|
所在平台: Coursera |
课程主页: https://www.coursera.org/learn/data-warehousing-business-intelligence
课程评论：没有评论
课程名称：数据仓库与商业智能 课程概述：本课程在“数据的性质与关系数据库设计”基础上，进一步拓展数据的捕获与处理过程，通过数据仓库和数据挖掘进行深化。一旦事务数据经过ETL（提取、转换、加载）处理后，就会存储在数据仓库中，以供管理决策使用。数据挖掘是将存储在数据仓库中的数据转化为可操作见解的关键工具，从而实现更快速、更有效的决策制定。 课程结束时，学生将能够解释数据仓库及其在商业智能中的应用，阐述不同的数据仓库架构和多维数据建模，并开发预测性数据挖掘模型，包括分类和估算模型。此外，学生还将能够开发解释性数据挖掘模型，包括聚类和关联模型。 课程大纲： 1. 数据仓库概述 - 介绍数据仓库及其架构，定义ETL过程，涉及云端数据仓库并通过简短测验进行实践。活动中，学生将区分Kimball和Inmon的数据仓库架构设计方法。 2. 数据仓库的多维建模 - 探讨数据仓库的数据建模，学习构建多维数据模型的步骤，区分星型模式与雪花模式，并通过简短测验进行实践。最后，学生在活动中创建一个规范的雪花模式。 3. 预测与解释的数据挖掘 - 概述数据挖掘过程及其方法，识别数据挖掘过程中的步骤，区分不同的数据挖掘方法，并通过简短测验进行实践。活动中，学生将选择适合特定数据集的数据挖掘方法。 4. 聚类与关联的数据挖掘 - 探讨用于解释建模的无监督数据挖掘，学习聚类与分割、K均值聚类、关联及市场篮分析的定义，并通过简短测验进行实践。在活动中，学生将练习在数据集中识别聚类。
Name:Overview of Data Warehousing
Description:Welcome to Module 1, Overview of Data Warehousing. In this module, we will overview data warehousing and data warehousing architectures. We will also define the Extract, Transform, Load (ETL) process as well as touch on data warehousing in the cloud and practice these through a short quiz. Finally, in our activity we will differentiate between the Kimball and Inmon design approaches for data warehouse architecture.
Name:Multidimensional Modeling for Data Warehousing
Description:Welcome to Module 2, Multidimensional Modeling for Data Warehousing. In this module, we will go over data modeling for data warehousing. We will also learn the steps needed to construct a multidimensional data model and differentiate between star schema and snowflake schema. These will be practiced through a short quiz. Finally, we will create a normalized snowflake schema in our activity.
Name:Data Mining for Prediction and Explanation
Description:Welcome to Module 3, Data Mining for Prediction and Explanation. In this module, we will overview the data mining process and data mining methods. We will also identify the steps in a data mining process and differentiate between data mining methods. We will practice identifying these through a short quiz. In our activity, we will also select what data mining methods are best for a particular data set.
Name:Data Mining for Clustering and Association
Description:Welcome to Module 4, Data Mining for Clustering and Association. In this module, we will go over unsupervised data mining for explanatory modeling. We will also learn the definitions for clustering and segmentation, K-means clustering, association, and market basket analysis and practice these through a short quiz. Finally we will practice identifying clusters in a dataset through our activity.
This course builds on “The Nature of Data and Relational Database Design” to extend the process of capturing and manipulating data through data warehousing and data mining. Once the transactional data is processed through ETL (Extract, Transform, Load), it is then stored in a data warehouse for use in managerial decision making. Data mining is one of the key enablers in the process of converting data stored in a data warehouse into actionable insight for better and faster decision making. By the end of this course, students will be able to explain data warehousing and how it is used for business intelligence, explain different data warehousing architectures and multidimensional data modeling, and develop predictive data mining models, including classification and estimation models. IN addition, students will be able to develop explanatory data mining models, including clustering and association models.
