MIS 6302.X02: Analytics and Information Technology The University of Texas at Dallas Spring 2014 Professors Office Phone e-mail Office Hours Indranil Bardhan JSOM 3.414 972-883-2736 bardhan@utdallas.edu W: 5-6:30pm & by appointment Syam Menon JSOM 3.421 972-883-4779 syam@utdallas.edu Tue: 2-3:30, & by appointment Pre-requisites There are no pre-requisites for this class. However, some knowledge of basic statistics and probability can be helpful. Course Description Most of you have heard of Big Data, and the wonders it can do from recommendations at Amazon and Netflix to Nate Silver s predictions in the 2012 presidential election. While there is no consensus on a precise definition of this term, we know that it has something to do with large volumes and new sources of data. How does this data turn into insight? That is where analytics comes in, usually in the form of advanced modeling and visualization techniques. Approximately half of this course will focus on analytics, and on related decision making. Students will gain experience using BI software, in the form of SAS Enterprise Miner. At the end of the course, you will not only appreciate the substantial opportunities that exist in the business intelligence realm, but also learn techniques that will allow you to exploit these opportunities. The knowledge of these methods will allow you to ask relevant questions when similar problems present themselves in the course of your work. The ability to ask the right questions will be further enhanced in the other half of the course, which addresses strategy in the information technology (IT) realm. Cases and discussions covering various aspects of IT will hone your ability to narrow in on the key questions when given a specific context. Learning Outcomes To gain a general understanding of business intelligence, and to appreciate the data rich environment of today's global economy. To gain a practical understanding of key methods integral to data mining and an understanding of the choice of technique for different real-world situations. Apply the concepts of business analytics in developing and evaluating information technologybased strategies for companies. Required Textbook Data Mining for Business Intelligence: Concepts, Techniques, and Applications in Microsoft Office Excel with XLMiner by Galit Shmueli, Nitin Patel, Peter Bruce. ISBN-10: 0470526823, ISBN-13: 978-0470526828. This
will be useful reference, particularly for the material covered in sessions 2-5. This book is available as an ebook from the UT-Dallas library. Required Software SAS 9.3 - There will be hands-on exercises and homework using SAS Enterprise Miner. Some illustrations will also involve MS_Excel. Suggested Reading The following cases will be used in the course for classroom discussion. All students are expected to review and become familiar with all cases as you may be called upon to engage in class discussion. 1. Cemex: Global Growth through Superior Information Capabilities (HBS) 2. Google, Inc. (HBS) 3. Air France Internet Marketing: Optimizing Google, Yahoo!, MSN and Kayak Sponsored Search (HBS) 4. Oracle versus Salesforce.com (HBS) 5. Social Media Strategy for the Minnesota Wild (University of Minnesota) 6. The Global Delivery Model at Infosys (Georgia Tech University) Reading articles that are available in your course pack include the following: 1. Business Models: An Introduction. Haim Mendelson, Stanford Graduate School of Business, July 2013. 2. Competing on Analytics. Thomas Davenport, HBR, 2006. 3. Why IT Fumbles Analytics. Marchand and Peppard, HBR, Jan-Feb. 2013. 4. Successfully governing business process outsourcing relationships. Mani, Barua, and Whinston, MIS Quarterly Executive. March 2006. 5. Strategies for Two-sided markets. Eisenmann, Parker, and Van Alstyne, HBR, 2006. Additional readings will be posted on elearning during the course. Grading Criteria Weights Assessment Points Weightage Homework 1 100 10 % Homework 2 150 15 % Final Exam 250 25 % Project Report and Presentation 250 25 % Group Case (incl. peer evaluation) 150 15% Class Participation 100 10 % Total 1,000 100% Spring 2014 Page 2 of 6
Grading Scale Homework Overall Course Total Letter Grade 920 1000 A 890 919 A- 860 889 B+ 820 859 B 790 819 B- 750 789 C+ 680 749 C 0 679 F There are two group homework assignments in this course. Homework 1 will contribute 100 points (10%) and homework 2 will contribute 150 points (15%) towards the final grade. They will involve the use of SAS Enterprise Miner, and you will have at least 1 week to work on each homework assignment. The first homework will be assigned in the 3 rd session (02/07/2014), and will be due in the 4 th session (02/21/2014). The second homework will be assigned in the 4 rd session (02/21/2014), and will be due in the 5 th session (03/07/2014). Final Exam A 2-hour examination will be administered on 05/03/2014, and will be worth 250 points (25% of the final grade). It will cover all material covered in the course until that time. Project 250 points (25%) of the final grade will be determined by your performance on a group project. Each group will have to obtain data from one of their member s companies for this project, with the objective of providing some useful insights in return. If obtaining data from work is not possible, you may use a dataset from one of the competitions at kaggle.com. The primary objective of the project is to encourage you to explore and think about potential applications of the techniques you will learn in this class. It offers you an opportunity to apply your BI knowledge to real-life data and to mine managerially-relevant insights. You are required to identify a topic that is not only interesting, but also understandable to you. You should also have a clear idea about your data mining objective it should be closely related to the methods covered in this course. It is acceptable to use methods not covered in the course, but only as a means to do pre- or post-processing the primary focus should remain one (or more) of the methods taught in the course. Spring 2014 Page 3 of 6
Project Deliverables 1) One-page Proposal: A 1-page proposal is due on or before the 4 th session (02/21/2014) and must be approved by the instructors; failure to submit a proposal by this date will result in a grade of zero for the paper. The proposal should clearly indicate the key question(s) you are trying to address and how you propose to obtain the data. A brief description of the data should also be included, so that it is clear that you have done enough work to ensure that you will get the data, and know what to do with it. If you wish, you may attach the list of all attributes and definitions (though this is not required). We will make a quick assessment whether this project will be feasible and interesting enough based on this proposal. Flags will be raised if there are any concerns. All paperwork required to ensure availability of the data (like non-disclosure agreements) need to be in place and submitted along with the proposal. 2) Midterm Report: A halfway report is due by the 7 th session (03/22/2014). This report should describe your data mining objective and the data. Data should be available and submitted along with this report, so the instructors can look at the data if needed. Each team should have conducted at least one successful trial of data mining on this dataset (so we know that you can do it), and should report your preliminary findings (so we know that you are on the path to obtain something interesting). 3) Final Report: The final paper is due on or before the 10 th session (04/26/2014). This should be a professionally prepared report that has at least the following parts: cover page, executive summary, project motivation/background, data description, your BI model, Enterprise Miner diagrams used, your findings and managerial implications/conclusions, an references. Feel free to add other sections if needed. If you feel any of the above required parts should not be included in the final report, you need to get the instructors prior approval by explaining why. There is no page limit (but please be reasonable). What really matters is whether you successfully discovered useful knowledge from a dataset relevant to one of your jobs, and whether you presented it well to readers if you can do so in a concise way, that is preferred. The final report will be submitted and examined through the plagiarism detection tool, Turnitin.com. For more information and assistance on using Turnitin.com, please go to: http://www.turnitin.com/static/training.html. 4) Presentation: Part of the evaluation of the project will be based on a group presentation in the 10 th session (04/26/2014). All members of the group are expected to participate in a useful way in the presentation, either by presenting, or responding to questions, or both. The analysis and development of the presentation should be a contributory effort with full participation by all group members. The paper should be in 12-point font, double-spaced, and between 10-12 pages in length. It should include a minimum of five appropriate references and citations (Wikipedia is not an appropriate research source). A professional paper in terms of style and mechanics (spelling, etc.) is expected. The grade for the project will be determined primarily on (i) the relevance of the topic, (ii) the quality and originality of the ideas and the extent of analysis presented, (iii) evidence to support (iv) the organization and presentation, Spring 2014 Page 4 of 6
and (v) other factors (input from Turnitin.com, professionalism, grammar, spelling, proper referencing, etc.) Peer Evaluation Each student will evaluate himself/herself as well as other group members, on all group work, using a Peer Evaluation Form. The completed form should be submitted by the 10 th session (04/26/2014). Class Participation Class participation is encouraged the class participation grade is subjective and will be based on a student s level of preparation for the class (has the student done all the required work needed for a class?), participation in class (is the student making a positive contribution to the analysis and discussions, and therefore, to the class?), and attendance (preparation and participation cannot be evaluated without your presence!). You are also expected to read any supplemental readings distributed by the instructors. Case Analysis The class will be split into 7 teams. Each group will prepare, analyze, and submit their analysis for one assigned case. However, all groups are responsible for reading all assigned cases and be prepared to contribute to a vigorous case discussion. Case analysis is a group effort, and each group presentation should be approximately 20 minutes in duration. Each group will submit its written analyses in Powerpoint format to the Instructor before class. Course Policies Make-ups: There are no make-ups in this course for any of the graded material. Extra Credit: There will be no work available for extra-credit. Late Work: Late submissions will not be accepted. Please refer to the following web page link for other policies related to this course: http://coursebook.utdallas.edu/syllabus-policies Spring 2014 Page 5 of 6
ACADEMIC CALENDAR SESSION LECTURE TOPICS ASSIGNMENTS 1: 01/17 Introduction to Business Technology, Big Data & Analytics Creating Value through Business Model Innovation, Competing on Analytics, Introduction to SAS Enterprise Miner Cemex case 2: 01/25 Unsupervised Methods Lectures, In Class Exercises HW1 assigned Air France Internet Marketing case 3: 02/07 Supervised Methods Lectures, In Class Exercises HW1 due; HW2 assigned 4: 02/21 Social Media 5: 03/08 Recommender Systems 6: ONLINE Global IT Delivery Models 7: 03/22 Enterprise 2.0 Social Media strategy and Networks Lectures, Amazon.com Recommendations Outsourcing & Governance in global IT delivery teams, Risk management Platform-based ecosystem, Online search & Advertisingsupported business models Minnesota Wild Case HW2 due; Project Proposal due Infosys Case AV recorded lectures available on elearning Google Case Project Half-way report due 8: ONLINE Future Information Technology Trends Software as a Service, Cloud Computing Mobile & the app. economy Audio lectures will be available on elearning 9: 04/25 Discussion; Group Presentations Final Project Reports Due Oracle vs. Salesforce case FINAL EXAM: In-class Examination on Saturday, May 3 Spring 2014 Page 6 of 6