Business Analytics with R

www.edureka.co/r-for-analytics
View Business Analytics with R course details at www.edureka.co/r-for-analytics
Business Analytics with R

www.edureka.co/r-for-analyticsSlide 2 Twitter @edurekaIN, Facebook /edurekaIN, use #AskEdureka for Questions
Objectives
What is data mining
What is Business Analytics
Stages of Analytics / data mining
What is R
overview of Machine Learning
 What is Clustering
What is K-means Clustering
Use-case
At the end of this session, you will be able to

Slide 3 www.edureka.co/r-for-analyticsTwitter @edurekaIN, Facebook /edurekaIN, use #AskEdureka for Questions
Data mining ??
Generally, data mining is the process of studying data from maximum possible dimensions and summarizing it into
useful information
Technically, data mining is the process of finding correlations or patterns among dozens of fields in large data
generated from business
Or you can say, data mining is the process finding useful information from the data and then devising knowledge
out of it for improving future of our business
» Data ??
Data are any facts, numbers, or text is getting produced by existing system
» Information ??
The patterns, associations, or relationships among all this data can provide information
» Knowledge ??
Information can be converted into knowledge about historical patterns and future trends. For example summary of
sales in off season may help to start some offers in that period to increase sales

Business Analytics(BA)
Refers to the skills, technologies, practices for iterative study and investigation of historical business data to
gain insight and drive business planning
Study of data through statistical and operations analysis
Makes use of past data and statistical methods to understand business performance and hence makes us
take necessary steps to improve it
Injects intelligence into the business planning
Intersection of business and technology

Business Analytics
Why Business Analytics is getting popular these days ?
Cost of storing data Cost of processing data

Cross Industry standard Process for data mining ( CRISP – DM )
Stages of Analytics / Data Mining

Knowledge discovery and data mining ( KDD)
Stages of Analytics / Data Mining

What is R : Programming Language
 You do data analysis in R by writing scripts and functions
in the R programming language.
 R has also quickly found the following because
statisticians, engineers and scientists without computer
programming skills find it easy to use.
R is Programming Language

What is R : Data Analysis Software
 Data Scientists, Statisticians, Analysts, Quants, and
others who need to make sense of data use R for
statistical analysis, data visualization, and
predictive modelling.
 Rexer Analytics’s Annual Data Miner Survey is the
largest survey of data mining, data science, and
analytics professionals in the industry.
 It has concluded that R's popularity has increased
substantially in recent years.
R is Data Analysis Software

What is R : Environment for Statistical Analysis
 R language consists of functions for almost every
data manipulation, statistical model, or chart that a
data analyst could ever need.
 For statisticians, however, R is particularly useful
because it contains a number of built-in mechanisms
for organizing data, running calculations on the
information and creating graphical representations of
data sets.
R is Environment for Statistical Analysis

R : Characteristics
Effective and fast data handling and storage facility
A bunch of operators for calculations on arrays, lists, vectors etc
A large integrated collection of tools for data analysis, and visualization
Facilities for data analysis using graphs and display either directly at the computer or paper
A well implemented and effective programming language called ‘S’ on top of which R is built
A complete range of packages to extend and enrich the functionality of R

Data Visualization in R
This plot represents the
locations of all the traffic
signals in the city.
It is recognizable as
Toronto without any other
geographic data being
plotted - the structure of
the city comes out in the
data alone.

Who Uses R : Domains
 Telecom
 Pharmaceuticals
 Financial Services
 Life Sciences
 Education, etc

Machine Learning
We have so many algorithms for data mining which can be used to build systems that can read past data and can
generate a system that can accommodate any future data and derive useful insight from it
Such set of algorithms comes under machine learning
Machine learning focuses on the development of computer programs that can teach themselves to grow and change
when exposed to new data
Train data
ML
model
Algorithms

Types of Learning
Supervised Learning Unsupervised Learning
1. Uses a known dataset to make
predictions.
2. The training dataset includes
input data and response values.
3. From it, the supervised learning
algorithm builds a model to make
predictions of the response
values for a new dataset.
1. Draw inferences from datasets
consisting of input data without
labeled responses.
2. Used for exploratory data analysis
to find hidden patterns or grouping
in data
3. The most common unsupervised
learning method is cluster analysis.
Machine Learning

Common Machine Learning Algorithms
Types of Learning
Supervised Learning
Unsupervised Learning
Algorithms
 Naïve Bayes
 Support Vector Machines
 Random Forests
 Decision Trees
Algorithms
 K-means
 Fuzzy Clustering
 Hierarchical Clustering
Gaussian mixture models
Self-organizing maps

What is Clustering?
Organizing data into clusters such that there is:
 High intra-cluster similarity
 Low inter-cluster similarity
 Informally, finding natural groupings among objects
http://en.wikipedia.org/wiki/Cluster_analysis

K-means Clustering

www.edureka.in/hadoopSlide 19
K-Means Clustering
The process by which objects are classified into
a number of groups so that they are as much
dissimilar as possible from one group to another
group, but as much similar as possible within
each group.
The objects in group 1 should be as similar as
possible.
But there should be much difference between an
object in group 1 and group 2.
The attributes of the objects are allowed to
determine which objects should be grouped
together.
Total population
Group 1
Group 2 Group 3
Group 4

How it works
1. Given n object set, randomly initialize k cluster centers from the existing set
2. Assign the objects from the set to these randomly selected cluster center based on closets Euclidean distance
from the center.
3. Set the position of each cluster to the mean of all data points belonging to that cluster
4. Repeat steps 2-3 until cluster center changes no more and cluster size remains constant

We have marks of 17 students in a class. Their ratings are :
{1,2,2,4,5,6,6,7,8,10,10,11,11,12,13,13,13}
Group the students in three categories i.e. good, average and bad.
K-means example with one dimensional data

Randomly initialize 3 cluster centers:
Iteration 1
Good
(centroid=3)
Average
(centroid=2)
Bad
(centroid=1)
4,5,6,6,7,8,
10,10,11,11,
12,13,13,13
2,2 1

Iteration 1 summary
Cluster 1 (Good):
No 0f items = 14
Sum of items = 129
mean = 129/14 = 9
Cluster 1 (Average):
No 0f items = 2
Sum of items = 4
mean = 4/2 = 2
Cluster 1 (Bad):
No 0f items = 1
Sum of items = 1
mean = 1/1 = 1
Change
detected
Good Average Bad
(centroid=9) (centroid=2) (centroid=1)
New cluster center after iteration 1

Good
(centroid=9)
Average
(centroid=2)
Bad
(centroid=1)
6,6,7,8,
10,10,11,11,
12,13,13,13
2,2,4,5 1
Iteration 2

Cluster 1 (Good):
No 0f items = 12
Sum of items = 120
mean = 120/12 = 10
No 0f items = 4
Sum of items = 13
mean = 13/4= 3
Cluster 1 (Bad):
No 0f items = 1
Sum of items = 1
mean = 1/1 = 1
Change
detected
Good Average Bad
Change
detected
Iteration 2 summary

Good
(centroid=10)
Average
(centroid=3)
Bad
(centroid=1)
7,8,
10,10,11,11,
12,13,13,13
6,6,2,2,4,5 1
Iteration 3

Cluster 1 (Good):
No 0f items = 10
Sum of items = 108
mean = 108/11 = 11
No 0f items = 6
Sum of items = 25
mean = 13/4= 4
Cluster 1 (Bad):
No 0f items = 1
Sum of items = 1
mean = 1/1 = 1
Change
detected
Good Average Bad
Change
detected
Iteration 3 summary

Good
(centroid=11)
Average
(centroid=4)
Bad
(centroid=1)
8,
10,10,11,11,
12,13,13,13
7,6,6,4,5 1,2,2
Iteration 4 summary

Cluster 1 (Good):
No 0f items = 9
Sum of items = 101
mean = 108/11 = 11
No 0f items = 5
Sum of items = 28
mean = 28/5= 6
Cluster 1 (Bad):
No 0f items = 3
Sum of items = 5
mean = 5/3 = 2
No Change
detected
Good Average Bad
Change
detected
Change
detected
Iteration 4 summary

Good
(centroid=11)
Average
(centroid=6)
Bad
(centroid=2)
10,10,
11,11,
12,13,13,13
8,7,6,6,4,5 1,2,2
Iteration 5

Cluster 1 (Good):
No 0f items = 8
Sum of items = 93
mean = 93/8 = 12
No 0f items = 6
Sum of items = 36
mean = 36/6= 6
Cluster 1 (Bad):
No 0f items = 3
Sum of items = 5
mean = 5/3 = 2
Change
detected
Good Average Bad
No Change
detected
No Change
detected
Iteration 5 summary

Good
(centroid=12)
Average
(centroid=6)
Bad
(centroid=2)
10,10,
11,11,
12,13,13,13
8,7,6,6,4,5 1,2,2
Iteration 6

Cluster 1 (Good):
No 0f items = 8
Sum of items = 93
mean = 93/8 = 12
No 0f items = 6
Sum of items = 36
mean = 36/6= 6
Cluster 1 (Bad):
No 0f items = 3
Sum of items = 5
mean = 5/3 = 2
No Change
detected
Good Average Bad
No Change
detected
No Change
detected
Iteration 6 summary

G
O
O
d
A
v
g
B
a
d
10 10
11
11
12
13
13
13
4
5
6
6
7
8
1
2
2

Use Cases

www.edureka.co/r-for-analyticsTwitter @edurekaIN, Facebook /edurekaIN, use #AskEdureka for Questions
Demo
More Information on R setup and applications at:
http://www.edureka.in/blog/category/business-analytics-with-r/

 Module 1
» Introduction to Business Analytics
 Module 2
» Introduction to R Programming
 Module 3
» Data Manipulation in R
 Module 4
» Data Import Techniques in R
 Module 5
» Exploratory Data Analysis
 Module 6
» Data Visualization in R
Course Topics
 Module 7
» Data mining: Clustering Techniques
 Module 8
» Data Mining: Association rule mining and
Sentiment analysis
 Module 9
» Linear and Logistic Regression
 Module 10
» Annova and Predictive Analysis
 Module 11
» Data Mining: Decision Trees and Random forest
 Module 12
» Final Project Business Analytics with R class –
Census Data

Business Analytics with R

Recomendados

Recomendados

Mais conteúdo relacionado

Mais procurados

Mais procurados (20)

Destaque

Destaque (20)

Semelhante a Business Analytics with R

Semelhante a Business Analytics with R (20)

Mais de Edureka!

Mais de Edureka! (20)

Último

Último (20)

Business Analytics with R