Data Science for Decision Making
Become a Data Smart Manager
3 Weeks Live Sessions | 4th Weekend Campus Immersion
Scaling statistical algorithms to data sets and understanding the underlying rationale behind the same is one of the most critical pieces in the Data Science puzzle. The core skill sets required for Data Analysis problem definition, selecting the right statistical model, making the right assumptions and inferring the correct conclusions are emphasized in this bootcamp. This bootcamp will provide the necessary foundation to start your Data Science career.
Talk to Admissions: 9740-991-601
Promotions & Dynamic Coupon Creation for a Retail Store
In this skillathon, your CEO has tasked you to create a model to predict the features that influence sales. You should be able to predict the outcome of a promotional campaign that your marketing team has devised. Your model should be specific to any changes made to the campaign for example changing the coupon price by $1 or adding promotional fliers to each cart.
Methods/Models: Multiple regression, Exploratory Data Analysis (EDA)
Customer Renewal Churn Detection for a Cloud based Product
You are a data scientist at a SaaS company that provides a subscription service to businesses. In a B2B scenario the number of customers is low and revenue per customer is high as compared to a B2C scenario.
Your CEO is concerned that some of the customers might leave (churn) and he wants you to identify which customers are most likely to churn. Your model should also identify the drivers that contribute most to the customer defection. This would help the customer service group to specifically work on these customers and avoid giving random and costly discounts without the insights that your model will likely provide.
Methods/Models: Logistic Regression
Industry: Information Technology
Lifetime Value Prediction for E Commerce Customers
In this skillathon, you will calculate the customer lifetime value which is an estimate of how much a customer is worth to a company. It is the present value of the future cash flows attributed to a customer during her entire relationship with the company. As a data scientist, you will inform the management about how valuable each customer is to the business and what are the chances that they will purchase in a given time period.
Methods/Models: Recency Frequency and Monetary Value (RFM) analysis
Industry: E Commerce
Cancer Detection for a Hospital
Early diagnosis is critical for a successful cancer treatment. Your task as a data scientist is to diagnose breast cancer using digitized methods of collecting tissue samples. The CEO of the hospital where you work is very keen to find a cheap and quick way of diagnosing cancer at an early stage. This could serve as a source of competitive advantage for him in this highly competitive space. Using the dataset you will build a model that predicts cancer based on the features as accurately as you can.
Methods/Models: Naive Bayes Classifier
Why this Bootcamp
Work on Real-life Data Science Problems
Take your career head-on by working on projects using a competency based learning paradigm. Quality of time spent and the outcome is far more important than the quantity.
Work 1:1 with a Mentor
We pair you with a mentor who has extensive professional and academic knowledge of the field. You will have one-on-one conversations with your mentor, and receive constructive feedback on your work.
We Will Keep You Engaged
Our mentors are here to keep you motivated, answer questions, provide feedback, and help deepen your understanding of essential tools and techniques. Learn with live online classes and face to face sessions. Learning is best when you are able to ask the questions and clarify your doubts with the faculty.
What You Will Learn
Unit 1: Introduction to Statistics
■ Measures of Central Tendency
■ Probability & Probability Distributions
Unit 2: Sampling and Hypothesis Testing, ANOVA
■ Sampling Distributions, Estimation
■ Hypothesis Testing (t, Chi-Square, F Sampling Distributions)
■ ANOVA, Statistical Significance
Unit 3: Correlation and Regression
■ Correlation and Simple Linear Regression
■ Multiple Linear Regression
■ Quantile and Logistic Regression
Unit 4: Time Series and Bayesian Statistics
■ Stochastic Processes
■ Autoregressive-Moving Average Models ARMA
■ Box-Jenkins Model
■ Bayesian Inference and Regression
1. Build a purchase propensity model for your favourite brand of chocolate.
2. Use logistic regression to uncover suspicious activities on credit cards and reduce chargebacks.
3. Predict lifetime value/risk of churn for telecom using call center database.
Ability to be learn hands on with real industry data and delivering insights to industry jury is the best part of the program. Data Science and its application for Decision Science with practitioner faculty is the biggest highlight of the program. Strongly Recommend it.