# Introduction to Machine Learning

Machine learning is making the machine learn from the data. This is similar to how human learns. To derive the analogy, let's consider an example. If you wish to learn to make shahi paneer, you go to your mom and ask her to demonstrate the recipe as well as browse the internet for a good recipe. You follow her and next time when your friends turn up you decide to cook shahi paneer for them. Similarly, when you want your computer to do something, instead of writing the program you just provide an example data of how to accomplish the task. The computer processes the data and figures out the way to do it.Today machine learning has accomplished a lot and is being put to use in almost every field be it science, finance, marketing, social media. Every company is utilizing machine learning in one or the other way. Machine learning industry is booming. There are ample number of opportunities in this field.

The outstanding examples of this area includes Google's Deepmind, Google's self driving cars, IBM Watson, Siri etc.

## Machine learning course contents

Introduction |

Definition |

Application |

Process |

Machine learning techniques |

Supervised learning |

Unsupervised learning |

Regression |

Definition and examples |

Linear regression |

Tuning the parameters: Gradient descent algorithm |

Evaluating the model: Loss function |

Prediction |

Project: Apply linear regression to a dataset. Implement the algorithm in python and compare the results with library implementation. |

Some important concepts |

Feature extraction |

Training and testing data |

Cross validation |

The concept of Overfitting and underfitting |

Regularization |

Classification |

General overview and examples |

Naive Bayes classifier |

Error function |

Sigmoid function |

Logit function: Logistic Regression |

Support Vector Machines in detail: Kernels and parameter tuning |

Project: Implement Naive Bayes classifier and apply it to dataset and find the accuracy of the model. |

Project: Implement Logistic regression and apply it to dataset and find the accuracy of the model. |

Project: Apply SVM to dataset and find the best parameters and accuracy of the model. |

Project: Comparative study of the different models. |

Neural networks |

General overview |

Training a neural network |

Customizing the parameters and output function |

Effect of hidden layers |

Project: Train a neural network on given dataset. |

Image classification |

Dataset: Images |

Train different models and find the best model by tuning the parameters. |

### Course expectations

By the end of the course you will be able to successfully apply the different machine learning algorithms on your dataset and evaluate the performance of different models.You will also be able interpret the results and draw appropriate inferences. You will also be able to understand the application of different models and identify what model should be used in which situation.#### Course duration

The course is expected to be completed in 2 to 3 months.The fee for the course is 15k INR.

##### Course Resources

Topic | PPT | Video lecture |
---|---|---|

Introduction to machine learning and machine learning process | Introduction to machine learning | Watch Video |