ADVANCED DATA MINING

AIM

To recognize the basic concepts and methods of data mining, interpret and apply common data mining methods including clustering and classification, design a data mining model for a given problem, and apply, and interpret the main clinical and managerial decision support systems in healthcare.

CONTENT

This course contains; Introduction to Data Mining,Data Mining Process,Data Discovery and Visualization,Feature Selection and Data Transformation,Clustering Methods,Exercise: Clustering ,Classification Methods - Decision Trees,Exercise: Classifications,Exercise: Classifications,Association Rule Mining,Exercise: Association rule mining ,Exercise: Problem oriented data mining,Midterm project presentations,Midterm project presentations.

1. 1

   Use the association rules with KNIME

   Taught by: Discussion Method, Question - Answer Technique, Lecture Method

2. 2

   Use the classification methods with KNIME

   Taught by: Discussion Method · Assessed by: Oral Exam

3. 3

   Use the clustering methods with KNIME

   Taught by: Demonstration Method, Case Study Method, Concept Map Technique, Experiential Learning, Lecture Method

4. 4

   Apply the data preprocess method with KNIME

   Taught by: Discussion Method, Problem Baded Learning Model, Experiential Learning, Lecture Method · Assessed by: Project Task

5. 5

   Use an open source data mining tool (KNIME)

   Taught by: Self Study Method, Question - Answer Technique, Project Based Learning Model, Inquiry-Based Learning, Cooperative Learning, Experiential Learning, Flipped Classroom Learning, Lecture Method · Assessed by: Homework, Project Task

6. 6

   Explain the data mining and its sub-processes

   Taught by: Discussion Method, Demonstration Method, Lecture Method · Assessed by: Oral Exam

7. 7

   Explain the learning-based DSSs

   Taught by: Experiential Learning, Lecture Method

8. 8

   Explain the knowledge-based DSSs

   Taught by: Lecture Method

9. 9

   Explain the clinical and management decision support systems.

   Taught by: Self Study Method, Micro Teaching Technique, Inquiry-Based Learning, Cooperative Learning, Experiential Learning, Lecture Method · Assessed by: Traditional Written Exam

10. 10

    Interpret the most possible methods for a given data set with respect to the data types and pattern

    Taught by: Discussion Method, Case Study Method, Question - Answer Technique, Project Based Learning Model, Inquiry-Based Learning, Lecture Method

11. 11

    Propose a correct (supervised/unsupervised) method for a given data mining problem

    Taught by: Case Study Method, Brainstorming Technique, Experiential Learning, Lecture Method

12. 12

    Distinguish the differences between descriptive and predictive methods

    Taught by: Question - Answer Technique, Lecture Method · Assessed by: Performance Task

13. 13

    Distinguish the appropriate method for a given data mining problem.

    Taught by: Discussion Method, Case Study Method, Question - Answer Technique, Inquiry-Based Learning, Experiential Learning, Lecture Method · Assessed by: Traditional Written Exam

14. 14

    Explain classification methods

    Taught by: Question - Answer Technique, Lecture Method

15. 15

    Explain clustering methods

    Taught by: Discussion Method, Question - Answer Technique, Lecture Method

16. 16

    Explain common data mining methods

    Taught by: Case Study Method, Self Study Method, Question - Answer Technique, Micro Teaching Technique, Experiential Learning, Flipped Classroom Learning, Lecture Method · Assessed by: Traditional Written Exam

17. 17

    Describe the data mining process

    Taught by: Discussion Method, Case Study Method, Self Study Method, Lecture Method · Assessed by: Homework

18. 18

    Explain basic concepts, including data, database, data wearehouse, etc

    Taught by: Lecture Method

19. 19

    Define data mining and its objectives

    Taught by: Discussion Method, Question - Answer Technique, Lecture Method

20. 20

    Explain the basic concepts and processes in data mining

    Taught by: Question - Answer Technique, Micro Teaching Technique, Lecture Method · Assessed by: Traditional Written Exam

WEEKLY PLAN

1. WEEK 1

   Introduction to Data Mining

   Preparation: Basic database concepts

2. WEEK 2

   Data Mining Process

   Preparation: The usage ofSQL as DML; TSQL; data warehouse architectures, the reasons yield data manipulation

3. WEEK 3

   Data Discovery and Visualization

   Preparation: The graph types in data visualization and the components of a graph, such as dimension, measure, etc.

4. WEEK 4

   Feature Selection and Data Transformation

   Preparation: Data types, generalization, specialization of data

5. WEEK 5

   Clustering Methods

   Preparation: Main clustering approaches, such as Hierachical Clustering, Centroid-based Clustering, Density-based Clustering, Distribution-based Clustering.

6. WEEK 6

   Exercise: Clustering

   Preparation: The excercises with KNIME for especially clustering

7. WEEK 7

   Classification Methods - Decision Trees

   Preparation: The main differences between clustering and clalssifications, how to generate a decision tree and the main decision tree algorithms, such as ID 3 and C4.5

8. WEEK 8

   Exercise: Classifications

   Preparation: The excercises with KNIME for especially classification

9. WEEK 9

   Exercise: Classifications

   Preparation: The excercises with KNIME for especially classification

10. WEEK 10

    Association Rule Mining

    Preparation: Market box analysis

11. WEEK 11

    Exercise: Association rule mining

    Preparation: The excercises with KNIME for especially ARM

12. WEEK 12

    Exercise: Problem oriented data mining

    Preparation: Clustering, decision trees, association rule mining

13. WEEK 13

    Midterm project presentations

    Preparation: A data mining solution having feature selection, data transformation, data mining application and evaluation of the result

14. WEEK 14

    Midterm project presentations

    Preparation: A data mining solution having feature selection, data transformation, data mining application and evaluation of the result

ASSESSMENT

• Rate of Midterm Exam to Success50%
• Rate of Final Exam to Success50%

WORKLOAD

READING

• Lecture notes and lab sheets (will be shared regularly on the lecture pages) - Veri Madenciliği Teori Uygulama ve Felsefesi, Dr. İlker KÖSE (2015) - Kavram ve Algoritmalarıyla Temel Veri Madenciliği, Dr. Gökhan SİLAHTAROĞLU - Veri Madenciliği Yöntemleri, Dr. Yalçın ÖZKAN - Han Jiawei and Kamber Micheline (2006), Data Mining: Concepts and Techniques, Morgan Kaufmann Publisher San Francisco - Pang-Ning Tan, Michael Steinbach, Vipin Kumar, Introduction to Data Mining, Addison Wesley, (2005)

TEACHING STAFF

• Assist.Prof. Kevser Banu KÖSECOORDINATOR
• Assist.Prof. Kevser Banu KÖSE