Topic outline

  • ANNOUNCEMENTS

  • Course Introduction

    Welcome to data mining course!


    Instructions/Guideline for the course:

        • All the students registered for this course have to enroll in Moodle
        • Students can find all the course materials from Moodle.
        • One discussion or feedback forum is created under each of the lecture Students have to give their feedback on these forum and marks will be given for their feedback
        • Any announcement regarding the class will be posted on Moodle. So they have to keep themselves always active on Moodle.
        • All the quizzes and presentation will be held on face to face class and may be few of the class will be held on online (Moodle) and it will be announced before the class
        • The question pattern and the syllabus for the quizzes, midterm and final exam is given here under each of the section (quizzes, midterm and final)

        • There are midterm and final exam preparation forum under these sections where students can discuss with each other about their midterm and final exam syllabus, any problem regarding the exam etc.

    Course Objectives:

    This course introduces students to the applications of data mining, different stages in data mining and different data mining algorithms. Upon completion of this course, students will be able to do the following:

    (1) Appreciate the necessity of data mining in everyday life
    (2) Apply the concept of data mining in solving problems
    (3) Demonstrate applications of data mining using tools
    (4) Apply knowledge of data mining in project work


    Course Contents:

    Introduction and basics of data mining
    - Data pre-processing
    - Classification techniques, decision tree
    - Clustering algorithms
    - Association rules mining

  • Introduction to data mining

    Objectives of this Chapter: To teach the importance of data & data mining in various field of science, technology & business.

    Chapter Outcomes: At the end of this chapter, students will be able to :

    -        Appreciate the needs of data mining
    -        Provide examples of different types of data files and their volumes

    -        Visualization of different data mining tasks

    Chapter Contents:

    -        Introduction to data mining
    -        Why data mining is a discipline?
    -        Overview of data mining tasks: Clustering, Classifications, Rules learning etc

    -        Data characteristics

  • Data pre-processing

    Objectives of this Chapter: To teach the need to data pre-processing in order to produce quality data mining results

    Chapter Outcomes: At the end of this chapter, students will be able to :

    -        Explain issues associated with raw data
    -        Appreciate the need of solving the issues by various pre-processing methods

    -        Visualization of different data preprocessing methods

    Chapter Contents:

    -        Issues involved with raw data

    -        Various methods to solve the issues

  • Classification & prediction

    Objectives of this Chapter: To teach how the classification & prediction problem can be solved using data mining approach

    Chapter Outcomes: At the end of this chapter, students will be able to :

    -        Explain what is classification and the applications of classification algorithms in supervised learning
    -        Provide differences between bi-class & multi-class applications
    -        Perform prediction based on entropy & gini-indexed based decision tree

    -        Explain confusion matrix to know the accuracy of the prediction

    Chapter Contents:

    -        Classification based on entropy based decision tree
    -        Classification based on gini based decision tree

    -        Confusion matrix

  • Midterm Week

    Midterm exam includes introduction to data mining, data pre-processing, classification and prediction (decision tree)

  • Associate Rules Mining

    Objectives of this Chapter: To teach how association rules mining is useful for different data mining applications

    Chapter Outcomes: At the end of this chapter, students will be able to :

    -        Provide scenarios where association rules mining can be used
    -        Apply apriori algorithm on suitable datasets and generate meaningful rules
    -        Apply fp-growth & fp-tree algorithm on suitable datasets and generate meaningful rules

    -        Explain the strengths & weaknesses of association rules mining

    Chapter Contents:

    -        Basic concepts
    -        Apriori algorithm
    -        Different data formats for mining
    -        Mining with multiple minimum supports
    -        Mining class association rules

    -        FP-growth/FP-tree

  • Clustering algorithms

    Objectives of this Chapter: To teach clustering example as an example of unsupervised learning

    Chapter Outcomes: At the end of this chapter, students will be able to :

    -        Visualization on clustering and application of clustering

    Chapter Contents:

    -        Types of clustering
    -        Common distance measure

    -        K-means clustering with examples

  • Assignment

    Instructions:

    • Your Assignment will be taken as viva
    • It will be based on your basic understanding and short questions will be asked
    • Some samples of short questions are given in the "Course Introduction" section
    • Syllabus will be given in the class

    • Presentation

      Instructions:

      • In presentation, marks will be given on: getup & outfit, body language, English command, eye contact, content knowledge, Q/A
      • Try to prepare the slides with more figures and less text. The font size of text should be larger so that it can be seen from the back bench of your class
      • Give slide number
      • More instructions will be discussed in the class

    • Final-Exam Week

      Final Exam Includes: Association Rules Mining & Clustering Algorithm, Basics of data mining