Data Science and big data Analysis

Day One 

Introduction to R

• Exploratory data analysis with R.
• Load the data into R
• Data query in R
• Data processing in R
• Clean up the raw data
• Dimensional reduction

Facilitate Good Analytical Thinking through Data Visualization

• Check the dataset properties
• Plot data distributions
• Identify outliers in the data

Day Two 

Work with Large, Unstructured Data Sets

• Mining unstructured data
• Pre-processing unstructured data in preparation
• Describe a group of documents with a document term matrix

Dealing with the Additional Complexities of Big Data

• Examine the MapReduce and Hadoop architectures
• Merge R and Hadoop with RHadoop

PREDICTING OUTCOMES WITH REGRESSION TECHNIQUES

• Estimating future values with linear and logistic regression
• Modeling the relationship between an output variable and several input variables
• Correctly interpreting coefficients of continuous and categorical data

Regression Techniques

• Overcoming issues of volume with RHadoop
• Creating regression modules for RHadoop

Day Three

Classification Techniques

• Automate the naming of new data items
• Predict target values using decision trees
• Build a model from existing data for future predictions
• Combine predictions of trees and random forests in RHadoop

Evaluate the Performance of the Model

• Visualize model performance with a ROC curve
• Evaluation of workbooks using confusion matrices

Uncover patterns in complex data

• Identification of previously unknown clusters within the dataset
• Customer market segmentation
• Determine similarity with appropriate distance scales
• Build tree
• Compilation of text documents and Tweets

Discover Connections

• Capture important connections with social network analysis
• Explore how social networking results can be used in marketing

Day Four 

Use transaction data

• Building and evaluating association rules
• Capture real customer preferences in transactional data
• Calculating support, confidence, and lift
• Distinguish between actionable, trivial, and inexplicable rules
• Build recommendation engines
• Cross-selling, up-selling and exchange as incentives
• Leverage recommendations based on collaborative filtering

Day Five

Implement Analytics

• Expand analytical capabilities
• Break down big data analytics into manageable steps
• Integrate analytics into existing business processes
• Spark, MLib, and Mahout Machine Learning Review

Publication Policies

• Examining ethical questions related to privacy in big data
• Disseminate results to different types of stakeholders