Natural language processing with ML.NET: Introduction - ML.NET Tutorial

- Welcome to Natural Language Processing with ML.NET course, and thanks for joining. I'm Carlotta Castelluccio, a Cloud Advocate at Microsoft, focusing on machine learning and artificial intelligence technologies, and my job's mission is to help every student, developer, or entrepreneur succeed with AI by building innovative and responsible AI solutions. To achieve this goal, I develop technical content and host skilling sessions, enabling my audience to take the most out of AI technologies. In today's fast-paced world, machine learning has become an integral part of our daily lives. The newest advancements in the field of artificial intelligence and in particular of natural language processing unlocked new scenarios and possibilities. In this context, acquiring the skills needed to integrate AI capabilities into your solutions has become increasingly important. In this course, we are going to demystify the main concepts of machine learning and natural language processing, and we're going to put them into practice with some hands-on examples using the ML.NET framework. What I love about the ML.NET framework is that it's designed to democratize the art of machine learning to all developers, and it enables a smoother integration of machine learning models into .NET applications. The first lesson of this course provides an overview of the main concepts and terminology we'll be using throughout the whole course. Don't worry if you are a complete beginners because here's where we set the basis of what machine learning is, what the .NET ecosystem is, and how ML.NET is placed within it. We explore common real-world use cases in which we can leverage AI capabilities, and we discover how developers can address these scenarios with ML.NET framework. We also set up the main prerequisites We need to complete the course exercises, including the tools, such as Visual Studio and Visual Studio Code, and the extensions, such as Model Builder in Visual Studio and Polyglot Notebooks in Visual Studio Code. In the second lesson, we deal with one of the most common applications of machine learning algorithms, which is classification. We also look at how we can build, evaluate, and consume a classification model with ML.NET, with both a low-code approach using Model Builder and a code-first approach using Polyglot Notebooks and AutoML. After that, with the third lesson, we dive deep into a specific kind of classification, which is text classification. Text classification is one of the applications in the domain of natural language processing, and so we cover what is natural language processing and the inner workings of a language model. At this point, we are ready to get our fingers dirty with some code, so we go over an example of text classification in ML.NET. This time, we do not only build and evaluate a model, starting from a pre-trained one provided by the framework, but we also look at hyperparameters tuning to improve our first result. The third lesson also includes an advanced scenario of natural language processing application, which is sentence similarity. Sentence similarity is very useful in semantic search type of scenarios or information retrieval. Alongside covering the main concept about what is sentence similarity and how you can implement it with ML.NET framework, we also see this in practice with hands-on exercises. Training and evaluating machine learning model is the core part of a machine learning lifecycle. However, it's good to have an overview of the whole process of building, deploying, and maintaining a machine learning model following MLOps principles and practices and using key tools, like the cloud, a versioning system and continuous integration/continuous deployment pipelines. This is covered by the fourth and last lesson of the course, which also includes a hands-on demo. And looking at the whole development lifecycle, it's also essential to be aware of artificial intelligence's risk and limitations and be accountable for the solutions we built. Well, that's the plan. We are ready to kick off, and I hope you enjoy the journey.