MGMT 47300

Data Mining
Matthew Lanham - Online - Summer - 2019

About Course Insights
Course Insights is a platform for faculty to share detailed information about the courses they teach. This page is not to be interpreted as being the official course syllabus for this course.
About the Course

This Data Mining course will show you how to follow a structured analytical process to understand and identify relationships within data, and build supervised and unsupervised learning models to support business problems. Data science is an exciting field and in high demand, but the work can be cruel (e.g. dirty data, big data, misleading trends, non-replicability, poor model performance, etc.). The sooner you realize this by getting your hands dirty in the data and software, the better prepared you will be when you are tasked to deliver data-driven results in practice.

To perform data mining/science successfully requires the ability to pull, summarize, visualize, explore, and model relationships among many variables, which should help you begin to formulate a story about what the data is suggesting, and possibly provide analyses that support decision-making in your role post-Purdue.

This course is designed for Krannert School of Management students, but those in other schools may take this course as well, provided they have a introductory statistics or business analytics course. Assignments are both conceptual and hands-on exercises.

Course Goals/Learning Objectives
  1. Ability to follow a structured analytical process for framing and solving business problems.

  2. Ability to use JMP Pro to analyze data, build models, and properly evaluate them.

  3. Ability to perform EDA, data visualization, data preparation, and feature selection.

  4. Understanding and hands-on usage of popular prediction/estimation algorithms.

  5. Understanding and hands-on usage of popular clustering algorithms

  6. Understanding of cross-validation designs, modeling tradeoffs, model evaluation approaches, and consideration of statistical and business performance measures.

MGMT 47300

Course Catalog
MGMT 47300 Data Mining

Description
Credit Hours: 3.00. Students follow a structured analytical process using popular industry tools (e.g., RStudio, Tableau, SQL) to identify, visualize, and summarize relationships within large data sets to support business problems. More focus is on descriptive analytics which includes business segmentation and clustering methods, but also introduces two predictive analytic methods, decision trees, and neural networks. Typically offered Fall.
3.000 Credit hours
Levels: Graduate, Professional, Undergraduate
Schedule Types: Distance Learning, Lecture
Offered By: School of Management
Department: School of Mgmt Adm & Instr
Course Attributes
Upper Division
May be offered at any of the following campuses: West Lafayette
Learning Objectives
1. Demonstrate working knowledge of relationship databases, SQL queries, and efficient sampling. 2. Demonstrate working knowledge of RStudio to perform EDA, visualization, and data mining techniques. 3. Demonstrate working knowledge of data visualization and exploration using Tableau. 4. Demonstrate practical understanding of popular unsupervised and supervised learning methods.
Prerequisites
Undergraduate level MGMT 30500 Minimum Grade of C- or Undergraduate level STAT 35000 Minimum Grade of C- or Undergraduate level STAT 41600 Minimum Grade of C- or Undergraduate level STAT 50100 Minimum Grade of C- or Undergraduate level STAT 51100 Minimum Grade of C-
Other Information
Restrictions: Must be enrolled in one of the following Programs: Management-BSIM Management-BS Accounting-BS Economics-BS
Having trouble? Self Help Knowledgebase | For assisted support: itap@purdue.edu or 765-494-4000
Purdue University is an equal access/equal opportunity university. If you have trouble accessing
this page because of a disability, please contact ITaP Customer Service at itap@purdue.edu.