Program Overview

Business Intelligence from Web Data Analytics and Data Mining using R and AI

Overview

The Business Intelligence from Web Data Analytics and Data Mining using R and AI program is designed to introduce students to the rapidly evolving area of business intelligence. This course provides an introduction to the business intelligence process, including data collection, data exploration, data mining, data analysis, and model evaluation.

Program Details

• ECTS: 5
• Level: Bachelor
• Location: Aarhus
• Type of course: Summer University
• Primary programme: Bachelor's Degree Programme in Economics and Business Administration
• Related programmes: Bachelor's Degree Programme in Economics, Bachelor's Degree Programme in Global Management and Manufacturing
• Department: Department of Economics and Business Economics
• Faculty: Aarhus BSS
• Maximum number of participants: 40 (10 seats reserved for international exchange students from AU partner universities)

Course Content

The course focuses on building the skills and competence of students by covering topics such as:

• Data collection
• Data Preparation/Data Preprocessing
• Exploratory Data Analysis
• Single Regression/Multiple Regression
• K-Nearest Neighbor
• Decision Trees
• Neural Networks
• Clustering
• Handling Missing Data
• Evaluating Different Models
• Using AI programming tools

Description of Qualifications

Upon successful completion of this course, students will develop a broad appreciation for and a basic understanding of different aspects of Business Intelligence. These include:

• Knowledge:
  • The overall framework of the business intelligence process using the CRISP-DM process.
  • The range of activities associated with the business intelligence process.
  • Security and privacy issues in the business intelligence process.
  • Evaluate the different data analysis models.
  • Understand how AI tools are changing the landscape of the programming world.
• Skills:
  • The data collection process from web-based data collection, open data collection, to automated data collection.
  • The model evaluation process, including error rates, false positives, false negatives, sensitivity, specificity, gain charts, and lift charts.
  • Run data collection using data collection tools.
  • Run data analysis using tools to aid in data analysis techniques such as data mining.
• Competences:
  • The nature of common data collection and analysis hazards.
  • The basic operation and limitations of business intelligence.
  • Recognize and analyze the business intelligence process.
  • Understand the limitations of business intelligence.
  • Garner useful, actionable information from the data presented.
  • Recognize the limitations of AI tools, when to use AI, and when human programming is needed.

Teaching

• Form of instruction: Classroom instruction
• Instructor: Hirotoshi Takeda
• Course coordinator: Ana Alina Tudoran
• The course will include a mix of ordinary classroom lectures, case-based teaching, break-out sessions involving group exercises, and computer demonstrations using data mining and data analysis tools.

Examination

• Form of examination: Take-home assignment (Assign)
• Form of co-examination: No co-examination
• Assessment: 7-point grading scale
• Permitted exam aids: All
• Duration: 4 hours

Requirements for Taking the Exam

• Attendance requirement: 80% attendance is required to participate in the exam.
• Compulsory activities:
  • Activity 1: Find an Open Data Set that can be used for Data Mining. Each student will need to search for a dataset and identify a dataset. Length: maximum 1,200 characters. (Individual Activity)
  • Activity 2: Run Exploratory Data Analysis on the Dataset from Activity 1. The groups will choose one dataset to use for their group activity (activity 2 and 3). This can be one of the students' dataset found in Activity 1 or a new dataset. Groups will run Exploratory Data Analysis on their chosen dataset. Length: maximum 12,000 characters. (Group Activity)
  • Activity 3: Run various data mining methodologies on the Dataset from Activity 1 and continue analysis from Activity 2. The groups run three of the following data mining methods covered in class on their data: Regression, K-Nearest Neighbor Analysis, Decision Tree Analysis, Neural Networks, or Clustering. Length: maximum 24,000 characters. (Group Activity)

Expected Student Workload

• Classroom attendance: 52 hours
• Preparation: 75 hours
• Feedback activity: 5 hours
• Papers (prerequisites): 15 hours
• Exam: 10 hours

Literature

• Daniel T. Larose (2014). Discovering Knowledge in Data: An Introduction to Data Mining, 2nd Edition, Wiley, ISBN: 978-
• Paul Torfs & Claudia Brauer (2014) “A (very) short introduction to R
• Jonathan Baron “R reference card”

Re-exam

• Re-exam: Written take-home exam (max. 36,000 characters including spaces).
• Dates for the first retake: 27th October, 12:00 noon: You will receive your exam question via WISEflow. 3rd November, 12:00 noon: Deadline for submitting via WISEflow.
• Dates for the second retake: January 30th, 12:00 noon: You will receive your exam question via WISEflow. February 6th, 12:00 noon: Deadline for submitting via WISEflow.