Computer Science (COMP) 682

Overview

Our ability to generate and collect data has been increasing rapidly. The widespread use of information technology in our lives has flooded us with a tremendous amount of data. This explosive growth of stored and transient data has generated an urgent need for new techniques and automated tools that can assist us in transforming this data into useful information and knowledge. Data Mining has emerged as a multidisciplinary field that addresses this issue.

This course discusses the techniques for preprocessing data before mining, and defines the concepts related to data warehousing, on-line analytical processing (OLAP), and data generalization. It presents methods for mining frequent patterns, associations, and correlations. It also describes methods for data classification and prediction, and data-clustering approaches.

Course Objectives

• Interpret the contribution of data warehousing and data mining to the decision support level of organizations;
• Evaluate different models used for OLAP and data pre-processing;
• Categorize and carefully differentiate between situations for applying different data mining techniques: mining frequent pattern, association, correlation, classification, prediction, and cluster analysis;
• Design and implement systems for data mining;
• Evaluate the performance of different data mining algorithms;
• Propose data mining solutions for different applications.

Outline

Unit 1: Overview of Data Mining

• This unit provides some background on data objects and statistical concepts. It also discusses the type of data to be mined and presents a general classification of data-mining tasks.

Unit 2: Data Preprocessing

• This unit introduces techniques for preprocessing data before mining. Concepts such as data cleaning, data integration, data reduction, data transformation, and data discretization are discussed.

Unit 3: Overview of Data Warehousing and OLAP

• This unit provides a solid introduction to data warehousing, OLAP, and data generalization.

Unit 4: Data Cube Computation and multidimensional data analysis

• This unit presents a detailed study of methods for data cube computation, advanced queries processing and multidimensional data analysis.

Unit 5: Mining Frequent Patterns, Associations, and Correlations

• This unit presents methods for mining frequent patterns, associations, and correlations.

Unit 6: Classification

• This unit discusses methods such as decision tree induction, Bayesian classification, rule-based classification, neural networks, support vector machines, associative classification, k-nearest neighbor classifier, case-based reasoning, genetic algorithms, rough sets, and fuzzy set approaches.

Unit 7: Cluster Analysis

• This unit describes methods for data clustering. We discuss several major data-clustering approaches, such as partitioning methods, hierarchical methods, density-based methods, grid-based methods, and model-based methods.

Unit 8: Outlier detection

• This unit describes methods for anomaly detection. We discuss several major approaches, such as statistical approaches, proximity-based approaches,clustering-based approaches, and classification-based approaches.

Evaluation

In order to receive credit for COMP 682, you must achieve a cumulative course grade of "B-" (70 percent) or better, and must achieve an average grade of at least 60 percent on the assignments and project and achieve a grade of at least 60 percent on the Final Examination. Your cumulative course grade will be based on the following assessment.

The weighting of the composite grade is as follows: