|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/natural-language-processing-with-machine-learning-in-python/
课程评论：没有评论
课程名称：使用Python进行机器学习的自然语言处理 概述：欢迎来到这门课程！本课程精心设计，旨在让您学习自然语言处理的基础知识，并逐步掌握解决复杂NLP问题的机器学习技术。课程内容均为实践操作，您将通过亲身实践来学习，无需具备自然语言处理、机器学习或Python的先前经验，但需熟悉至少一种编程语言。Python作为解决机器学习问题的优选语言，也适用于本课程。若您对Python不熟悉，不用担心，我会在使用之前向您简要介绍所需的知识。 在课程中，我们将使用Google Colab运行代码，您无需在自己的设备上安装或配置任何工具，只需联网即可。这意味着无论您的操作系统或硬件规格如何，都可以参与学习。如果您有兴趣，也可以在自己的计算机上安装Jupyter Notebook并运行相同的代码。 课程内容包括： 1. **自然语言处理基础概念**：学习标记化、词干提取和词形还原等基本概念，使用NLTK库。同时，我们会讨论不同方法的优缺点。 2. **数据预处理技术**：处理停用词、空格、标点符号等文本清理技术。 3. **使用SpaCy库**：探索NLP管道，以及更高级的概念，例如命名实体识别和句法依赖，这些技术使得代码能够自动理解文本信息中的概念，比如钱、时间、公司、产品和地点等。 4. **词性标注**：让代码自动为文本中的单词分配相应的词性，是构建智能语言系统的重要部分。 5. **文本向量化**：学习如何将文本转化为计算机能理解的格式，介绍两种常见的机制：计数向量化和TF-IDF向量化。 6. **文本分类**：运用机器学习进行自然语言处理，构建一个分类IMDb电影评论的模型，包括数据清理、预处理、特征工程、模型训练和测试，探索多种机器学习算法，如逻辑回归、朴素贝叶斯和线性支持向量机，提升模型表现应对实际NLP问题，如评论分类和垃圾邮件检测。 7. **情感分析**：学习如何使用内置的情感分析工具（例如TextBlob和VADER），并构建自己的情感分析器，包括预处理、特征工程、训练和测试。 8. **整合Twitter API**：学习如何使用Twitter的数据，Twitter在文本数据方面是最强大的社交媒体平台，一些投资者和银行已在使用Twitter数据理解市场情绪。 自然语言处理成为技术行业中日益重要的技能集。本课程将帮助您开启自己的NLP之旅。还等什么？今天就开始您的NLP专家之路吧！所有课程提供30天退款保证，所以您可以无风险尝试。我在课程中等你！
Welcome!This course is carefully designed for you to learn the fundamentals of Natural Language Processing and then to advance gradually and to solve complex NLP problems using Machine Learning. Everything taught in this course is completely hands on. So you will be able to learn things by doing them yourself. You don't need prior experience in Natural Language Processing, Machine Learning or even Python. But you should be comfortable with programming, and should be familiar with at least one programming language. Python is by far one of the best programming language to work on Machine Learning problems and it applies here as well. If you're new to Python, don't worry, I'll explain what you need to know, just before using it.In this course, we use Google Colab to run our code. So, you don't have to install or configure anything in your machine. It doesn't matter what's your OS or hardware spec, as long as you have access to the Internet. But if you're interested, the same code can be run on Jupyter Notebook, installed in your machine.First we will explore the basic concepts of Natural Language Processing, such as tokenization, stemming and lemmatization using NLTK. You will learn more than one way to get these things done, so you can understand the pros and cons of different approaches. Then we will study some pre-processing techniques for removing stop-words, whitespaces, punctuations, symbols, new lines, etc.Next we will move to SpaCy - a state of the art NLP library heavily used in the industry. We will explore the NLP pipeline, and more advanced concepts such as Named Entity Recognition and Syntactic Dependencies. These techniques allow your code to automatically understand concepts like money, time, companies, products, locations, and many more simply by analysing the text information.There we will cover Part-of-Speech tagging as well, where your code will be able to automatically assign words in text to their appropriate part of speech, such as nouns, verbs, adverbs and adjectives, an essential part of building intelligent language systems.After that, you will learn how to transform text into a format where the computer can understand. This process is called vectorization. There're more than one way to do this, and you will learn the two most common mechanisms. Count vectorization and TF-IDF vectorization. Next, we will move to Text Classification, where we will start using Machine Learning for Natural Language Processing. We will build a fully functioning model to classify IMDb movie reviews. There you will learn how to perform data cleansing, pre-processing, feature engineering, model training and testing. We will try out few different machine learning algorithms from the scikit-learn library, such as Logistic Regression, Naive Bayes and Linear SVC, and we will explore how to improve the performance on each case. You will be able to use the learnings from this section to address real world NLP problems, such as review classifications or spam detection.Then we will move to one of the most demanding areas of Natural Language Processing, which is Sentiment Analysis. First we will explore how to use some built-in sentiment analysis tools such as TextBlob and VADER. Then we will start building our own Sentiment Analyzer using Logistic Regression and Naive Bayes. There we will go through all the steps required to build a sentiment analyser from the scratch, including pre-processing, feature engineering, training and testing. Finally we will complete this course by learning how to integrate Twitter's APIs to pull Twitter data. Twitter is by far the strongest social media when it comes to text data. Some investors, banks and hedge funds are already using Twitter data to understand the market sentiment. So why not learn how to use this valuable resource, as the data source for your NLP problem. Natural Language Processing is becoming one of the highly demanding skillset in the technology industry, and this course will help you to start your NLP journey. What are you waiting for? Start your journey to become an expert in NLP today!All of this comes with a 30 day money back garuantee, so you can try the course risk free.I will see you inside the course.
