CS 593: Knowledge Discovery in Databases
Stevens Institute of Technology
Khasha Dehnad
kdehnad@stevens.edu Khasha.dehnad@aimsinfo.com
Spring 2013
1
Course Requirements
Recommended Prerequisites:
Familiarity with the principals of statistics and probabilities and Data Mining; for example, completion of MGT 502 (no credit).
Optional Hardware and Software:
Lap top with internet access and ability to install software (admin rights).
Students will be installing SAS on their computers
Books, Notes, and Manuals:
• Data Mining, Methods and Models, D. T. Larose, Wiley– Interscience, Latest Edition
• Mining of Massive Datasets, A. Rajaraman, J.D Ullman, Stanford University, Cambridge University Press, Third Edition
• Lecture Notes and Handouts
• Real world projects and case studies
2
Text books
3
Course Overview
Big Data refers to data sets whose volume (amount of data collected, number of data sources), velocity (rate at which data is collected) and variety (heterogeneity of data and data sources) are so extreme that advanced Data Mining Algorithms are needed to process and discover useful patterns in data for actionable intelligent decisions, in a reasonable amount of time. The purpose of this course is to introduce theoretical as well as practical aspects of advanced, as well as, well established algorithms for mining massive datasets. Topics include: Naïve Bayes & Bayesian Networks , Stream Data Mining, Big Data Definition,Dimension Reduction techniques e.g. Principal Component Analysis (PCA), and recommendation systems.
4
Course Schedule
Introduction
Linear Algebra Review Intro to SAS
Week 1 Week 2 Week 3 Week 4
Week 5
Intro to SAS (continued) and Basic Statistics Review,
Principal Component Analysis and Factor Analysis
Introduction to Big Data , Massive Data sets Map-Reduce,
Relational Algebra in Big Data environment
in Databases I
5
Course Schedule
Big Data , Massive Data sets (continued) Linear Algebra in Big Data environment Recommendation System
Mining Data Streams And Sensor Data Link and Social Network Analysis
Affinity and Market Basket Analysis Linear Regression
Multiple Linear Regression
Logistic Regressions
SpecialTopics
Student Projects and Final Exam
Week 6
Week 7 Week 8
Week 9
Week 10 Week 11 Week 12 Week 13 &14
6
Assignments and Grading
Assignments
Exercises
Mid-term
Final
Final project /research paper
Total Grade
Grade Percent
30%
20%
20%
30%
100%
7
Project Case Study
Project:
A real world data mining project (problem statement, data, methodology/algorithm), software, execution and analysis, references, documentation, and presentation). The problem statement, sample data, relevant methodology/algorithm).
Case Study:
A case study from literature/books, prepare and deliver a comprehensive presentation including, problem statement (‘profound question’), data source(s), methodology, data mining, result, suggestions for future work, and references.
8