neural network in matlab

What is Neural Network?

Neural network is a type of machine learning models. Artificial Intelligence is the other name of Neural Network. Any technology that includes some short of intelligence is a form of artificial intelligence. Machine Learning is a technique that figures out model out of data. That data can be form of documents, audio, images, etc. Machine Learning analyses the data and extract model which assembles the knowledge.

For Example, we want to develop a model that can recognize cats, so we need data that include the picture of cats. The Machine learning algorithm will train the model from these images, after the training is completed if we take an input image that contains the cat picture, the model will able to identify the cat.

Neural Network Architecture:-

  1. The architecture of a neural network is consists of a large number nodes called Neurons which are interconnected and organized into different layers.
  2. An individual node is connected to several other nodes in the previous and the next layer. They receive the inputs form one layer and processed to generate the output and they pass to the next layer.
  3. The first layer is known as input layer and the last layer is known as the output layer which produces the output and other layer between input and output layer is known as hidden layers.

Work flow for Neural Network in MATLAB:-

  1. Collect data
  2. Create the network 
  3. Configure the network 
  4. Initialize the weights and biases
  5. Train the network 
  6. Validate the network
  7. Use the network

Neural Network app in MATLAB

  1. Fitting App
  2. Pattern Recognition App
  3. Clustering App
  4. Time Series App

Steps for using all Apps:-

  1. Open the neural network toolbox using nnstart command.
  2. Choose the app according to your analysis.
  3. Choose the dataset from workspace or use inbuilt dataset.
  4. Then after choosing dataset train the model.
  5. Check the output by net command.
  6. If output is not satisfied then add the more data to get better accuracy.