This article will list some of the best resources for beginners to learn about Machine Learning.
Programming libraries
The following are the best open source Machine Learning programming libraries today. I don't think it's all suitable for use in real systems, but they are the ideal base for you to learn, learn and create prototyping.
Start with a library in a language you know well, then switch to other more powerful libraries. If you are a good programmer, you can easily switch from one language to another. All have the same logic, only different in syntax and APIs.
- R Project for Statistical Computing : This is the environment and scripting language is similar to lisp. All statistics will be provided in R language, including graphing. In Machine Learning section on CRAN (Machine Learning packages of third parties), there is a code section written by experts in this field. If you want to create a prototype, you must learn R
- WEKA : This is a Data Mining workbench that provides APIs, some command lines and a graphical interface for the entire data mining life cycle (data mining). You can prepare data, visualize, classify, regress and group models and many algorithms are provided through third-party plugins. Unrelated to WEKA, Mahout is a very good Java framework for Machine Learning on Hadoop infrastructure. If you're just starting out with big data and machine learning , stick with WEKA and learn it part by part.
- Scikit Learn : Machine Learning in Python is built on NumPy and Scipy. This library is very suitable for Python or Ruby programmers. It is very easy to use, powerful and comes with great documentation. Orange will be a good choice if you want to test something new.
- Octave : If you're familiar with Matlab or a NumPy programmer looking for something different, try Octave. Octave is an environment that supports numerical computation, similar to Matlab, can write programs that easily solve linear and non-linear problems. If you already have a technical background, you should start with Octave.
- BigML : You don't want to write code. You can use tools like WEKA that do not require any programming at all or use services such as BigML to provide interface learning machines on the web, which help you explore all construction models. in the browser.
Choose a platform and use it to practice while learning about machine learning. Don't just read the theory smoothly, start working.
Video courses
- Stanford Machine Learning : includes Stanford 's Andrew Ng lectures on Coursera. After enrollment, you can view all lectures at any time and get practical knowledge like in the Stanford CS229 course. This course has both homework and a test that focuses on linear algebra and how to use Octave.
- Caltech Learning from Data : taught by Yaser Abu-Mostafa. All lectures and course materials are available on the CalTech website . Similar to Stanford, depending on the speed you want, you can study and complete assignments. Caltech Learning from Data includes detailed topics and focuses on math. For beginners, homework will be a big challenge.
- Machine Learning Category on VideoLectures.Net : Look for videos that are widely viewed and rated the most because there is a lot of content here.
- Talk by Jeremy Howard – Talk by Jeremy Howard : The talk is very valuable for a group of R group users.
TopDev Techtalk # 52: Technology of the Future: What Breakthrough from Machine Learning?
* Ho Chi Minh: 18:00 – 21:00 on December 2, 2016.
Some other documents
- The Discipline of Machine Learning : This is a white book (white book is a report or a guide of a competent authority with the aim of helping readers understand a problem, solving a problem or making a decision ) defines Machine Learning principles that are punctuated by author Tom Mitchell. This is part of the document that Mitchell used to convince the CMU university principal to create an independent Machine Learning department (see a short interview with Tom Mitchell here ).
- Useful Things to Know about Machine Learning : Great material refers to specific algorithms and promotes a number of important issues such as feature selection generalizability and simplicity of model. These are all important knowledge for beginners
Book about Machine Learning for beginners
- Programming Collective Intelligence: Building Smart Web 2.0 Applications : The book is written for the programmer object. It is light in theory, heavy on programming examples as well as practical solutions and problems on the web.
- Machine Learning for Hackers : I recommend reading this book after reading the Programming Collective Intelligence (above). Machine Learning for Hackers provides practical examples, but is prone to data analysis and using R. I really like this book!
- Machine Learning: An Algorithmic Perspective : This book is like an advanced version of Programming Collective Intelligence (above), aimed at programmers who started studying Machine Learning. Books also include math and examples written in Python. You should read this book after reading the Programming Collective Intelligence .
- Data Mining: Practical Machine Learning Tools and Techniques, Third Edition : I started with this book with the first version published in 2000. At that time I was a Java programmer so this book was combined with the library. WEKA becomes a perfect environment for me to test and implement my algorithms.
- Machine Learning : An old book of recipes and lots of references. The content of the product is clear and easy to access through algorithms.
Of course there are many other great books about Machine Learning out there, but I think these documents are really suitable for those who are new to this field.
Source: Techmaster