Scribd
Upload
49 views

Naive Bayes Algorithm

The document provides information about three different machine learning algorithms: Naive Bayes, Support Vector Machines (SVM), and Artificial Neural Networks (ANN). The Naive Bayes algorithm is a classification technique based on Bayes' theorem that considers all features to independently contribute to the probability of a class. SVM is a classification algorithm that determines decision boundaries to optimally separate classes with the largest possible distance. ANN is inspired by biological neural networks and uses interconnected "neurons" that can perform computations and machine learning through training to approximate unknown functions.

Uploaded by

SaranKumar
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
49 views

Naive Bayes Algorithm

The document provides information about three different machine learning algorithms: Naive Bayes, Support Vector Machines (SVM), and Artificial Neural Networks (ANN). The Naive Bayes algorithm is a classification technique based on Bayes' theorem that considers all features to independently contribute to the probability of a class. SVM is a classification algorithm that determines decision boundaries to optimally separate classes with the largest possible distance. ANN is inspired by biological neural networks and uses interconnected "neurons" that can perform computations and machine learning through training to approximate unknown functions.

Uploaded by

SaranKumar
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Naive Bayes Algorithm

The Naive Bayes Classifier technique is based on the Bayesian theorem. It is


particularly used when the dimensionality of the inputs is high. The Bayesian
Classifier is capable of calculating the possible output. That is based on the
input. It is also possible to add new raw data at runtime and have a better
probabilistic classifier.
This classifier considers the presence of a particular feature of a class. That is
unrelated to the presence of any other feature when the class variable is given.
For example, a fruit may consider to be an apple if it is red, round.
Even if these features depend on each other features of a class. A naive Bayes
classifier considers all these properties to contribute to the probability. That it
shows this fruit is an apple. Algorithm works as follows,
Bayes theorem provides a way of calculating the posterior probability, P(c|x),
from P(c), P(x), and P(x|c). Naive Bayes classifier considers the effect of the
value of a predictor (x) on a given class (c). That is independent of the values of
other predictors.
P(c|x) is the posterior probability of class (target) given predictor (attribute) of
class.
P(c) is called the prior probability of class.
P(x|c) is the likelihood which is the probability of predictor of given class.
P(x) is the prior probability of predictor of class.

SVM Algorithm
SVM has attracted a great deal of attention in the last decade. It also applied to various
domains of applications. SVMs are used for learning classification, regression or
ranking function. SVM is based on statistical learning theory and structural risk
minimization principle. And have the aim of determining the location of decision
boundaries. It is also known as a hyperplane. That produces the optimal separation of
classes. Thereby creating the largest possible distance between the separating
hyperplane. Further, the instances on either side of it have been proven. That is to
reduce an upper bound on the expected generalization error.
The efficiency of SVM based does not depend on the dimension of classified entities.
Though, SVM is the most robust and accurate classification technique. Also, there are
several problems. The data analysis in SVM is based on convex quadratic
programming. Also, expensive, as solving quadratic programming methods. That need
large matrix operations as well as time-consuming numerical computations.
Also See – Data Mining Applications and Use Cases
Training time for SVM scales in the number of examples. So researchers strive all the
time for more efficient training algorithm. That resulting in several variant based
algorithm.
SVM can also extend to learn non-linear decision functions. That is by first projecting
the input data onto a high-dimensional feature space. As by using kernel functions and
formulating a linear classification problem. The resulting feature space is much larger
than the size of a dataset. That is not possible to store on popular computers.
Investigation of this issues leads to several decomposition based algorithms. The basic
idea of decomposition method is to split the variables into two parts:
a set of free variables called as a working set. That can update in each iteration and set
of fixed variables. That are fix during a particular. Now, this procedure have to repeat
until the termination conditions are met
The SVM was developed for binary classification. And it is not simple to extend it for
multi-class classification problem. The basic idea to apply multi-classification to SVM.
That is to decompose the multi-class problems into several two-class problems. That
can address using several SVMs.
Follow this link to know more about Data Mining Algorithms-SVM

ANN Algorithm
This is the types of computer architecture inspire by biological neural networks.
They are used to approximate functions. That can depend on a large number of inputs
and are generally unknown.
They are presented as systems of interconnected “neurons”. That can compute values
from inputs. Also, they are capable of machine learning as well as pattern recognition.
Due to their adaptive nature.
An artificial neural network operates by creating connections between many different
processing elements. That each corresponding to a single neuron in a biological brain.
These neurons may actually construct or simulate by a digital computer system. Each
neuron takes many input signals. Then based on an internal weighting. That produces a
single output signal that is sent as input to another neuron.
The neurons are interconnected and organized into different layers. The input layer
receives the input and the output layer produces the final output. In general, one or more
hidden layers are sandwiched between the two. This structure makes it impossible to
forecast or know the exact flow of data.
Artificial neural networks start out with randomized weights for all their neurons. This
means that they need to train to solve the particular problem for which they are
proposed. A back-propagation ANN is trained by humans to perform specific tasks.
During the training period, we can test whether the ANN’s output is correct by
observing a pattern. If it’s correct the neural weightings produce that output is
reinforced. if the output is incorrect, those weightings responsible diminish.
To know More about Data Mining Algorithm – Artificial Neural Network, follow
this link
Implemented on a single computer, a network is slower than more traditional solutions.
The ANN’s parallel nature allows it to built using many processors. That gives a great
speed advantage at very little development cost. The parallel architecture allows ANNs
to process amounts of data very in less time. It deals with large continuous streams of
information. Such as speech recognition or machine sensor data. ANNs can operate
faster as compared to other algorithms.
An artificial neural network is useful in a variety of real-world applications. Such as
visual pattern recognition and speech recognition. That deals with complex often
incomplete data. Also, recent programs for text-to-speech have utilized ANNs. Many
handwriting analysis programs are currently using ANNs.

You might also like