TWO DECADES OF TEACHING DATA MINING AND ADVANCED DATA MINING - Olivia Sheng Olivia Sheng - Simon Business School
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
TWO DECADES OF TEACHING DATA MINING AND ADVANCED DATA MINING Olivia Sheng Operations and Information Systems © Olivia Sheng
EARLIER DAYS ▪ Data mining (DM) before 2012 ▪ Only one class in business school ▪ Only two to three classes on campus ▪ Advanced data mining (ADM) before 2016 ▪ Only additional data mining course in business school ▪ Overlaps with other courses in business school ▪ Decision analysis - decision trees ▪ MBA IS core – overview of the DM methods Operations and Information Systems © Olivia Sheng
CHANGING LANDSCAPE (BIZ SCHOOL) ▪ A sea of data mining, business analytics, case oriented ML, text analytics, regression, case oriented, data analysis, big data ▪ Mostly in R ▪ Overlaps in methods and related content ▪ More practices maybe beneficial for students ▪ Increasing concerns Operations and Information Systems © Olivia Sheng
AVOIDING OVERLAPS ▪ Moving targets or difficult for pre-recorded online content ▪ What I have done ▪ Big Data – Learning from data at scale (2018) ▪ First adoption of Databricks cloud for teaching data processing, DM, structured streaming (twitter, e.g.) and a big deep learning with pre-trained image models in pyspark ▪ Advanced Data Mining – avoid text analytics and re- prep different topics in python Operations and Information Systems © Olivia Sheng
DM LEARNING OBJECTIVES ▪ Apply ML/AI methods to address common business questions ▪ Data selection, preparation and exploration ▪ Method understanding ▪ Classification, numeric prediction, association rule mining and clustering ▪ Concepts and how a method works ▪ Extremely limited math required Operations and Information Systems © Olivia Sheng
DM LEARNING OBJECTIVES ▪ Evaluate, understand, improve, and explain models and performance ▪ Select evaluation metrics ▪ Select features ▪ Select hyper-parameter settings ▪ Feature importance and inherent glassbox explanation ▪ rstudio.cloud (since 2020) Operations and Information Systems © Olivia Sheng
ADM LEARNING OBJECTIVES ▪ Topics – to fit evolving cybersecurity programs ▪ Mainly mining rare but interesting data ▪ Other topics ▪ graph mining and temporal data patterns ▪ Insufficient time and practices ▪ Better to be removed ▪ Select hyper-parameter settings ▪ Feature importance and inherent glassbox explanation Operations and Information Systems © Olivia Sheng
ADM LEARNING OBJECTIVES ▪ Enhance preparation for tackling real world data mining tasks ▪ Topics – to fit evolving cybersecurity programs ▪ Mainly mining rare but interesting data ▪ Other topics ▪ graph mining and temporal data patterns ▪ Insufficient time and practices ▪ Better to be removed ▪ Python and scikit-learn with Google Colab Operations and Information Systems © Olivia Sheng
DELIVERY AND PEDAGOGY ▪ Online and cloud IDE ▪ Very helpful for learning technical content ▪ Any place any time ▪ Very hands-on ▪ Relative to other ML/AI courses in business school ▪ No quizzes and final exam in ADM Operations and Information Systems © Olivia Sheng
DELIVERY AND PEDAGOGY ▪ Project-oriented learning ▪ A firm believer ▪ Not a fan of group projects ▪ Cannot accommodate projects in DM courses with a number number of students ▪ Need to know the project and data well ▪ Assign optional data sets in ADM Operations and Information Systems © Olivia Sheng
EVOLVING IDEAS ▪ DM ▪ Change from R to Python ▪ Some students need personalized requirements ▪ Still not sure about how to accommodate projects ▪ ADM ▪ Optional presentations to benefit students’ projects ▪ additional topics, methods and studies ▪ Research-oriented guidance for students who plan to apply for a Ph.D. study Operations and Information Systems © Olivia Sheng
INDUSTRY EXPECTATIONS FOR TRANSFORMING IS EDUCATION— DISCUSSION ON AACSB MACUDE IS TASK FORCE FINDING Industry Expectations for Transforming IS Education—Discussion on AACSB MaCuDE IS Task Force Finding The sec ▪ ond web inar in a seri es The second webinar in a series of seminars organized by AIS and AACSB MaCuDE of sem inar s org aniz ed by AIS and AAC SB Ma CuD E proj ect’ s IS task forc e pro vide s the me mb ers of the IS com mu nity project’s IS task force provides the members of the IS community a preliminary a preli min ary ove rvie w of the findi ngs of Sta ge II of the proj ect. The disc ussi on will part icul arly focu s on the role of adv anc ed anal ytic s and AI in futu re busi nes s sch overview of the findings of Stage II of the project. The discussion will particularly focus ool curri cula and the way s in whi ch dev elop me nts in thes e are as of stud y and prac tice are imp acti ng the field of IS in the busi nes s sch ool cont on the role of advanced analytics and AI in future business school curricula and the ext. The outc om es of the con vers atio n will info rm the Ma CuD E IS task forc e’s Sta ge II rep ort to this AAC SB proj ect as a who le. The sem inar will also ways in which developments in these areas of study and practice are impacting the field incl ude com me ntar y not es fro m som e emi nen t indu stry exp erts . Plea se join us to help ens ure that the voic e of the AIS com mu nity is hea rd of IS in the business school context. The outcomes of the conversation will inform the in this imp orta nt disc ussi on! Wh en: Jun e 9, 202 1 at 10 a.m . East ern Tim eRS VP: http s:// us0 2we b.zo om. us/ web inar /reg ister /W MaCuDE IS task force’s Stage II report to this AACSB project as a whole. The seminar N_s LiZy Q- SR8 ixVJ re8 KFn 3g will also include commentary notes from some eminent industry experts. Please join us to help ensure that the voice of the AIS community is heard in this important discussion! ▪ When: June 9, 2021 at 10 a.m. Eastern Time ▪ RSVP: https://us02web.zoom.us/webinar/register/WN_sLiZyQ-SR8ixVJre8KFn3g Operations and Information Systems © Olivia Sheng
Operations and Information Systems © Olivia Sheng
You can also read