifier

In statistics, naive Bayes classifiers are a family of simple "probabilistic

classifiers" based on applying Bayes' theorem with strong (naive) independence
assumptions between the features (see Bayes classifier). They are among the
simplest Bayesian network models,[1] but coupled with kernel density estimation,
they can achieve high accuracy levels.[2]

Naive Bayes classifiers are highly scalable, requiring a number of parameters

linear in the number of variables (features/predictors) in a learning problem.
Maximum-likelihood training can be done by evaluating a closed-form
expression,[3]: 7 18  which takes linear time, rather than by expensive iterative
approximation as used for many other types of classifiers. Example of a naive Bayes classifier
depicted as a Bayesian Network
In the statistics literature, naive Bayes models are known under a variety of
names, including simple Bayes and independence Bayes.[4] All these names
reference the use of Bayes' theorem in the classifier's decision rule, but naive Bayes is not (necessarily) a Bayesian
method.[3][4]

Introduction
Naive Bayes is a simple technique for constructing classifiers: models that assign class labels to problem instances,
represented as vectors of feature values, where the class labels are drawn from some finite set. There is not a single
algorithm for training such classifiers, but a family of algorithms based on a common principle: all naive Bayes classifiers
assume that the value of a particular feature is independent of the value of any other feature, given the class variable. For
example, a fruit may be considered to be an apple if it is red, round, and about 10 cm in diameter. A naive Bayes classifier
considers each of these features to contribute independently to the probability that this fruit is an apple, regardless of any
possible correlations between the color, roundness, and diameter features.

In many practical applications, parameter estimation for naive Bayes models uses the method of maximum likelihood; in
other words, one can work with the naive Bayes model without accepting Bayesian probability or using any Bayesian
methods.

Despite their naive design and apparently oversimplified assumptions, naive Bayes classifiers have worked quite well in
many complex real-world situations. In 2004, an analysis of the Bayesian classification problem showed that there are
sound theoretical reasons for the apparently implausible efficacy of naive Bayes classifiers.[5] Still, a comprehensive
comparison with other classification algorithms in 2006 showed that Bayes classification is outperformed by other
approaches, such as boosted trees or random forests.[6]

An advantage of naive Bayes is that it only requires a small number of training data to estimate the parameters necessary
for classification.[7]

Probabilistic model
Abstractly, naive Bayes is a conditional probability model: it assigns probabilities for each of the K
possible outcomes or classes given a problem instance to be classified, represented by a vector
encoding some n features (independent variables).[8]

The problem with the above formulation is that if the number of features n is large or if a feature can take on a large
number of values, then basing such a model on probability tables is infeasible. The model must therefore be reformulated
to make it more tractable. Using Bayes' theorem, the conditional probability can be decomposed as:
In plain English, using Bayesian probability terminology, the above equation can be written as

In practice, there is interest only in the numerator of that fraction, because the denominator does not depend on and the
values of the features are given, so that the denominator is effectively constant. The numerator is equivalent to the joint
probability model

which can be rewritten as follows, using the chain rule for repeated applications of the definition of conditional
probability:

Now the "naive" conditional independence assumptions come into play: assume that all features in are mutually
independent, conditional on the category . Under this assumption,

Thus, the joint model can be expressed as

where denotes proportionality.

This means that under the above independence assumptions, the conditional distribution over the class variable is:

where the evidence is a scaling factor dependent only on , that is, a

constant if the values of the feature variables are known.

Constructing a classifier from the probability model

The discussion so far has derived the independent feature model, that is, the naive Bayes probability model. The naive
Bayes classifier combines this model with a decision rule. One common rule is to pick the hypothesis that is most
probable so as to minimize the probability of misclassification; this is known as the maximum a posteriori or MAP
decision rule. The corresponding classifier, a Bayes classifier, is the function that assigns a class label for some k
as follows:
Parameter estimation and event models
A class's prior may be calculated by assuming equiprobable classes, i.e.,
, or by calculating an estimate for the class probability from the
training set:

To estimate the parameters for a feature's distribution, one must assume a Likelihood functions ,
distribution or generate nonparametric models for the features from the training Confusion matrix and ROC curve. For
set.[9] the naive Bayes classifier and given
that the a priori probabilities are
The assumptions on distributions of features are called the "event model" of the the same for all classes, then the
naive Bayes classifier. For discrete features like the ones encountered in decision boundary (green line) would
document classification (include spam filtering), multinomial and Bernoulli be placed on the point where the two
distributions are popular. These assumptions lead to two distinct models, which probability densities intersect, due to
are often confused.[10][11]

.
Gaussian naive Bayes

When dealing with continuous data, a typical assumption is that the continuous values associated with each class are
distributed according to a normal (or Gaussian) distribution. For example, suppose the training data contains a continuous
attribute, . The data is first segmented by the class, and then the mean and variance of is computed in each class. Let
be the mean of the values in associated with class , and let be the Bessel corrected variance of the values in
associated with class . Suppose one has collected some observation value . Then, the probability density of given a
class , i.e., , can be computed by plugging into the equation for a normal distribution parameterized
by and . Formally,

Another common technique for handling continuous values is to use binning to discretize the feature values and obtain a
new set of Bernoulli-distributed features. Some literature suggests that this is required in order to use naive Bayes, but it is
not true, as the discretization may throw away discriminative information.[4]

Sometimes the distribution of class-conditional marginal densities is far from normal. In these cases, kernel density
estimation can be used for a more realistic estimate of the marginal densities of each class. This method, which was
introduced by John and Langley,[9] can boost the accuracy of the classifier considerably.[12][2]

Multinomial naive Bayes

With a multinomial event model, samples (feature vectors) represent the frequencies with which certain events have been
generated by a multinomial where is the probability that event i occurs (or K such multinomials in the
multiclass case). A feature vector is then a histogram, with counting the number of times event i
was observed in a particular instance. This is the event model typically used for document classification, with events
representing the occurrence of a word in a single document (see bag of words assumption). The likelihood of observing a
histogram x is given by:
 where .

The multinomial naive Bayes classifier becomes a linear classifier when expressed in log-space:[13]

where and .

If a given class and feature value never occur together in the training data, then the frequency-based probability estimate
will be zero, because the probability estimate is directly proportional to the number of occurrences of a feature's value.
This is problematic because it will wipe out all information in the other probabilities when they are multiplied. Therefore,
it is often desirable to incorporate a small-sample correction, called pseudocount, in all probability estimates such that no
probability is ever set to be exactly zero. This way of regularizing naive Bayes is called Laplace smoothing when the
pseudocount is one, and Lidstone smoothing in the general case.

Rennie et al. discuss problems with the multinomial assumption in the context of document classification and possible
ways to alleviate those problems, including the use of tf–idf weights instead of raw term frequencies and document length
normalization, to produce a naive Bayes classifier that is competitive with support vector machines.[13]

Bernoulli naive Bayes

In the multivariate Bernoulli event model, features are independent Booleans (binary variables) describing inputs. Like
the multinomial model, this model is popular for document classification tasks,[10] where binary term occurrence features
are used rather than term frequencies. If is a boolean expressing the occurrence or absence of the i'th term from the
vocabulary, then the likelihood of a document given a class is given by:[10]

where is the probability of class generating the term . This event model is especially popular for classifying
short texts. It has the benefit of explicitly modelling the absence of terms. Note that a naive Bayes classifier with a
Bernoulli event model is not the same as a multinomial NB classifier with frequency counts truncated to one.

Semi-supervised parameter estimation

Given a way to train a naive Bayes classifier from labeled data, it's possible to construct a semi-supervised training
algorithm that can learn from a combination of labeled and unlabeled data by running the supervised learning algorithm in
a loop:[14]

Given a collection of labeled samples L and unlabeled samples U, start by training a naive
Bayes classifier on L .
Until convergence, do:

Predict class probabilities for all examples x in .

Re-train the model based on the probabilities (not the labels) predicted in the previous step.

Convergence is determined based on improvement to the model likelihood , where denotes the parameters of
the naive Bayes model.
This training algorithm is an instance of the more general expectation–maximization algorithm (EM): the prediction step
inside the loop is the E-step of EM, while the re-training of naive Bayes is the M-step. The algorithm is formally justified
by the assumption that the data are generated by a mixture model, and the components of this mixture model are exactly
the classes of the classification problem.[14]

Discussion
Despite the fact that the far-reaching independence assumptions are often inaccurate, the naive Bayes classifier has
several properties that make it surprisingly useful in practice. In particular, the decoupling of the class conditional feature
distributions means that each distribution can be independently estimated as a one-dimensional distribution. This helps
alleviate problems stemming from the curse of dimensionality, such as the need for data sets that scale exponentially with
the number of features. While naive Bayes often fails to produce a good estimate for the correct class probabilities,[15] this
may not be a requirement for many applications. For example, the naive Bayes classifier will make the correct MAP
decision rule classification so long as the correct class is predicted as more probable than any other class. This is true
regardless of whether the probability estimate is slightly, or even grossly inaccurate. In this manner, the overall classifier
can be robust enough to ignore serious deficiencies in its underlying naive probability model.[16] Other reasons for the
observed success of the naive Bayes classifier are discussed in the literature cited below.

Relation to logistic regression

In the case of discrete inputs (indicator or frequency features for discrete events), naive Bayes classifiers form a
generative-discriminative pair with multinomial logistic regression classifiers: each naive Bayes classifier can be
considered a way of fitting a probability model that optimizes the joint likelihood , while logistic regression fits
the same probability model to optimize the conditional .[17]

More formally, we have the following:

Theorem — Naive Bayes classifiers on binary features are subsumed by logistic regression classifiers.

Proof

Consider a generic multiclass classification problem, with possible classes , then the
(non-naive) Bayes classifier gives, by Bayes theorem:

The naive Bayes classifier gives

where
 This is exactly a logistic regression classifier.

The link between the two can be seen by observing that the decision function for naive Bayes (in the binary case) can be
rewritten as "predict class if the odds of exceed those of ". Expressing this in log-space gives:

The left-hand side of this equation is the log-odds, or logit, the quantity predicted by the linear model that underlies
logistic regression. Since naive Bayes is also a linear model for the two "discrete" event models, it can be reparametrised
as a linear function . Obtaining the probabilities is then a matter of applying the logistic function to
, or in the multiclass case, the softmax function.

Discriminative classifiers have lower asymptotic error than generative ones; however, research by Ng and Jordan has
shown that in some practical cases naive Bayes can outperform logistic regression because it reaches its asymptotic error
faster.[17]

Examples

Person classification

Problem: classify whether a given person is a male or a female based on the measured features. The features include
height, weight, and foot size. Although with NB classifier we treat them as independent, they are not in reality.

Training

Example training set below.

Person height (feet) weight (lbs) foot size(inches)

male 6 180 12

male 5.92 (5'11") 190 11

male 5.58 (5'7") 170 12

male 5.92 (5'11") 165 10

female 5 100 6
female 5.5 (5'6") 150 8

female 5.42 (5'5") 130 7

female 5.75 (5'9") 150 9

The classifier created from the training set using a Gaussian distribution assumption would be (given variances are
unbiased sample variances):

Person mean (height) variance (height) mean (weight) variance (weight) mean (foot size) variance (foot size)
−2 2
male 5.855 3.5033 × 10 176.25 1.2292 × 10 11.25 9.1667 × 10−1

female 5.4175 9.7225 × 10−2 132.5 5.5833 × 102 7.5 1.6667

The following example assumes equiprobable classes so that P(male)= P(female) = 0.5. This prior probability distribution
might be based on prior knowledge of frequencies in the larger population or in the training set.

Testing
Below is a sample to be classified as male or female.

Person height (feet) weight (lbs) foot size(inches)

sample 6 130 8

In order to classify the sample, one has to determine which posterior is greater, male or female. For the classification as
male the posterior is given by

For the classification as female the posterior is given by

The evidence (also termed normalizing constant) may be calculated:

However, given the sample, the evidence is a constant and thus scales both posteriors equally. It therefore does not affect
classification and can be ignored. The probability distribution for the sex of the sample can now be determined:

where and are the parameters of normal distribution which have been previously
determined from the training set. Note that a value greater than 1 is OK here – it is a probability density rather than a
probability, because height is a continuous variable.

Since posterior numerator is greater in the female case, the prediction is that the sample is female.

Document classification

Here is a worked example of naive Bayesian classification to the document classification problem. Consider the problem
of classifying documents by their content, for example into spam and non-spam e-mails. Imagine that documents are
drawn from a number of classes of documents which can be modeled as sets of words where the (independent)
probability that the i-th word of a given document occurs in a document from class C can be written as
(For this treatment, things are further simplified by assuming that words are randomly distributed in the document - that is,
words are not dependent on the length of the document, position within the document with relation to other words, or
other document-context.)

Then the probability that a given document D contains all of the words , given a class C, is

The question that has to be answered is: "what is the probability that a given document D belongs to a given class C?" In
other words, what is ?

Now by definition

and

Bayes' theorem manipulates these into a statement of probability in terms of likelihood.

Assume for the moment that there are only two mutually exclusive classes, S and ¬S (e.g. spam and not spam), such that
every element (email) is in either one or the other;

and

Using the Bayesian result above, one can write:

Dividing one by the other gives:

Which can be re-factored as:

Thus, the probability ratio p(S | D) / p(¬S | D) can be expressed in terms of a series of likelihood ratios. The actual
probability p(S | D) can be easily computed from log (p(S | D) / p(¬S | D)) based on the observation that p(S | D) + p(¬S |
D) = 1.

Taking the logarithm of all these ratios, one obtains:

(This technique of "log-likelihood ratios" is a common technique in statistics. In the case of two mutually exclusive
alternatives (such as this example), the conversion of a log-likelihood ratio to a probability takes the form of a sigmoid
curve: see logit for details.)

Finally, the document can be classified as follows. It is spam if (i. e., ),

otherwise it is not spam.

See also
AODE
Bayes classifier
Bayesian spam filtering
Bayesian network
Random naive Bayes
Linear classifier
Logistic regression
Perceptron
Take-the-best heuristic

References
1. McCallum, Andrew. "Graphical Models, Lecture2: Bayesian Network Representation" (https://people.cs.u
mass.edu/~mccallum/courses/gm2011/02-bn-rep.pdf) (PDF). Archived (https://ghostarchive.org/archive/2
0221009/https://people.cs.umass.edu/~mccallum/courses/gm2011/02-bn-rep.pdf) (PDF) from the original
on 2022-10-09. Retrieved 22 October 2019.
2. Hastie, Trevor. (2001). The elements of statistical learning : data mining, inference, and prediction : with
200 full-color illustrations. Tibshirani, Robert., Friedman, J. H. (Jerome H.). New York: Springer. ISBN 0-
387-95284-5. OCLC 46809224 (https://www.worldcat.org/oclc/46809224).
3. Russell, Stuart; Norvig, Peter (2003) [1995]. Artificial Intelligence: A Modern Approach (2nd ed.). Prentice
Hall. ISBN 978-0137903955.
4. Hand, D. J.; Yu, K. (2001). "Idiot's Bayes — not so stupid after all?". International Statistical Review. 69
(3): 385–399. doi:10.2307/1403452 (https://doi.org/10.2307%2F1403452). ISSN 0306-7734 (https://www.
worldcat.org/issn/0306-7734). JSTOR 1403452 (https://www.jstor.org/stable/1403452).
5. Zhang, Harry. The Optimality of Naive Bayes (http://www.cs.unb.ca/profs/hzhang/publications/FLAIRS04
ZhangH.pdf) (PDF). FLAIRS2004 conference.
6. Caruana, R.; Niculescu-Mizil, A. (2006). An empirical comparison of supervised learning algorithms.
Proc. 23rd International Conference on Machine Learning. CiteSeerX 10.1.1.122.5901 (https://citeseerx.i
st.psu.edu/viewdoc/summary?doi=10.1.1.122.5901).
7. "Why does Naive Bayes work better when the number of features >> sample size compared to more
sophisticated ML algorithms?" (https://stats.stackexchange.com/q/379383). Cross Validated Stack
Exchange. Retrieved 24 January 2023.
8. Narasimha Murty, M.; Susheela Devi, V. (2011). Pattern Recognition: An Algorithmic Approach.
ISBN 978-0857294944.
 9. John, George H.; Langley, Pat (1995). Estimating Continuous Distributions in Bayesian Classifiers (http
s://dl.acm.org/doi/10.5555/2074158.2074196). Proc. Eleventh Conf. on Uncertainty in Artificial
Intelligence. Morgan Kaufmann. pp. 338–345. arXiv:1302.4964 (https://arxiv.org/abs/1302.4964).
10. McCallum, Andrew; Nigam, Kamal (1998). A comparison of event models for Naive Bayes text
classification (http://www.kamalnigam.com/papers/multinomial-aaaiws98.pdf) (PDF). AAAI-98 workshop
on learning for text categorization. Vol. 752. Archived (https://ghostarchive.org/archive/20221009/http://w
ww.kamalnigam.com/papers/multinomial-aaaiws98.pdf) (PDF) from the original on 2022-10-09.
11. Metsis, Vangelis; Androutsopoulos, Ion; Paliouras, Georgios (2006). Spam filtering with Naive Bayes—
which Naive Bayes? (https://www.researchgate.net/publication/221650814). Third conference on email
and anti-spam (CEAS). Vol. 17.
12. Piryonesi, S. Madeh; El-Diraby, Tamer E. (2020-06-01). "Role of Data Analytics in Infrastructure Asset
Management: Overcoming Data Size and Quality Problems". Journal of Transportation Engineering, Part
B: Pavements. 146 (2): 04020022. doi:10.1061/JPEODX.0000175 (https://doi.org/10.1061%2FJPEODX.
0000175). S2CID 216485629 (https://api.semanticscholar.org/CorpusID:216485629).
13. Rennie, J.; Shih, L.; Teevan, J.; Karger, D. (2003). Tackling the poor assumptions of naive Bayes
classifiers (http://people.csail.mit.edu/~jrennie/papers/icml03-nb.pdf) (PDF). ICML. Archived (https://ghost
archive.org/archive/20221009/http://people.csail.mit.edu/~jrennie/papers/icml03-nb.pdf) (PDF) from the
original on 2022-10-09.
14. Nigam, Kamal; McCallum, Andrew; Thrun, Sebastian; Mitchell, Tom (2000). "Learning to classify text from
labeled and unlabeled documents using EM" (http://www.kamalnigam.com/papers/emcat-aaai98.pdf)
(PDF). Machine Learning. 39 (2/3): 103–134. doi:10.1023/A:1007692713085 (https://doi.org/10.1023%2F
A%3A1007692713085). S2CID 686980 (https://api.semanticscholar.org/CorpusID:686980). Archived (htt
ps://ghostarchive.org/archive/20221009/http://www.kamalnigam.com/papers/emcat-aaai98.pdf) (PDF)
from the original on 2022-10-09.
15. Niculescu-Mizil, Alexandru; Caruana, Rich (2005). Predicting good probabilities with supervised learning
(https://web.archive.org/web/20140311005243/http://machinelearning.wustl.edu/mlpapers/paper_files/ic
ml2005_Niculescu-MizilC05.pdf) (PDF). ICML. doi:10.1145/1102351.1102430 (https://doi.org/10.1145%2
F1102351.1102430). Archived from the original (http://machinelearning.wustl.edu/mlpapers/paper_files/ic
ml2005_Niculescu-MizilC05.pdf) (PDF) on 2014-03-11. Retrieved 2016-04-24.
16. Rish, Irina (2001). An empirical study of the naive Bayes classifier (http://www.research.ibm.com/people/
r/rish/papers/RC22230.pdf) (PDF). IJCAI Workshop on Empirical Methods in AI. Archived (https://ghostar
chive.org/archive/20221009/http://www.research.ibm.com/people/r/rish/papers/RC22230.pdf) (PDF) from
the original on 2022-10-09.
17. Ng, Andrew Y.; Jordan, Michael I. (2002). On discriminative vs. generative classifiers: A comparison of
logistic regression and naive Bayes (http://papers.nips.cc/paper/2020-on-discriminative-vs-generative-cla
ssifiers-a-comparison-of-logistic-regression-and-naive-bayes). NIPS. Vol. 14.

Further reading
Domingos, Pedro; Pazzani, Michael (1997). "On the optimality of the simple Bayesian classifier under
zero-one loss" (http://citeseer.ist.psu.edu/domingos97optimality.html). Machine Learning. 29 (2/3): 103–
137. doi:10.1023/A:1007413511361 (https://doi.org/10.1023%2FA%3A1007413511361).
Webb, G. I.; Boughton, J.; Wang, Z. (2005). "Not So Naive Bayes: Aggregating One-Dependence
Estimators" (https://doi.org/10.1007%2Fs10994-005-4258-6). Machine Learning. 58 (1): 5–24.
doi:10.1007/s10994-005-4258-6 (https://doi.org/10.1007%2Fs10994-005-4258-6).
Mozina, M.; Demsar, J.; Kattan, M.; Zupan, B. (2004). Nomograms for Visualization of Naive Bayesian
Classifier (http://eprints.fri.uni-lj.si/154/01/PKDD_camera_mozina.pdf) (PDF). Proc. PKDD-2004.
pp. 337–348.
Maron, M. E. (1961). "Automatic Indexing: An Experimental Inquiry". Journal of the ACM. 8 (3): 404–417.
doi:10.1145/321075.321084 (https://doi.org/10.1145%2F321075.321084). hdl:2027/uva.x030748531 (htt
ps://hdl.handle.net/2027%2Fuva.x030748531). S2CID 6692916 (https://api.semanticscholar.org/CorpusI
D:6692916).
 Minsky, M. (1961). Steps toward Artificial Intelligence. Proc. IRE. Vol. 49. pp. 8–30.

External links
Book Chapter: Naive Bayes text classification, Introduction to Information Retrieval (http://nlp.stanford.ed
u/IR-book/html/htmledition/naive-bayes-text-classification-1.html)
Naive Bayes for Text Classification with Unbalanced Classes (http://www.cs.waikato.ac.nz/~eibe/pubs/Fr
ankAndBouckaertPKDD06new.pdf)
Benchmark results of Naive Bayes implementations (http://tunedit.org/results?d=UCI/&a=bayes)
Hierarchical Naive Bayes Classifiers for uncertain data (http://www.biomedcentral.com/1471-2105/7/514)
(an extension of the Naive Bayes classifier).
Naïve Bayes Algorithm | Maximum A Posteriori in Machine Learning (https://medium.com/@rizwanayasm
een06/na%C3%AFve-bayes-algorithm-maximum-a-posteriori-in-machine-learning-2d965a2fbd58)

Software

Naive Bayes classifiers are available in many general-purpose machine learning and NLP packages,
including Apache Mahout, Mallet (http://mallet.cs.umass.edu/), NLTK, Orange, scikit-learn and Weka.
IMSL Numerical Libraries Collections of math and statistical algorithms available in C/C++, Fortran, Java
and C#/.NET. Data mining routines in the IMSL Libraries include a Naive Bayes classifier.
An interactive Microsoft Excel spreadsheet Naive Bayes implementation (https://downloads.sourceforge.
net/naivebayesclass/NaiveBayesDemo.xls?use_mirror=osdn) using VBA (requires enabled macros) with
viewable source code.
jBNC - Bayesian Network Classifier Toolbox (https://jbnc.sourceforge.net/)
Statistical Pattern Recognition Toolbox for Matlab (http://cmp.felk.cvut.cz/cmp/software/stprtool/).
ifile (http://people.csail.mit.edu/jrennie/ifile/) - the first freely available (Naive) Bayesian mail/spam filter
NClassifier (https://nclassifier.sourceforge.net/) - NClassifier is a .NET library that supports text
classification and text summarization. It is a port of Classifier4J.
Classifier4J (https://classifier4j.sourceforge.net/) - Classifier4J is a Java library designed to do text
classification. It comes with an implementation of a Bayesian classifier.
JNBC (http://naivebayesclassifier.org/) Naive Bayes Classifier running in-memory or using fast key-value
stores (MapDB, LevelDB or RocksDB).
Blayze (https://github.com/Tradeshift/blayze) - Blayze is a minimal JVM library for Naive Bayes
classification written in Kotlin.

Retrieved from "https://en.wikipedia.org/w/index.php?title=Naive_Bayes_classifier&oldid=1163543509"