Data Science Foundation Program

• What is Data Science?
• How data can be used for Decision Making?
• Introduction to Machine Learning
• Introduction to Artificial Intelligence
• Introduction to Big Data
• Applications in Data Science and Machine Learning

• Fundamentals: R overview. Installation. Packages & walkthrough. Data structures (Vector, array, factors, data frames, lists). Vector calculation. Arithmetic & logical operators. Subsetting. Missing, indefinite & infinite values.
• Control Flow Basics: For loops. While loops. Nested loops. The disadvantages of using loops. Alternates to loops
• Functions: Understand the structure of the function. Build your own function. Usage of parameters and default values. Usage of return.
• Packages: How to search & choose a new package. Package installation & updates. Help and learn. Access package functions. Hack a function. Build your own package.
• Environments Objects: Save, load & delete objects.
• Data Import and Export: Import & export from Excel. Import & export from MySQL. Import & export from the text file. Export to image & PDF. Present output in HTML webpage
• Data manipulation basics: Sort & rank. Data Aggregation. Merging.
• Data manipulation advanced: Apply, Lapply, Tapply, By, Replicate functions. Dplyr. Tidyr.
• Data Visualization fundamentals: Plot function. Changing parameters. Drawing basic charts. Adding chart elements.
• Data Visualization advanced: Qplot, Ggplot, Maps.

• Linear Regression: Introduction to linear regression technique & its uses. Details of ordinary least squares estimation technique. Modeling steps. Variable handling. Model statistics interpretation. Validation of linear regression assumptions. Metrics to measure model performance.
• Logistic Regression: Introduction to logistic regression technique & its uses. Maximum likelihood estimation technique. Modeling steps. Dependent variable definition. Variable handling. Weight of Evidence & Information Value. Variable reduction. Model statistics interpretation. Metrics to measure model performance.
• Time series forecasting: Learn basic concepts of time series modeling. Basic techniques for forecasting. Smoothing techniques. Decomposition. Understanding the fundamentals of ARIMA. ARIMA modeling, model estimation & interpretation. Forecasting with regression and time series data. ARIMAX or dynamic regression models to build forecasting models with multiple regressors.
• Clustering: Introduction to clustering. Types of clustering & their uses. Data Standardization. K-Means clustering. Cluster profiling & validation

• Preview to machine learning: How do machines learn? Types of machine learning algorithm. Choosing a machine learning algorithm. Using R for machine learning.

Predicting House prices in Melbourne

In Melbourne, real estate sector is going through major fluctuations. As a result, the prices of residential properties have become really unpredictable. This has become a serious issue among real estate agencies. The agencies are not able to estimate the right price of the property hence not able to maintain consistent margins over the sales.

Homes1221, one such real state agency wants to predict the prices of the houses in Melbourne city using the data for 2014-2015 so that no house is being sold at unfair prices.

Predicting Churn for a DTH company

Telecommunications industry is one of those industries where attrition/churn rates of customers is really high. This is primarily due to availability of multiple options to users with a wide variety of offers and promotions running round the year. In any industry, the cost of acquisition of a customer is much higher than cost of customer retention. Hence, organizations lure their customers with attractive offers and schemes to retain them. However, it is difficult to identify customers who are on the verge of attrition.

Apprehensive Entertainment Pvt. Ltd. Is one such DTH Company that is facing such challenges. So the company needs to identify the customers who may leave the company so that they can target them with retention offers.

Forecast monthly sales for a retail franchise

Black Sheep is a dairy franchise which has number of stores across the globe in India, US, Europe, Dubai and Singapore. The sales were seen to increase when there were promotions or on holidays. But there were times when there were very low number of sales despite having promotions in the stores. It became difficult for the company to manages the stock accordingly in different stores.

Black Sheep decided to build a time series forecasting model which would help them predict the sales and so that they can accommodate their stocks in the respective store accordingly and all the stores has inventory stock according to the sales predicted for that particular store.

Customer segmentation for an insurance company

Modern technologies have brought the promotion of products and services to a qualitatively new level. Different customers tend to have specific expectations for the insurance business. Insurance marketing applies various techniques to increase the number of customers and to assure targeted marketing strategies. In this regard, customer segmentation proves to be a key method.

InsuredU, a car insurance company wants to do customer segmentation based on customer profile, vehicle profile, customer purchase behaviour etc. to help them build right marketing strategies for different customer segments.

Capstone Project 1

Apprehensive Entertainment Pvt. Ltd. Is a DTH Company. In was founded in the year 2011, with offices in US, Dubai and India. At present it has around 350, 000 worldwide subscribers.  Due to the rise in competition, many of its subscriber are leaving the company. Though this problem is across the countries but it is prominent in India. Mr. Srinivas (VP, Sales) has thought of an innovative idea for customer retention.

Srinivas want to execute a strategy for a select group of subscribers who are most likely to leave. He needs your help in identifying the customers who may leave the company.     

Capstone Project 2

As Italy is famous for wine, there are a lot of producers producing top quality wines. But to produce the best quality and best-tasted wine, it needs to analyse the amount of chemicals they need to club together to produce the same. So they need to classify the properties of wine according to the taste, quality etc.  Le Smith, a wine producer needs to analyse the chemicals grown and in what quantity it needs to be present in the wine. So the company hires you to cluster the wine data according to the wine properties so that they can segregate the wines accordingly.

Your task is to build an appropriate model to accomplish the business objective.   

Capstone Project 3

MRV International is a retail chain, situated in UK. It has 10 stores located across the country. Recently it is facing issues in inventory management. It wants to ensure that each store should have just the right amount of stock available such that there should not be any stock-outs and at the same time there should not be any extra inventory management cost incurred by the company.  In this case study, you will help MRV international by doing demand forecasting. You have to predict 3 months of sales for 50 different items at all the stores of MRV international.

Build an appropriate model to predict the claim amount of the  insurance