ADV 08 – DATA MINING

First Semester, SY 2023 - 2024

I. Course Description: This course provides an in-depth exploration of the fundamental concepts, techniques,
and applications of data mining. Students will gain a comprehensive understanding of
how data mining plays a crucial role in various domains, the process involved, and the
ethical considerations that arise in the context of data mining.

II. Course Objectives:

By the conclusion of this course, students will be capable of:
1. Comprehending the fundamental principles and significance of data mining
within the context of Information Technology.
2. Preparing and refining raw data to render it suitable for mining operations.
3. Implementing diverse data mining techniques, including classification,
clustering, and association analysis.
4. Analyzing and deriving meaning from the results of data mining algorithms.
5. Demonstrating proficiency in utilizing data mining tools and software
effectively.
6. Recognizing ethical implications and privacy concerns inherent to data
mining procedures.
7. Applying data mining techniques to real-world IT scenarios, and making
informed decisions.
8. Communicating effectively regarding data mining principles and findings.

III. Course Outline and Time Frame

Week Topic
1 Aldersgate College vision & mission, core values, institutional goals & objectives
Program vision & mission, educational objectives & program outcomes
Course description & outcomes, course outline
2 Importance and Applications of Data Mining
Topic 1: Importance of Data Mining
 Understanding the value of extracting knowledge from large datasets.
 Identifying domains where data mining has made significant impacts.
Topic 2: Applications of Data Mining
 Exploring real-world applications such as business, healthcare, finance, and more.
 Highlighting case studies to illustrate the practical utility of data mining.
Topic 3: Data Mining Process and Tasks
 Step-by-step overview of the data mining process: from data collection to
interpretation.
 Explanation of various data mining tasks: classification, clustering, association,
and more.
3 Data Preprocessing
Topic 1: Data Cleaning and Transformation
 Identifying and addressing inconsistencies, errors, and outliers in data.
 Techniques for transforming raw data into suitable formats for analysis.
Topic 2: Handling Missing Data
 Strategies for dealing with missing values, including imputation and deletion.
 Understanding the impact of missing data on analysis outcomes.
Topic 3: Data Integration and Reduction
 Techniques to merge data from multiple sources while resolving conflicts.
 Methods for dimensionality reduction to improve computational efficiency.
4 Classification Techniques
Topic 1: Decision Trees and Rules
 Exploring decision tree construction, pruning, and interpretation.
 Rule based classification and its application in generating understandable models.
Topic 2: Naïve Bayes Classification
 Understanding the probabilistic foundation of Naïve Bayes.
 Application of Naïve Bayes in text classification and other domains.
 Topic 3: Support Vector Machines
 Concept of hyper plane-based classification and maximum margin.
 SVM kernel trick and handling nonlinearly separable data.

5 Clustering Techniques
Topic 1: Kmeans Clustering
 Understanding the Kmeans algorithm and its convergence properties.
 Practical considerations and initialization methods.
Topic 2: Hierarchical Clustering
 Exploring agglomerative and divisive hierarchical clustering.
 Dendrogram interpretation and linkage criteria.
Topic 3: Density based Clustering
 Introduction to DBSCAN and its density-based clustering approach.
 Handling noise and detecting clusters of arbitrary shapes.
6 Association Analysis
Topic 1: Market Basket Analysis
 Uncovering associations and patterns in transactional data.
 Basket analysis applications in retail and ecommerce.
Topic 2: Apriori Algorithm
 Working principles of the Apriori algorithm for frequent itemset mining.
 Handling large item-sets and candidate generation.
Topic 3: Sequential Pattern Mining
 Mining sequential patterns in timeseries and sequence data.
 Applications in recommendation systems and web usage analysis.
7 Ethics in Data Mining
Topic 1: Privacy Concerns and Anonymization
 Addressing privacy challenges in data mining, including deidentification
techniques.
 Implications of reidentification attacks and privacy-preserving methods.
Topic 2: Bias and Fairness in Data Mining
 Identifying sources of bias in data and algorithms.
 Strategies to mitigate bias and ensure fairness in analysis outcomes.
Topic 3: Legal and Ethical Considerations
 Overview of regulations and laws governing data mining and privacy.
 Ethical responsibilities of data miners and the implications of data misuse.
8 Real-world Applications
Topic 1: Customer Segmentation
 Using data mining to segment customers based on behavior and preferences.
 Customized marketing strategies and customer relationship management.
Topic 2: Fraud Detection
 Application of data mining in identifying fraudulent activities.
 Techniques for anomaly detection and fraud prevention.
Topic 3: Healthcare Analytics
 Leveraging data mining to extract insights from medical and healthcare data.
 Predictive modeling for disease diagnosis and patient outcomes.
9 Communication Skills
Topic 1: Presenting Data Mining Results
 Effective communication strategies for presenting complex findings.
 Tailoring presentations to different audiences.
Topic 2: Effective Data Visualization
 Principles of data visualization for enhancing understanding.
 Choosing appropriate visualization techniques for different types of data.
Topic 3: Interpreting Technical Findings
 Translating technical data mining outcomes into actionable insights.
 Collaborating with domain experts to interpret results accurately.

IV. Required Reading

V. Suggested Readings and References

VI. Grading System

1. Averaging-cumulative
2. Components
Quizzes (AQ) 20%
Periodic Examinations (PE) 30%
Requirements (Req) 50%
Total 100%
3. Computation
AQ+ PE+ REQ
Prelim Grade( PG)=
3

( AQ+ PE+ REQ ) 2+ PG

Midterm Grade ( MG )=
3

( AQ+ PE+ REQ ) 2+ MG

Final Grade ( FG ) =
3
VII. Classroom Policies

VIII. Consultation Hours: