|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/decision-tree-theory-application-and-modeling-using-r/
课程评论：没有评论
课程名称：决策树 - 理论、应用及使用R建模 课程概述：本课程旨在教授决策树模型的构建，这是分析领域中最常用的技术之一。决策树模型开发迅速且易于理解，学习难度低。许多商业场景，如贷款、通信和汽车等领域，都需要决策树模型的构建。通过本课程，学生将了解什么是决策树、决策树的应用场景、所带来的好处、决策树背后的各种算法、如何在R中开发决策树的步骤以及如何解读R的决策树输出。 课程标签：决策树、CHAID、CART、目标细分、预测分析、ID3、GINI。 课程内容：课程中提供高清视频，制作所用的演示文稿可下载为PDF格式，使用的Excel文件和R程序也可供下载。 课程时长：大约需要8小时来理解概念并熟悉使用R进行决策树建模。 课程结构： - 第1部分 - 动机与基本理解 - 理解需要分类结果的决策树的商业场景 - 查看示例决策树输出 - 理解决策树带来的收益 - 了解决策树与基于逻辑回归评分的不同 - 第2部分 - 实践（针对分类输出） - 安装R的过程 - 安装R Studio的过程 - 理解R Studio/包/库 - 在R中开发决策树，深入分析输出 - 第3部分 - 决策树背后的算法 - 节点的GINI指数 - 分裂的GINI指数 - 变量和分裂点选择程序 - 实施CART模型 - 数据挖掘场景中的决策树开发和验证 - 自动修剪技术 - 理解R中自动修剪的程序 - 理解CHAID与CART的区别 - 理解CART针对数值结果的应用 - 解读与CART相关的R平方意义 - 第4部分 - 决策树的其他算法 - ID3 - 节点的熵 - 分裂的熵 - 随机森林方法 为什么选择本课程？通过本课程，可以全面理解决策树建模，熟练掌握使用R进行决策树开发，实际操作R包的输出，并理解决策树的应用场景。
What is this course? Decision Tree Model building is one of the most applied technique in analytics vertical. The decision tree model is quick to develop and easy to understand. The technique is simple to learn. A number of business scenarios in lending business / telecom / automobile etc. require decision tree model building. This course ensures that student get understanding of what is the decision tree where do you apply decision tree what benefit it brings what are various algorithm behind decision tree what are the steps to develop decision tree in R how to interpret the decision tree output of R Course Tags Decision Tree CHAID CART Objective segmentation Predictive analytics ID3 GINI Material in this course the videos are in HD format the presentation used to create video are available to download in PDF format the excel files used is available to download the R program used is also available to download How long the course should take? It should take approximately 8 hours to internalize the concepts and become comfortable with the decision tree modeling using R The structure of the course Section 1 - motivation and basic understanding Understand the business scenario, where decision tree for categorical outcome is required See a sample decision tree - output Understand the gains obtained from the decision tree Understand how it is different from logistic regression based scoring Section 2 - practical (for categorical output) Install R - process Install R studio - process Little understanding of R studio /Package / library Develop a decision tree in R Delve into the output Section 3 - Algorithm behind decision tree GINI Index of a node GINI Index of a split Variable and split point selection procedure Implementing CART Decision tree development and validation in data mining scenario Auto pruning technique Understand R procedure for auto pruning Understand difference between CHAID and CART Understand the CART for numeric outcome Interpret the R-square meaning associated with CART Section 4 - Other algorithm for decision tree ID3 Entropy of a node Entropy of a split Random Forest Method Why take this course? Take this course to Become crystal clear with decision tree modeling Become comfortable with decision tree development using R Hands on with R package output Understand the practical usage of decision tree
