Overview

Description

CALENDAR DESCRIPTION:

Current topics in database and information systems depending on faculty and student interest.

COURSE DETAILS:

This course introduces Data Mining, an area that plays a key role in Big Data analytics. Data mining applies machine learning methods for the efficient discovery of useful patterns in large datasets. This course focuses on fundamental data mining tasks and algorithms as well as key applications. It will prepare you both for developing your own data mining application and for starting your data mining research. Students taking this course are expected to have taken an algorithms course and to have an understanding of basic statistics equivalent to an entry-level course. The programming assignments and the course project require programming in Python, and students are expected to be proficient with this programming language. In the course project, groups of students will go through all steps of a data mining project on a dataset of their own choice.

Topics

• Introduction
• Data preprocessing: data cleaning, completion, transformation, normalization
• Classification: evaluation, decision trees, Bayesian classification, NN, SVM, ensemble methods
• Cluster analysis: partitioning, hierarchical, density-based methods, subspace clustering
• Outlier detection: probabilistic and distance-based methods, LOF, non-parametric methods
• Frequent pattern mining: association rules, Apriori, FP-growth, pattern summarization
• Impact of data mining
• Optional - Research directions in data mining: deep neural networks, transfer learning and continual learning, causal discovery and causal inference

Grading

Materials

REQUIRED READING:

REQUIRED READING NOTES: