You are on page 1of 4

A neural network is a series of algorithms that endeavors to recognize underlying

relationships in a set of data through a process that mimics the way the human brain

operates. In this sense, neural networks refer to systems of neurons, either organic or

artificial in nature. Neural networks can adapt to changing input; so, the network

generates the best possible result without needing to redesign the output criteria.

Neural networks are also ideally suited to help people solve complex problems in real-

life situations. They can learn and model the relationships between inputs and outputs

that are nonlinear and complex; make generalizations and inferences; reveal hidden

relationships, patterns and predictions; and model highly volatile data (such as financial

time series data) and variances needed to predict rare events (such as fraud detection).

III. Artificial Neural Networks (ANN)

• In Artificial Neural Networks, nodes or neurons are generally arranged in layers.

• Minimum- 3 layers – Input Layer, Hidden Layer, Output layer.

The three layers in Artificial Neural Network are:

• Input layer

• Hidden layer

• Output layer
Each layer is further divided into several blocks called nodes. Each node is

assigned a particular task to perform and that is further passed to the next layer.

Input Layer:

• It is the first layer of a neural network.

• It is used to take data as input.

• The input layer is responsible to feed in the input and no processing takes place

here.

Hidden Layer:

• It is a layer between the input and output layers.

• This layer is responsible for processing the input fed into the system.

• Processing is not visible to anyone and thus is called the hidden layer.

• Each node of the hidden layer has its own functionality, and hence, the defined

machine learning algorithm gets executed on the data received from the input

layer.

• The processed information is then given to the subsequent hidden layer in the

network.

• The number of hidden layers in a neural network system totally depends on the

complexity of the function for which the network has been designed.

• The number of nodes in each layer also varies.

Output Layer:

• It is the last layer of the neural network.

• It displays the results as given by the system.

• The last hidden layer passes the final processed data to the output layer which

then gives it to the user as the final output.

• Similar to the input layer, output layer too does not process the data which it

acquires.

• It is meant for user-interface.

Nodes:

• Node is trainable (weight and threshold – can be changed).


• Each node is assigned to a particular task and it is further passed to the next

layer.

• A node can take several inputs. Each input has a desired output. If undesired

output is given, then it seems some mistake in processing. So weight/ threshold

will be adjusted to correct the mistake.

• Data enters in ANN, through input layer. Input layer communicates the data to

the hidden layer. Hidden layer process the data and send the output to another

hidden layer or output layer. The output layer provides the output.

Creating brain like neuron networks is not easy, primarily due to following reasons:

• The problem of structure. Understanding about the connection between the

neurons is difficult.

• Problem of weights and Threshold.

• The functioning of neurons might be complex.

Supervised learning:

In this type of model, the algorithm learns from a dataset which is

labelled, and the algorithm uses the answer keys to evaluate its accuracy on the

training data. Classification and Regression models are types of supervised learning.

Unsupervised learning:

In this type of model, the algorithm learns and makes sense by

extracting features/patterns from the unlabelled dataset provided (The system will

evaluate by itself) Eg: Clustering and Association

Reinforcement learning:
In this type of model, the algorithms work towards accomplishing

the goal or try to improve the performance in a particular task. This is used in

gaming.

• Learning based on the feedback.

• Positive feedback – helps the system to recognise right or correct output.

• Negative feedback – helps the system to recognise incorrect/wrong

output.

You might also like