As more data are collected by businesses and scientific institutions, knowledge exploration techniques are needed to gain useful business intelligence. This course covers the theory and practice of the analysis (mining) of extremely large datasets.

Data mining is for relatively unstructured data for which more sophisticated techniques are needed. The course aims to cover powerful data mining techniques including clustering, association rules, and classification. We also breifly introduce  high volume data processing mechanisms by modeling warehouse schemas such as snowflake  and star. OLAP query retrieval techniques are also introduced.  We learn the basics of Data Warehousing structure and query formulation, as it impacts a data miner.  We do not query against  an actual  data warehouse.  There are other courses for that, if you are interested in Data Warehousing.

Outcome

Students will be able to define, apply and critically analyze data mining approaches for fields such as security, health care, science, marketing and text analysis.

COMP 251: Introduction to Database Systems or COMP 271: Data Structures  

Required:  For general reference (language and platform independent); for homework problems, lectures and examples:

This book is useful for reference and conceptual examples (language independent, with nice illustrations) of some of the underlying concepts/algorithms (such as apriori, basic classification and clustering algorithms and more).  The book doesn't have a more recent edition, but it is something of a classic as a data mining text.  Since you can find this quite inexpensively on the internet, I am including it here for reference.

Title: Data Mining: Concepts and Techniques, Third Edition
Authors: Jiawei Han, Micheline Kamber, Jian Pei
Publisher: Morgan Kaufmann; 3^rd edition (2012)
ISBN-10: 0123814790  or try this

Required:  For the RapidMiner and R lab part of the course:\

           Data Mining for the Masses

****There is an excellent fourth edition out that is an online and updated version of the third edition.  It has update powerpoint slides, short explanatory videos, review questions and other support materials.  Some students have really liked this version of the lab text, so I have created a course link where you can purchase this online text.

            Support site for the third edition

Reference:  or implementations in R (assignments, labs, cases, etc.), for reference and some good examples:
Title: Data Mining For Business and Analytics
Concepts, Techniques and Applications in R.
Authors: Shmueli, Bruce, Yahav, Patel, Lichtendahl
Publisher: Morgan Kaufmann; 3^rd edition (July 6, 2011)
ISBN-10: 1118879368
ISBN-13: 978- 1118879368

For ggplot examples: (You don't have to buy the book.  It is based off of his website.  Illustrated examples, if you are interested in Data Visualization for your project presentation.)  (Or just take the DataViz class.)
Alboukadel Kassambara.  Guide to Create Beautiful Graphics in R,  STHDA, 2013.  isbn:  9781532916960.  Most examples, with small modifications, are available on his wonderful website and his R support website.

Course Objectives and Goals

After taking this course, students should be able to:

• Know basic techniques for both supervised and unsupervised knowledge discovery.
• Know and use software package techniques for mining.
• Apply R to data preparation and basic DM algorithms
  
• Have a good grasp of data mining techniques: classification, association rules, clustering, etc.
• Understand the impact of  data warehousing schemas on  the selection of Data Mining algorithms 

• This is course is not a programming course in the data mining algorithms.
• Neither is it a strictly applications course, where you don't have to know the algorithms.
• What this course is:  This course is a (hopefully) happy medium, where you learn what the algorithms are and how they work, so that you can then use them appropriately and interpret the results.  The course keeps evolving towards a more application orientation.  We use R and we also use RapidMiner, a GUI data mining software product.

Students with Disabilities:

Loyola University Chicago provides reasonable accommodations for students with disabilities. Any student requesting accommodations related to a disability or other condition is required to register with the Student Accessibility Center (SAC). Professors will receive an accommodation notification from SAC, preferably within the first two weeks of class. Students are encouraged to meet with their professor individually in order to discuss their accommodations. All information will remain confidential.  Please note that in this class, software may be used to audio record class lectures in order to provide equal access to students with disabilities.  Students approved for this accommodation use recordings for their personal study only and recordings may not be shared with other people or used in any way against the faculty member, other lecturers, or students whose classroom comments are recorded as part of the class activity.  Recordings are deleted at the end of the semester.  For more information about registering with SAC or questions about accommodations, please contact SAC at 773-508-3700 or SAC@luc.edu.

Students who are allowed to take their exams in the SAC office are encouraged to do so.  Should you choose to take the exam in the classroom, I cannot guarantee that the classroom environment will be quiet enough to provide you with the environment that your disability may require.  If you choose to take the exam in the classroom, you are taking that risk.

Online Recording Policy

In this class software may be used to record live class discussions. As a student in this class, your participation in live class discussions will be recorded. These recordings will be made available only to students enrolled in the class, to assist those who cannot attend the live session or to serve as a resource for those who would like to review content that was presented. All recordings will become unavailable to students in the class when the Sakai course is unpublished (i.e. shortly after the course ends, per the Sakai administrative schedule: https://www.luc.edu/itrs/sakai/sakaiadministrativeschedule/). Students who prefer to participate via audio only will be allowed to disable their video camera so only audio will be captured. Please discuss this option with your professor. The use of all video recordings will be in keeping with the University Privacy Statement shown below:
Privacy Statement
Assuring privacy among faculty and students engaged in online and face-to-face instructional activities helps promote open and robust conversations and mitigates concerns that comments made within the context of the class will be shared beyond the classroom. As such, recordings of instructional activities occurring in online or face-to-face classes may be used solely for internal class purposes by the faculty member and students registered for the course, and only during the period in which the course is offered. Students will be informed of such recordings by a statement in the syllabus for the course in which they will be recorded. Instructors who wish to make subsequent use of recordings that include student activity may do so only with informed written consent of the students involved or if all student activity is removed from the recording. Recordings including student activity that have been initiated by the instructor may be retained by the instructor only for individual use. 

• Attendance:  Attendance refers to physical attendance for the on-campus class and logging into Sakai for the online class.  Most of the attendance policies do NOT apply to the online class, although please be aware the Sakai does keep track of the frequency and duration of visits.  The lectures are available asynchronous and online.
• Late homeworks and labs:  If you have a very good reason for submitting something late, you may ask me, and I may allow you to submit it late for half credit. I will only allow this once or maybe twice during the semester.  Habitual lateness will earn you zeros.
• Labs:  Most weeks, we will be working on specific lab assignments, which must be completed by the specified due dates  These will be hands-on applications of the concepts and algorithms covered in class.   You must submit them on Sakai.   Data Mining is an iterative process, and in many labs, we will be explore the data together.  Not every lab is from the lab book--I supplement with additional labs.  The major labs, such as the text mining labs and the Affinity Marketing lab, are not from the text, and you must watch the videos to complete them.  Skip the videos at your own peril.  You should complete the installation of RapidMiner on your own as soon as possible, preferably before class begins.  RapidMiner is not installed in LUC labs.  You will have to install the software on your laptops.  If you do not have a laptop, team up with someone who does.  If that doesn't work, I can set up a VM for you, but I need to know that before class begins. 
  
• Homework Assignments:  I have cut back on the homework assignments, in favor of additional labs.  For your convenience, all homework assignments (except for the Orientation ones) are in a single Excel spreadsheet.  Please note that the homework problems have been reconstructed so that the values are not the same as in the text; nor are they the same as the last time I taught the course.  This means that you really have to do the homework.  For the concept questions that do not require calculations, please be very careful to state the answers in your own words.  Do not  copy from the textbook.  Do not copy from the Instructor's Manual.  Do not  copy from other students, either in this class or previous classes.  If you do, you will be subject to the penalties for Academic Honesty violations. There are many different models and algorithms covered in the course.  The purpose of the homeworks is to give you prompt application of the foundational concepts and algorithms, as well as prompt feedback on your understanding of the material.  Practically speaking, the homeworks will prepare you for the exam, which is not lab-based.
• Each tab in the homework spreadsheet is a separate problem.   You may add worksheets and submit your homework in the same spreadsheet (which may be helpful to you in preparing for the exam), or you may submit a separate document.  Or, you may keep your spreadsheet in Google Drive, Box.com or Github and submit a link on Sakai.  (That's probably a good idea!)  Handwritten work should be limited to those graphs or charts that may be difficult or cumbersome to create with software. Homework may be worked on with a team of your classmates, and may be submitted as one submission per team.  You must always submit something on Sakai:  either the homework itself, or a comment that a team member submitted the homework on your behalf.  If you are submitting on behalf of your team, please indicate in your comment the names of the students in the team.   If you do not, you may not get credit for the homework, even if your team member submitted it on your behalf.  It is not sufficient for your team member to submit your name.  You must also submit a comment and submit it as your assignment. That is your responsibility!!
  
• Exams:  There is one exam.  It is scheduled in Week 9, which is late enough to include at least some of the Classification algorithms, but early enough to allow you to withdraw from the course if you are not pleased with your performance.  The exam will be taken using Sakai and zoom with a web cam.  Exact dates/times will be determined after I poll students for availability. 
  
• Project:  You are required to design and implement a project that analyzes a large dataset, using some of the techniques and algorithms that are covered in the course.  The project presentations will be on 5/03.  Projects may be completed in teams; however, every team member is expected to fully participate in the project in order to earn the team grade.  There are more details about the project in the project assignments.
• Project Excellence:   A "very good" project that fulfills the requirements and is perfectly satisfactory can earn full points for all non-Excellence points for  the Project.    If you get full points on your project and do not earn any Project Excellence points, please don't ask me what you did "wrong".  The answer will most likely be  "nothing".  Project Excellence points can be earned for a project that is truly excellent, incorporating components well beyond the material covered in class, or simply having a real "Wow" factor.  A grade of "A" means "excellent".  There are many points in this course that could be considered free points.  Full points for Homework and Labs can be earned if significant effort is demonstrated:  you are graded for completeness, not necessarily for correctness (although grossly incorrect answers will be marked off).  But for a project to get an A, it must be excellent.  As I review your progress during project walkthroughs, I will tell you what you can do to earn excellence points for your project.
• Extra Credit:  There is no extra credit in this course, as this is not practical to the course, nor is it fair to other students.

Course Schedule

This schedule is a guide.  Exact dates and topics may be subject to change.  It is my best estimate, but we may have to adjust the schedule slightly.  You are responsible for all announcement/changes made in class or posted on Sakai.  

• s

• s

• s

• s

• s

• s

• s

• s