Are any of these true for you or your organization?
- You want to make an effective use of abundant business transactional and survey data within your organization
- You want to formulate a customer-focus marketing strategy by understanding their needs and preferences
- You want to know what potential customers are blogging about your products and services
- You want to understand who are your valued customers
- You want to predict which customers are unsatisfied and likely to leave soon
- You want to single out customers to market add-on products and services
- You want to predict where to invest or focus your business and associated risk
If the answer is yes then our business analytics training can help you to build descriptive, predictive and prescriptive models to fulfill your need.
Traditional business analytics have so far focused mostly on descriptive analyses of historical data using a myriad of sound statistical techniques. Numerical statistical techniques can be augmented/enriched with techniques from symbolic artificial intelligence (AI), machine learning (ML)/data mining and control theory for enhance descriptive, predictive, and prescriptive (a.k.a. decision support) analytics. The unique nature of this training course is its coverage of both traditional probabilistic/statistical and cutting-edge AI/ML-based approaches to descriptive and predictive analytics and associated decision support.
The training starts with a survey of business analytics covering big players, market size, state-of-the-art in statistics, and so on. Then it provides analytics practitioners with problem modeling guidance and appropriate modeling techniques and algorithms suitable for solving their problems at hand. The training also provides a detailed account of various types of uncertainties in data and techniques for handling them. A special emphasis is given to modeling problems that are time-dependent. The course makes use of various in-house (iDAS, aText, RiskAid, E5), commercial-off-the-shelf (e.g. SAS, Matlab, Hugin) and publicly available (e.g. CLIPS, OpenStat) tools to illustrate analytics concepts with examples.
Attendees receive comprehensive slides, texts, CDs and software tools to take away for future references. The content of these tutorials are drawn heavily from books by in-house experts, especially the forthcoming book “Computational Business Analytics” and two recent ones, namely, “High-Level Data Fusion” and “Foundations of Decision Making Agent: Logic, Modality and Probability”. Each attendee will receive a complimentary copy of the data fusion book.
- Designers and developers of analytics systems for any vertical (e.g. healthcare, finance and accounting, human resource, customer support, transportation) who work within any business organizations and BPO companies around the world.
- University students and teachers, especially those in business schools, who are studying and teaching in the field of analytics.
Subrata Das with contributions from other team members.
Lesson 1: Introduction – Analytics defined, Scope of analytics, Descriptive, predictive and prescriptive analytics, Big players and market size, State-of-the-art in statistics and beyond, Candidate architectures, Available open source and commercial-off-the-shelf tools.
Lesson 2: Statistics for Descriptive and Predictive Analytics – Descriptive statistics (distributions, central tendency, dispersions), Inferential statistics (generalization, test hypothesis, estimate, prediction or decision), Dependence methods (decision tress, CART/CHAID, linear, logistics and kernel regressions, auto-regression, linear discriminant analysis, factor analysis, survival analysis), Interdependence methods (hierarchical and k-means clustering, multidimensional scaling), Potential enhancement and augmentation of statistical techniques, Conjugacy for Bayesian inference, Stochastic process modeling, Many Excel and SAS based demonstrations.
Lesson 3: Analytics Problem Modeling in Symbolic Artificial Intelligence – Approaches to handling uncertainty, Deductive, inductive and abductive reasoning, Ontology and knowledge representation, Knowledge acquisition and its bottleneck, Rule-based system, Bayesian Belief Networks.
Lesson 4: Machine Learning/Data Mining for Descriptive and Predictive Analytics – Generative vs. discriminative models, Supervised, unsupervised and semi-supervised learning, Decision trees (C4.5), Inductive logic programming, Naïve Bayesian Classifier, Neural networks, Singular Value Decomposition, Latent Semantic Analysis, Support Vector Machine, Bagging and Boosting.
Lesson 5: Time-Series Modeling for Predictive Analytics – ARMA/ARIMA, ARCH/GARCH, Hidden Markov Models, Dynamic Bayesian Networks, Kalman filtering and extensions.
Lesson 6: Prescriptive Analytics and Decision support – Test hypothesis, Expected Utility Theory, Influence diagrams, Symbolic argumentation, Reinforcement Learning Markov Decision Process.
Lesson 7: Monte Carlo Sampling – Markov Chain Monte Carlo, Metropolis-Hastings, Gibbs, Particle filtering.
Lesson 8: Text Analytics – Supervised and unsupervised text classification techniques, Natural language processing for parsing and stemming, Information extraction and structuring.
Lesson 9: Case Studies – Lots of mini case studies in areas of credit risk assessment, fraud detection, clinical state estimation, sentiment analysis, and online loan processing.