Michel Ballings
Transcription
Michel Ballings
New items since Aug 1, 2014 are in blue Michel Ballings Assistant Professor of Business Analytics at The University of Tennessee Department of Business Analytics and Statistics 235 Stokely Management Center Knoxville, TN 37996-0525 mb@utk.edu ballings.co https://twitter.com/michelballings http://www.linkedin.com/in/michelballings Google Scholar: http://goo.gl/HGR1Vf http://bas.utk.edu/our-department/faculty/ballings.asp Summary Assistant Professor of Business Analytics at The University of Tennessee, interested in machine learning and predictive analytics for social media. Trained at the intersection of three disciplines: (a) statistics and machine learning, (b) programming and computing, and (c) business, management and marketing. Education 2008–2014 Ph.D. in Applied Economics, Ghent University Title: Advances and Applications in Ensemble Learning Supervisor: Prof. dr. Dirk Van den Poel (Dirk.VandenPoel@UGent.be) 2006–2008 Master in Business Economics, VLEKHO Business School Brussels (Magna cum laude, Ranking: 1st) 2003–2006 Bachelor in Business Administration, Leuven University College (Cum laude) Studied at Haute Ecole d’Enseignement Sup´erieur de Namur (one semester, 2005) Studied abroad at Universidad de M`alaga (one semester, 2005 - 2006) Teaching •UTK: The University of Tennessee, UGent: Ghent University Fall 2014 •UTK: STAT 583, section 2: Marketing Analytics (R) Spring 2014 •UGent: Marketing Models and Marketing Engineering (R) Spring 2013 •UGent: Marketing Models and Marketing Engineering (R) Fall 2012 •UGent: Marketing Information Systems - DB Marketing ((PL/)SQL, ODM) Fall 2011 •UGent: Marketing Information Systems - DB Marketing ((PL/)SQL, ODM) Fall 2010 •UGent: Marketing Information Systems - DB Marketing ((PL/)SQL, ODM) •UGent:Analytical Customer Relationship Management (SAS) Fall 2009 •UGent:Analytical Customer Relationship Management (SAS) Fall 2008 •UGent:Analytical Customer Relationship Management (SAS) Employment 2014–now Assistant Professor of Customer Analytics at The University of Tennessee 2008–2014 Doctoral researcher and teaching assistant at Ghent University 2008–2010 Consultant at Psilogy 2007–2008 Internship at AB InBev Languages Dutch Native speaker English Full professional proficiency French Full professional proficiency Spanish Elementary proficiency Skills Programming R SAS: BASE, STAT, MACRO Oracle: SQL, PL/SQL, Data Mining (ODM) SPSS-syntax LISREL Applications Reporting Predictive Analytics, Data Mining, Acquisition Modeling, Up-Sell Modeling, Cross-Sell Modeling, Retention Modeling, Segmenting Targeting Positioning, Out-of-stock prediction MS Word, MS PowerPoint, MS Excel, LATEX Software 2014 • Michel Ballings, and Dirk Van den Poel (2014). R package dummy: Automatic creation of dummy variables with support for predictive modeling 2013 • Michel Ballings, and Dirk Van den Poel (2013). R package kernelFactory: an ensemble of kernel machines. • Dirk Van den Poel, Michel Ballings, Andrey Volkov, Jeroen D’haen, and Michiel Van Herwegen (2013). R package aCRM: convenience functions for analytical customer relationship management. • Michel Ballings, Dauwe Vercamer, and Dirk Van den Poel. Hybrid Ensemble: An R package for Hybrid Ensemble classification) • Michel Ballings, and Dirk Van den Poel. AUC: Threshold independent performance measures for probabilistic classifiers Conference Talks 2014 • Michel Ballings. CRM in Social Media: Predicting Increases in Facebook Usage Frequency. INFORMS Annual Meeting 2014, Session Title: Predictive Analytics for Social Media, Cluster: Artificial Intelligence, San Francisco, USA, November 9-12, 2014 2013 • Michel Ballings. Evaluating multiple weight estimation methods in hybrid ensembles. The 2nd Annual Conference of the Dutch/Flemish Classification Society (VOC), Antwerp, Belgium, May 31, 2013 2012 • Michel Ballings. The dangers of using intention as a surrogate for retention in brand positioning decision support systems. The 36th Annual Conference of the German Classification Society (GFKL), Hildesheim, Germany, August 1-3, 2012 • Michel Ballings. Improving customer churn prediction by data augmentation using pictorial stimulus-choice data. International Symposium on Management Intelligent Systems, Salamanca, Spain, May 22-14, 2012 2011 • Michel Ballings. Data Augmentation in Customer Intelligence using Pictorials. 2011 Joint Statistical Meetings, Miami, USA, July 30- August 4, 2011 Seminar Talks 2014 • Michel Ballings. Social Media Analytics. Seminar Talk. The University of Tennessee, Haslam College of Business, Department of Management, November 4, 2014 • Michel Ballings. Advances in Social Media Analytics. Seminar Talk. The University of Tennessee, College of Engineering, Department of Industrial and Systems Engineering, October 3, 2014 2012 • Michel Ballings. Resolutions to Having Too Much or Too Little Data. Top Management aCRM Training, Ghent University, College of Economics and Business Administration, Department of Marketing, August 10, 2012. 2011 • Michel Ballings. Data Augmentation for CRM. Executive Course: Predictive Analytics for Customer Intelligence in Financial Services, Ghent University, College of Economics and Business Administration, Department of Marketing, August 18, 2011. Practical industry projects Analytics projects. Executed, coached or managed. 2014 • Hanes Brands Inc. (Project coach) • Club Brugge 2013 • Essent (Project manager) 2012 • Concentra 2011 • USG People 2010 • Friesland Campina 2009, 2010 • Vodafone • Carglass 2008 • Anheuser Busch InBev (internship project) 2014 • Faculty coach in BZAN 550 -Business Analytics Experience Capstone Service • Talk: The Twin Tools of Predictive Analytics and Testing, at FedEX Operation Research and Management Science (FORMS) Conference in Memphis, University of Memphis, Sep 16 • Search committee member for two new Faculty members at the Assistant Professor level Journal Articles 2013 • Michel Ballings and Dirk Van den Poel (2013). Kernel factory: An ensemble of kernel machines. Expert Systems with Applications, 40(8), 2904-2913. 2012 • Michel Ballings and Dirk Van den Poel (2012). Customer event history for churn prediction: How long is long enough? Expert Systems with Applications, 39(18): 13517-13522. Conference Papers 2013 • Michel Ballings and Dirk Van den Poel (2013). Using Eye-Tracking Data of Advertisement Viewing Behavior to Predict Customer Churn. In 2013 IEEE 13th International Conference on Data Mining Workshops (ICDMW 2013), 7 December 2013, Dallas, Texas, USA. • Michel Ballings and Dirk Van den Poel (2013). The dangers of using intention as a surrogate for retention in brand positioning decision support systems. In M. Spiliopoulou and L. SchmidtThieme, editors, Studies in Classication, Data Analysis, and Knowledge Organization, page 8. Hildesheim, Germany. 2012 • Michel Ballings, Dirk Van den Poel, and Emmanuel Verhagen (2012). Improving customer churn prediction by data augmentation using pictorial stimulus-choice data. In J. Casillas, F. J. Martinez Lopez, and J. M. Corchado, editors, Management Intelligent Systems, volume 171, pages 217-226. Springer-Verlag Berlin, Berlin. 2011 • Michel Ballings, Dries F. Benoit, and Dirk Van den Poel (2011). RFM variables revisited using quantile regression. In M. Spiliopoulou, W. Haixun, D. Cook, P. Jian, W. Wei, O. Zaiane, and W. Xindong, editors, 2011 IEEE International Conference on Data Mining Workshops, pages 1163-1169. Vancouver, Canada. Professional Affiliations Since 2014 • The Institute for Operations Research and the Management Sciences (INFORMS) Since 2011 • Foundation for Open Access Statistics (FOAS) Journal Reviews Since 2014 • Omega, The International Journal of Management Science • Expert Systems with Applications (ESWA) Since 2013 • Journal of the Association for Information Science and Technology Ph.D. students Started 06/2014 • Co-advisor of Matthijs Meire. Topic: Big Data Marketing Analytics and Social Media. Ghent University. Started 09/2014 • Co-advisor of Steven Hoornaert. Topic: Social Media Analytics. Ghent University. Started 09/2014 • Co-advisor of Matthias Bogaert. Topic: Social Media Analytics. Ghent University.