Wednesday, 2 April 2014

Applied Predictive Analytics

Einstein said "In theory, theory and practice are the same.  In practice, they are not."  This course is about the practice of predictive analytics, and is taught in partnership with a company with years of consulting experience with telecoms (predicting churn), large retailers (predicting holiday sales), online merchants (microtargeting), and more

This is a fully practical and applied course - you should have some familiarity with predictive modeling before taking it.  In part 1 of this course, you will get some hands-on experience with real-world data and the issues of problem definition and data prep.  In Part 2 you will take a defined dataset, develop models, assess them, and submit your best model.  Your instructor will serve as your personal guide in this effort.

Mr. Mohan Singh will present his course “Applied Predictive Analytics, in partnership with CrowdANALYTIX” online at For more details please visit at Applied Predictive Analytics, inpartnership with CrowdANALYTIX.

Aim of Course:
The goal of this course is to teach users (who have basic knowledge of R programming, predictive analytics and statistics) to apply machine learning techniques in real world case studies. This course provides hands on approach, presenting the opportunity to participate in a private educational competition hosted by CrowdANALYTIX.

Business Case Study: We will study data from the "daily deals" industry (consisting of websites like Groupon, Living Social etc. which source local deals to offer each day). The daily deals industry is emerging and highly competitive. The goal will be to predict the revenue from each offering using the given data.

Course Walkthrough: Each week, participants will be given a set of exercises and instructions to work on the raw data or processed data (details below). Users will apply their statistical/machine learning knowledge, along with their business understanding, to solve the problems and interpret the modeling results for the given business objective. The course will follow an iterative approach for problem solving, in which users are required to submit modeling responses multiple times.

Who Should Take This Course:
Business analysts, R users, SAS users, statistical analysts who want to learn to implement and apply machine learning techniques to solve predictive analytics problems in a real world business case. Data mining analysts who want extend their knowledge to include machine learning techniques for data modelling.

Course Program:
WEEK 1: Business Case Study, Introduction and Data Pre-processing
  • Course and business case study introduction
  • Data Cleaning, pre-processing & data visualization
  • Understanding business objective to be modelled based on the case study
Assignment: To read data in R and perform simple statistical measures. Perform data cleaning, data pre-processing and data manipulation as required by the case study problem.

WEEK 2: Classification Problem, Data Sampling, Feature Selection and Model Building
  • Introduction to machine learning
    • Supervised classification problem (Machine learning packages in R)
  • Perform feature selection and processing on given data
  • Model building based on classification techniques
Assignment: To perform simple exercises for machine learning packages in R. Converting raw data into features or creating new features, if needed for the modelling objective.

WEEK 3: Evaluation Metrics, Scoring Model Output, Leaderboard Prediction
  • Build machine learning model like decision trees, ensemble modelling etc.
  • Evaluation metric like confusion matrix, RMSE etc.
  • Leaderboard scoring for peer-2-peer feedback
Assignment: Perform basic to intermediate level machine learning model for the problem. These scores will be reflected on the private contest leaderboard for all participants.

WEEK 4: Visualization of Modeling Output and Analysis of Features & Model for Business Case Study
  • Model building and Leaderboard updates on a daily basis visualize the output of the model
  • Interpret modelling results for case study objective
Assignment: Final week assignment is based on understanding the various model outputs for the business objective.

The instructor, Mr. Mohan Singh, is a senior data scientist at CrowdANALYTIX with extensive research experience in healthcare and imaging.  He has taught data mining & computer science at University College Dublin and UC, Irvine.  Participants can ask questions and exchange comments with Mr. Singh via a private discussion board throughout the period.

You will be able to ask questions and exchange comments with the instructors via a private discussion board throughout the course.   The courses take place online at in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

We, the Center for eLearning and Training (C-eLT), Pune, partner with and offer these courses to Indian participants at special prices payable in INR.

For India Registration and pricing, please visit us at

Call: 020 6680 0300 / 322


No comments: