Understanding Machine Learning 2021

机器学习在所有行业中变得无处不在。现在已经确定了许多使用机器学习的应用程序。很少有例子包括垃圾邮件检测、人脸识别、情感分析、对象检测、信用卡欺诈检测、天气预报，而且这个列表几乎是无穷无尽的。几乎每天都有更多的新应用被不同的行业发现。

当我们应用机器学习时，不仅仅是将卓越的技术应用于传统问题。自从应用机器学习以来，这也与商业意义有关，我们可以使实验和应用更加经济。

本课程是我的项目团队在 IIT 坎普尔学习网络安全的队列中讨论的结果。我们已着手使用机器学习创建恶意软件检测产品。虽然我们所有人都在掌握恶意软件分析，但该团队需要一些机器学习的投入。为了填补空白，我与我们的项目团队成员就机器学习进行了一些会议。本课程是这些会话的录音的集合。

本课程讨论什么是机器学习算法。我们以随机森林算法和线性回归为例来了解机器学习中的模型。我们将看到如何使用 Python 实现此类模型。在讨论机器学习模型的开发过程中，我们讨论了数据预处理、归一化、缩放等各个步骤。我们触及了神经网络的基础知识，并稍微深入研究了回归。该课程包括对什么是过拟合、什么是超参数调整等概念的讨论。

本课程试图介绍创建使用机器学习的产品需要什么。我相信这些讨论可以让人们开始将机器学习应用于许多问题。

MP4 | Video: h264, 1280×720 | Audio: AAC, 44100 Hz
Language: English | Size: 669 MB | Duration: 1h 35m

What you’ll learn
What are models in Machine Learning?
How to build models for Machine Learning?
How does Machine Learning build a Linear Regression model?

Requirements
Some knowledge of programming in any language is essential.
Description
Machine Learning is becoming ubiquitous across all industries. Already many applications have been identified which use Machine Learning now. Few examples include Spam Detection, Face Recognition, Emotion Analysis, Object Detection, Credit Card Fraud Detection, Weather Prediction, and the list is almost endless. More new applications are being identified by different industries almost everyday.

It is not just about applying superior technology for traditional problems when we apply Machine Learning. It is also about business sense since applying Machine Learning, we can make experiments and applications much more economical.

This course is a result of a discussion among my Project Team from our cohort in IIT, Kanpur learning Cyber Security. We have embarked to create a product for Malware Detection using Machine Learning. While all of us are getting grips on Malware Analysis, the team needed some inputs of Machine Learning. To fill the gap, I conducted some sessions with our Project Team members on Machine Learning. This course is a collection of the recording of these sessions.

This course discusses what are Machine Learning Algorithms. We discuss Random Forest Algorithm and Linear Regression as examples to understand what are models in Machine Learning. We see how to implement such models using Python. During the discussion on the development of the Machine Learning models, we discuss the various steps like Data Preprocessing, Normalisation, Scaling, etc. We touch upon the basics of Neural Network and take a slight deep dive into Regression. The course includes discussion on concepts like what is overfitting, what is hyper-parameter tuning, etc.

This course tries to give an idea for what it takes to create a product which uses Machine Learning. I believe that the discussions can get one started to apply Machine Learning to many problems.

Who this course is for:
Students
Professionals
Engineers
Researchers