California State University- Sacramento MBA 264 Business Intelligence with Data Mining Fall 2014 Monday 6:00PM - 8:50PM INSTRUCTOR OFFICE PHONE E-MAIL OFFICE HOURS Tahoe Hall 1002/1009 Feng "Oliver" Liu, PhD TAH 2020 (916) 278-7106 liu@saclink.csus.edu Monday: 11:50 PM - 1:20 PM 2:50 PM - 4:20 PM Course Description and Objectives Because of the enormous, and rapidly growing amount of data flowing from, to, and through enterprises of all sorts, today’s successful enterprises are, typically, those that make effective use of the abundance of data to which they have access: to make better predictions, better decisions, and better strategies. This course covers the processes, methodologies and current practices used to transform business data into useful information and knowledge for IT-enabled managerial decision support and performance improvement. The students will learn how to ask the right questions and how to draw inferences from the data by using the appropriate data mining techniques, XLMiner/SPSS, and real business data. Overall, the course will enable students to approach business problems data-analytically, envision data-mining opportunities in organizations, and also follow up on ideas or opportunities that present themselves. The specific objectives of the course are: 1. To introduce the steps involved in data mining, from goal definition to model deployment. 2. To examine supervised learning methods, such as classification and prediction. 3. To understand unsupervised learning methods, such as association rules and clustering. 4. To discuss data visualization techniques. 5. To improve the student's communication and team working skills. Students will be afforded the opportunity to demonstrate their abilities in a number of different ways throughout the course, including individual and group work, written papers, oral presentations and participation in classroom discussions, among others. Course Materials a. Textbook: Data Mining for Business Intelligence, Second Edition, by Galit Shmueli, Nitin R. Patel and Peter C. Bruce, Wiley; ISBN-10: 0470526823, ISBN-13: 9780470526828. New copies of this book come bundled with a free 6-month license to use the software XLMiner, which is required for this course (see below). Since this software is otherwise expensive, you are recommended to purchase new copies of the book only. b. Software: XLMiner. Instructions for obtaining the free 6-month license using the access code available in the book can be found here: http://www.solver.com/xlminer/wiley.html. Since you will need the software for doing in-class exercises as well as homework assignments, you are advised to install the software on your laptops. c. Reading Materials: additional articles/papers/cases will be posted in the course contents page in SacCT. Grading HOMEWORK CLASS PARTICIPATION PROJECTS EXAM 45 % 15 20 20 100% Grading Scale: A = 93 or higher B- = 79 – 82 F = < 60 A- = 89 – 92 C = 70-78 B = 83 – 88 D = 60- 69 Homework: These will consist of problem sets that are designed to give the students valuable practice and enhance their understanding of the concepts covered in class. These will be individual assignments due by the times and dates designated in the schedule. Totally there will be ten assignments. If a student misses ONE or more of these assignments, he or she will lose 5 points. Otherwise, 5 points will automatically be rewarded to the student just for his or her efforts devoted to the assignments. The homework will be electronically submitted on SacCT on the due dates (see Class Schedule). For each student, five of the assignments will be graded (each one is worth 8 points), and the resulting grades will be documented. The student has the right to select which ones of hers or his to be graded simply by writing “TBG” (to be graded) on her or his submissions. More instructions/ guidelines on the assignments will be given in the classes. Projects: The purpose of the projects will be for the student to gain hands-on experience in solving a realistic problem using the data mining principles covered in class. These will be a group assignment. More specifically, four teams will be formed in this class. Each team will be assigned two case studies from chapter 18 of the text. The team is expected to conduct two research projects based on these two cases, and present its findings, both orally and in writing, by the end of the semester. Detailed instructions for the projects will be provided later. Final Exam: This will be a take-home exam—handed out on the last day of class. This will be an individual assignment due by the time and date designated in the schedule. Except for extreme cases (i.e., medical emergency), the professor does not give make-up exams. Class Participation: The student will be expected to ask questions in class, participate in class discussions and complete in-class exercises, and will be evaluated based on her or his involvement in exercises/ discussions in class. This is a subjective grade assigned by the professor based on both the quality and the quantity of the student's inputs to class discussions and activities. It also takes into account the student's attendance to the class. Policy on Academic Honesty Cheating devalues your degree. We all have the responsibilities for enforcing a fair and respectful learning environment. Please take a moment to review the university policies regarding academic dishonesty: http://www.csus.edu/umanual/AcademicHonestyPolicyandProcedures.htm. As a general principle, for individual assignments, there must be no sharing of coding and the implementation method among individuals in any manners. For group assignments, there must be no sharing of coding and the implementation method among groups in any manners. Each assignment is expected to be the unique and original work of a student or a group for this course in this term. Assignments which bear strong resemblance to others' will not be graded. Maximum penalty under the university policy will be enforced for academic dishonesty. ADA Compliance It is the policy of California State University to accommodate students with disabilities, pursuant to federal and state law. Students with disabilities who need classroom accommodations must make their requests by contacting the Services to Students with Disabilities (SSWD, http://www.csus.edu/sswd/) at 916‐278‐6955 (phone) 916‐ 278‐7239 (TDD), Lassen Hall Room 1008. CLASS SCHEDULE Date Sep 8 Sep 15 Sep 22 Sep 28 Oct 6 Oct 13 Oct 20 Oct 27 Nov 3 Nov 10 Nov 17 Nov 24 Dec 1 Dec 8 Dec 15 Topics Reading Syllabus/ Introduction Research Team Formation and Project Selection Overview of Data Mining / Statistics Refresher (Probability, Distribution, Hypothesis Testing, ANOVA, Correlation)/ Introduction to XLMiner Prediction: MLR HW #1 due Classification: Logistic Regression Ch. 1 & Ch. 2 Class notes Ch. 6 Ch. 10 Dimension Reduction HW #2 due K-Nearest Neighbors HW #3 due Naive Bayes HW #4 due Decision Trees HW #5 due Neural Nets HW #6 due Discriminant Analysis HW #7 due Project work day Ch. 4 Association Rules HW #8 due Forecasting Time Series Data Visualization HW #9 due Project Presentations HW #10 due Project Written Report Due Ch. 15, 16 Ch. 7 Ch. 8 Ch. 9 Ch. 11 Ch. 12 Ch. 13 Ch. 17 Ch. 3 Final Exam Due: Dec 18 * This is a tentative schedule and may change throughout the semester