|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/detect-fake-news-with-machine-learning-feature-engineering/
课程评论：没有评论
**Coursera课程：使用机器学习和特征工程检测虚假新闻** **课程概述：** 本课程是一门基于项目的综合性课程，将指导您一步步构建一个虚假新闻检测系统。课程结合了Python和机器学习，是提升数据科学技能的绝佳机会。我们将重点关注三个主要方面：数据分析（从多角度探索虚假新闻数据集）、预测建模（利用大数据构建虚假新闻检测系统）以及减轻虚假新闻检测模型的潜在偏见。 **课程内容亮点：** * **虚假新闻检测基础：** 了解虚假新闻检测模型的基本原理，包括伦理考量和常见挑战。 * **特征工程案例研究：** 学习如何在简单数据集上应用特征工程，预测新闻的真伪，例如识别虚假新闻中常用的特定词语，并根据新闻发布者的过往记录计算新闻为虚假的概率。 * **虚假新闻传播因素：** 探讨导致虚假新闻和错误信息广泛传播的因素，如确认偏见、社交媒体回音室和点击诱饵。 * **实战项目：** * **环境设置：** 指导您一步步设置Google Colab IDE。 * **数据集获取：** 学习如何从Kaggle查找和下载虚假新闻数据集。 * **数据分析与可视化：** 从不同角度探索和可视化数据集。 * **模型构建：** * 使用特征工程和逻辑回归构建虚假新闻检测系统。 * 使用随机森林模型。 * **模型评估：** 学习如何评估模型的准确性，包括使用混淆矩阵。 * **偏见缓解：** 学习如何通过多元化训练数据和进行公平性审计来减轻虚假新闻检测系统中的潜在偏见。 **为何学习构建虚假新闻检测系统？** 随着社交媒体的普及和在线信息量的爆炸式增长，未经核实、误导性或虚假新闻的传播日益严重，可能影响公众舆论、煽动恐惧，甚至带来政治和社会后果。掌握辨别准确信息与欺骗性内容的能力，在信息时代尤为宝贵。此外，构建复杂的机器学习模型也将为您带来更多职业发展机会。 **课程可期待的学习成果：** * 掌握虚假新闻检测模型的基础知识。 * 通过案例研究，学习应用特征工程预测新闻标题的真伪。 * 了解导致虚假新闻和错误信息传播的因素。 * 学会从Kaggle查找和下载数据集。 * 掌握数据清洗技巧，去除缺失值和重复值。 * 学习评估新闻来源的可靠性。 * 学会检测与虚假新闻相关的关键词。 * 能够进行新闻标题和长度分析。 * 学会检测虚假新闻中的煽动性内容。 * 利用NLP技术检测虚假新闻中的情绪。 * 掌握基于特征工程构建虚假新闻检测模型。 * 掌握基于逻辑回归构建虚假新闻检测模型。 * 掌握基于随机森林构建虚假新闻检测模型。 * 学会使用混淆矩阵评估虚假新闻检测模型。 * 学会通过公平性审计（如人口统计学差异）进行公平性评估。 * 学会缓解虚假新闻检测系统中的潜在偏见。
Welcome to Detecting Fake News with Machine Learning course. This is a comprehensive project based course where you will learn step by step on how to build a fake news detection system using feature engineering, logistic regression, and other models. This course is a perfect combination between Python and machine learning, making it an ideal opportunity to enhance your data science skills. The course will be mainly focusing on three major aspects, the first one is data analysis where you will explore the fake news dataset from multiple angles, the second one is predictive modeling where you will learn how to build fake news detection system using big data, and the third one is to mitigate potential biases from the fake news detection models. In the introduction session, you will learn the basic fundamentals of fake news detection models, such as getting to know ethical considerations and common challenges. Then, in the next session, we are going to have a case study where you will learn how to implement feature engineering on a simple dataset to predict if a news is real or fake. In the case study you will specifically learn how to identify the presence of specific words which are frequently used in fake news and calculate the probability of a news article is fake based on the track record of the news publisher. Afterward, you will also learn about several factors that contribute to the widespread of fake news & misinformation, for examples like confirmation bias, social media echo chamber, and clickbait incentives. Once you have learnt all necessary knowledge about the fake news detection model, we will begin the project. Firstly you will be guided step by step on how to set up Google Colab IDE. In addition to that, you will also learn how to find and download fake news dataset from Kaggle, Once, everything is ready, we will enter the main section of the course which is the project section The project will be consisted of three main parts, the first part is the data analysis and visualization where you will explore the dataset from various angles, in the second part, you will learn step by step on how to build a fake news detection system using logistic regression and feature engineering, meanwhile, in the third part, you will learn how to evaluate the model's accuracy. Lastly, at the end of the course, you will learn how to mitigate potential bias in fake news detection systems by diversifying training data and conducting fairness audits.First of all, before getting into the course, we need to ask ourselves this question: why should we build fake news detection systems? Well, here is my answer. In the past couple of years, we have witnessed a significant increase in the number of people using social media and, consequently, an exponential growth in the volume of news and information shared online. While this presents incredible opportunities for communication, however, this surge in information sharing has come at a cost, the rapid spread of unverified, misleading, or completely fabricated news stories. These stories can sway public opinion, incite fear, and even have political and social consequences. In a world where information is power, the ability to distinguish between accurate reporting and deceptive content is very valuable. Last but not least, knowing how to build a complex machine learning model can potentially open a lot of opportunities.Below are things that you can expect to learn from this course:Learn the basic fundamentals of fake news detection modelCase study: applying feature engineering to predict if a news title is real or fakeLearn factors that contribute to the widespread of fake news & misinformationLearn how to find and download datasets from KaggleLearn how to clean dataset by removing missing rows and duplicate valuesLearn how to perform news source credibilityLearn how to detect keywords associated with fake newsLearn how to perform news title and length analysisLearn how to detect sensationalism in fake newsLearn how to detect emotion in fake new with NLPLearn how to build fake news detection model with feature engineeringLearn how to build fake news detection model with logistic regressionLearn how to build fake news detection model with Random ForestLearn how to evaluate fake news detection model with confusion matrixLearn how to perform fairness audit with demographic parity differenceLearn how to mitigate potential bias in fake news detection
