Data Cleaning

This process is used to determine inaccurate, incomplete,

or unreasonable data and then improving the quality
through correction of detected errors and omissions. The
process may include format checks, completeness checks,
reasonableness checks, limit checks, review of the data to
identify outliers (geographic, statistical, temporal or
environmental) or other errors, and assessment of data by
subject area experts (e.g., taxonomic specialists).

1
Data Cleaning
These processes usually result in flagging, documenting
and subsequent checking and correction of suspect
records. Validation checks may also involve checking for
compliance against applicable standards, rules, and
conventions.
Data cleaning (or data cleansing) routines attempt to fill in
mission values, smooth out noise data while identifying
outliers and correct inconsistencies in the data.

2
Missing Data

Missing data is a common problem and people use

various methods to deal with it. There are statistical
techniques that generate values for missing data.
In statistics, missing data, or missing values, occur
-hen no data value is stored for variable in an
Observation. Missing data are a common
occurrence and can have a significant effect on the
conclusions that n be drawn from the data.
3
Outliers
An outlier is an observation point that is distant from other
observations. An outlier may be due to variability in the
measurement or it may indicate experimental error; the
latter are sometimes excluded from the data set.
An outlier is a piece of data that is an abnormal distance
from other points. In other words, it's data that lies outside
the other values in the set. If you had Pinocchio in a class
of children, the length of his nose compared to the other
children would be an outlier.
4
Outliers
For example, in following set of random numbers, 1 and 201 are
outliers: 1, 99, 100,101,103,109,110,201
Where 1 is extremely low value and 201 is an extremely high value.
Outliers can occur by chance in any distribution, but they often
indicate either measurement error or that the population has a
heavy-tailed distribution. In the former case one wishes to discard
them or use statistics that are robust to outliers, while in the latter
case they indicate that the distribution has high skewness and that
on should be very cautious in using tools or intuitions that assume a
normal distribution.

5
Outliers
Two key phases of outlier mining are:
1) Identifying the inconsistent data in the input database,
and
2) The extraction of the expected number of outliers or
deviant data points.
The process of outlier analysis is complicated in cases of
data involving multi-dimensions and categorical attributes.
Identifying exceptions with respect to numeric attributes
area lot easier in relation to their categorical counterparts
6
Types of Outliers
Outliers can be classified into three categories:
1)Point Outliers: If an individual data point can be considered anomalous with
respect to the rest of the data, then the datum is termed as a point outlier. This is
the simplest type of outlier and it is the focus of the majority of research on outlier
detection.
2) Contextual Outliers: If an individual data instance is anomalous in a specific
context (but not otherwise), then it is termed as a contextual (conditional) outlier.
The notion of a context is induced by the structure of the data set and has to be
specified as a part of the problem formulation. Each data point is defined with two
sets of attributes:
i) Contextual Attributes: They are used to determine the context (or
neighborhood). For example, time in time-series data, or longitude and latitude in
spatial data sets.
ii) Behavioral Attributes: They are used to define other characteristics of the data
point, specific to the problem in hand. 7
Types of Outliers

3. Collective outliers
If a collection of data points is anomalous with
respect to the entire data set, it is termed as a
collective data outlier. The individual data points
inside the collective outlier may not be outliers by
themselves alone, but their occurrence together as
a collection is anomalous. This type can occur only
in the data set in which the data points are some
how related. 8
DATA VISUALISATION
Data visualization refers to technologies that support
visualization and sometimes interpretation of data and
information at several points along the data processing
chain.
It includes digital images, GIS, graphical user interfaces,
graphs, virtual reality, dimensional presentations, videos,
and animation. Visual tools can help to identify
relationships such as trends.
Data visualization is easier to implement when the
necessary data are in a data warehouse or, better yet, in a9
multidimensional special database or server.
DATA VISUALISATION

Data visualization can also:

1) Identify areas that need attention or
improvement.
2) Clarify which factors influence customer
behavior.
3) Help you understand which products to place
were.
4) Predict sales volumes.
10
Functions of Data Visualization
Most good data visualization allows the user some key
abilities:
1) Ability to compare data.
2) Ability to control scale (looks from a high level or drill
down to detail).
3) Ability to map the visualization back to the detail data
that created it.
4) Ability to filter data to look only at subsets or sub
regions of it at a given time.
11
Applications of Big Data Analytics

Big data applications can help companies to make better

business decisions by analyzing large volumes of data and
discovering hidden patterns. These data sets might be from
social media, data captured by sensors, website logs,
customer feedbacks, etc. Organizations are spending huge
amounts on big data applications to discover hidden
patterns, unknown associations, market style, consumer
preferences, and other valuable business information

12
Applications of Big Data Analytics
Health care
There is a significant improvement in the healthcare domain by personalized
medicine and prescriptive analytics due to the role of big data systems.
Researchers analyze the data to determine the best treatment for a particular
disease, side effects of the drugs, forecasting the health risks, etc.
Media and entertainment
The media and entertainment industries are creating, advertising, and
distributing their content using new business models. This is due to customer
requirements to view digital content from any location and at any time. The
introduction of online TV shows, Netflix channels, etc. is proving that new
customers are not only interested in watching TV but are interested in
accessing data from any location.

13
Applications of Big Data Analytics
Internet of Things
IoT devices generate continuous data and send them to a server on a daily basis. These
data are mined to provide the interconnectivity of devices. This mapping can be put to
good use by government agencies and also a range of companies to increase their
competence.
Manufacturing
Predictive manufacturing can help to increase efficiency by producing more goods by
minimizing the downtime of machines. This involves a massive quantity of data for such
industries. Sophisticated forecasting tools follow an organized process to explore
valuable information for these data.
Government
By adopting big data systems, the government can attain efficiencies in terms of cost, output, and
novelty. Since the same data set is used in many applications, many departments can work in
association with each other. Government plays an important role in innovation by acting in all
these domains.

14
Data Mining Process
Any business problem will examine the raw data to build a model that
will describe the information and bring out the reports to be used by
the business. Building a model from data sources and data formats is
an iterative process as the raw data is available in many different
sources and many forms. Data is increasing day by day, hence when a
new data source is found, it can change the results.

15
Machine Learning Algorithms

The idea behind creating this guide is to

simplify the journey of aspiring data scientists
and machine learning enthusiasts across the
world. This will help to understand to work on
machine learning problems and gain from
experience. It can be provided that a high-
level understanding of various machine
learning algorithms along with R & Python
codes to run them.
16
Machine Learning Algorithms

Supervised Learning
How it works: This algorithm consist of a target /
outcome variable (or dependent variable) which
is to be predicted from a given set of predictors
(independent variables). Using these set of
variables, we generate a function that map
inputs to desired outputs. The training process
continues until the model achieves a desired
level of accuracy on the training data. 17
Machine Learning Algorithms

Unsupervised Learning
How it works: In this algorithm, we do not have any
target or outcome variable to predict / estimate. It
is used for clustering population in different
groups, which is widely used for segmenting
customers in different groups for specific
intervention. Examples of Unsupervised Learning:
Apriori algorithm, K-means.

18
Machine Learning Algorithms
Reinforcement Learning:
How it works: Using this algorithm, the machine is trained
to make specific decisions. It works this way: the
machine is exposed to an environment where it trains
itself continually using trial and error. This machine
learns from past experience and tries to capture the
best possible knowledge to make accurate business
decisions. Example of Reinforcement Learning: Markov
Decision Process
19
Challenges in data driven decision
making
• Common Challenges Of Data-Driven Decision Making
• A Lack Of Infrastructure And Tools. ...
• Poor Quality Data. ...
• Siloed Data. ...
• A Lack Of Organization-Wide Buy-In. ...
• Not Knowing How To Use Your Data. ...
• Being Unable To Identify Actionable Data. ...
• Too Much Focus On Data.

20
FINANCIAL ANALYTICS
Financial analytics is defined as the analysis of the
financial impact of business analytics. One aspect of
financial analytics is the opportunity of working with net
(final) figures, which are derived after taxes, duties, levies
and penalties, or capital charges are charged to the
business. It embodies the versatility of the risks of doing
business and also translates such risks to net turnout.

21
Objectives of Financial Analytics
The goal/objectives of financial analytics are as follows:
1) To increase the sales and revenue while decreasing capital
investment and operating cost and improving business processes.
2) To supports the financial planning, budgeting and forecasting.
3) To create the financial and managerial reports, financial budgets,
and forecasts can be created with the help of the financial analytics
processes.
4) To also support the profitability management, production and
service controlling, overhead cost controlling, payment-behavior
analysis, and investment & risk management.
5) To include the features, for analyzing product and service costs
and for legal and managerial consolidation.
22
Understanding Risk and Risk Analytics

Risk may be defined as the possibility of suffering an injury

or loss. It is present in every field or situation. In the
context of Business or Financial World', it represents the
uncertainty associated with an investment. In other words,
risk is the possibility that the actual return on an
investment may be different from the expected return.

23
Applications of Risk Analytics
1) Fraud Detection and Prevention: This includes analysis and control of risk,
identification, and bas removal of fraudulent factors in the banking and financial
services sectors.
2) Market Liquidity Risk Assessment: Effective gr market liquidity risk
management helps banks meet but their cash flow requirements.
3) Real-Time Situational Awareness: It helps organizations to achieve immediate
and long-term goals.
4) Product Portfolio Analysis: This is the study of the products of an organization
to improve market performance, and also minimize the product failure rate.
5) Credit Risk Analysis: It involves analysis of business and industry category,
financial statement analysis for strengths and weaknesses, cash flow analysis and
projections, and analysis of credit risk.
6) Benchmarking Risk Management: It involves comparison of risks measured
with the international risk management standard
24
Methods used in Risk Analytics
The various methods used in risk analytics are as follows:
1) Value at Risk (VaR),
2) Stress Testing,
3) Downside Risk,
4) Economic Factor Sensitivities,
5) Risk Decomposition, and 6) Scenario and "What-if"
Analyses

25
Importance of Risk Analytics
1) Get the Details: Risk analytics helps take the guesswork out of managing risk-
related issues by using a range of techniques and technologies to extrapolate
insights, calculate likely scenarios, and predict future events.
2) Understand the Complexity: An organization's exposure to risk is influenced by
increasing volumes of structured data such as databases-and unstructured data-
such as websites, social media, and blogs-that are available to an organization
internally and externally. Risk analytics can be leveraged to integrate this data into
a single, unified view, gather valuable information, and enable actionable insights.
3) Cross the Divide: In their scramble to build effective risk strategies, teams often
fail to consider the overall impact to the organization. Risk analytics pulls data
across the organization into one central platform, helping create a truly enterprise-
wide approach.
(4) Lay the Groundwork: Risk is such a wide-ranging issue, spilling across
organizational barriers, that it can be hard to know exactly what to do with risk
related in
26
Credit Risk Analysis
Credit risk is the possibility of a loss resulting from a borrower's
failure to repay a loan or meet contractual obligations. Traditionally,
it refers to the risk that a lender may not receive the owed principal
and interest, which results in an interruption of cash flows and
increased costs for collection. Excess cash flows may be written to
provide additional cover for credit risk. When a lender faces
heightened credit risk, it can be mitigated via a higher coupon rate,
which provides for greater cash flows.

27
Fraud Detection, Prevention and Analytics

Fraud is definitely not a recent phenomenon unique to

modern society, nor it is even unique to mankind. Animal
species also engage in what could be called fraudulent
activities. A person can classify the behavior as displayed
by, e.g., chameleons, stick insects, apes, and others rather
as manipulative behavior instead of fraudulent activities,
since wrongful or criminal are human categories

28
Methods of Fraud Detection
i) Sampling: It is mandatory for certain processes of fraud detection.
Sampling will be more effective where there a lot of data population
involved. But still, it has its own disadvantage. Sampling may not be
able to fully control the fraud detection as it takes only few
populations into consideration. Fraudulent transactions do not occur
randomly therefore an organization need to test all the transactions
to effectively detect fraud.
ii) Ad Hoc: It is nothing but finding out fraud by means of a
hypothesis. It allows the management to explore. They can test the
transactions and find out if there are any opportunities for fraud to
take place. They can have a hypothesis to test and find out if there
is any fraudulent activity occurring and then they can investigate on
the same. 29
Methods of Fraud Detection

iii) Repetitive or Continuous Analysis: It means creating and setting

up scripts to run against big volume of data to identify the frauds as
they occur over a period of time. Run the script every day to go
through all the transactions and get periodic notification regarding
the frauds. This method can help in improving the overall efficiency
and consistency of the organization’s fraud detection processes.

30
Methods of Fraud Detection in Banks
The basic approach to fraud detection is an expert sed approach,
meaning that it builds on the patience, intuition, and business or
domain knowledge of the fraud analyst. Such an expert sed
approach typically involves a manual ventilation of a suspicious
case, which may have signaled, e.g., by a customer complaining
charged for transactions he did not do. Such a transaction may
indicate a new fraud mechanism to have been discovered or
developed fraudsters, and therefore requires a detailed estimation
for the organization to understand and sequent address the new
mechanism.

31
Steps of fraud Detection

Step 1- Perform SWOT

Step 2- Prepare a dedicated fraud Management Team
Step – 3 Build or Buy option.
Step 4- Clean data
Step 5- Lay out relive

32
Frauds can be prevented by using following methods

1) Neural Nets: They are a form of modelling in which a computer attempts to

predict good and positive outcomes by use of previous and current activity. Neural
nets are not unique to fraud or fraud prevention; they are used in many different
industries today. The neural net is typically the primary engine behind more
sophisticated fraud-scoring applications.
2) Bayesian Models: These are a different approach to modelling where
behavioral characteristics are input to create predictive outcomes. A good practical
example of this approach is the way a physician diagnoses an illness by looking
for certain characteristics or symptoms. From a fraud perspective, the model
attempts to diagnose fraud by looking for certain symptoms or characteristics of
fraud.
3) Regression Analysis: This analysis compares the correlation ship between two
data elements with one do being fixed and the other changing in relationship to
the fixed data element (e.g., how many different names are associated with a
33
given phone number).
Frauds can be prevented by using following methods

4) Monitoring: It is the act of actively reviewing transactional or

account history to detect anomalies and outliers for review and
investigation. Monitoring also includes the benchmarking of existing
fraud controls for accuracy and reliability.
5) Historical Analysis "What if': This is an analytical tool to create
new rule sets or strategies to run "what if" analysis against known
outcomes on historical data.
6) Shadowing: It is an analytical tool to create and put a strategy or
rule set in live production to run in parallel to existing processes to
determine the effectiveness of a new fraud strategy.
7) Finally, once the model has been built, it will be interpreted and
evaluated by the fraud experts. 34
Analytics in Banking and Financial Services

Analytics is helping the banking industry become smarter in

managing the myriad challenges it faces. While basic reporting and
descriptive analytics continues to be a must-have for banks,
advanced predictive and prescriptive analytics are now starting to
generate powerful insights, resulting in significant business impact.
Advanced analytics-backed solutions are enabling banks to not only
manage the increasing cost of compliance, but also the risk (both
monetary and reputational) of non-compliance. Product and portfolio
optimization modelling is helping banks to achieve profitable growth
in an environment with significant volatility across asset classes and
rising losses in traditional banking products
35
Analytics in Banking and Financial Services

The institutions execute transactions directly with consumers, rather

than corporations or other banks. Services offered include: savings
and checking accounts, mortgages, personal loans, debit cards,
credit cards, and so forth.
The two immutable truths of retail banking are
1) The customer is at the heart of retail banking.
2) Knowing the customer and driving profitability through this
knowledge is the lifeblood of retail banking.

36
Analytics in Wealth Management
Wealth Management Analytics allows your business to
formulate investment strategies for your high-net worth
clientele. A flexible and powerful tool, IP uses simulation
and optimization techniques to come up with the best
portfolio model for your clients. It covers the entire gamut
of Financial Planning from prospecting to generating
proposals for the customers.

37
The features of wealth management analytics

1)Captures future contribution, distribution and life goals over

planning horizon
3) Identification of the optimal frontier for current asset classes
under consideration as historical risk/return
4) Create projections to develop and access the plan
5) Analytics to consider various what-if scenarios
6) Automated tools to create, analyze and optimize asset allocation
models based on industry-standard and self-defined indices and
criteria.
7) Predictive Analytics to calculate the probability of achieving the
goal 8) Interactive Proposal Generation Engine
38
The value proposition of wealth management analysis

1) Decision support tools help investment advisors

strategize wealth preservation and growth by considering
various scenarios using interactive screens.
2) Helps to models to their clients.
manage interactions prospects/clients in a streamlined
fashion.
3) Enables Investment Advisors to present pre-defined
4) Allows for a futuristic outlook

39
Demand Planning
Demand planning, which forms part of an overall management strategy and
affects many individuals and functions within an organisation, is just one element -
albeit an important one – within demand management. Demand planning is the
management process within an organisation which enables that organisation to
tailor its capacity, either production or service, to meet variations in demand or
alternatively to manage the level of demand using marketing or supply chain
management strategies to smooth out the peaks and troughs.
Demand drives the entire supply chain from suppliers to manufacturing,
marketing, inventory, distribution and service to customers. An organisation needs
to be able to forecast demand accurately but to do this needs to understand
demand patterns, and how factors such as new products, competition, and
changing market conditions affect these patterns.

40
Forecasting
Forecasting is the process of making predictions based on past and present data
and most commonly by analysis of trends. A commonplace example might be
estimation of some variable of interest at some specified future date. Prediction is
a similar, but more general term. Both might refer to formal statistical methods
employing time series, cross-sectional or longitudinal data, or alternatively to less
formal judgmental methods. Usage can differ between areas of application: for
example, in hydrology the terms "forecast" and "forecasting" are sometimes
reserved for estimates of values at certain specific future times, while the term
"prediction" is used for more general estimates, such as the number of times
floods will occur over a long period.
Risk and uncertainty are central to forecasting and prediction; it is generally
considered good practice to indicate the degree of uncertainty attaching to
forecasts. In any case, the data must be up to date in order for the forecast to be
as accurate as possible. In some cases the data used to predict the variable of
interest is itself forecast
41
Model Building
The model building process
1. Define the purpose of your model, the problem you are trying to
solve, or the story you are trying to tell. ...
2. Determine the model boundary. ...
3. Map the model. ...
4. Build the model. ...
5. Test the model. ...
6. Create an interface. ...
7. Share the model.

42
Supply Chain Analytics
Supply chain analytics refers to the processes organizations use to gain insight and extract
value from the large amounts of data associated with the procurement, processing and
distribution of goods. Supply chain analytics is an essential element of supply chain
management (SCM). The discipline of supply chain analytics has existed for over 100
years, but the mathematical models, data infrastructure, and applications underpinning
these analytics have evolved significantly. Mathematical models have improved with better
statistical techniques, predictive modeling and machine learning. Data infrastructure has
changed with cloud infrastructure, complex event processing (CEP) and the internet of
things. Applications have grown to provide insight across traditional application silos such
as ERP, warehouse management, logistics and enterprise asset management.
An important goal of choosing supply chain analytics software is to improve forecasting and
efficiency and be more responsive to customer needs. For example, predictive analytics on
point-of-sale terminal data stored in a demand signal repository can help a business
anticipate consumer demand, which in turn can lead to cost-saving adjustments to
inventory and faster delivery.
43
Supply Planning
Supply planning is the component of supply chain management
involved with determining how to best fulfill the requirements
created from the demand plan. The objective is to balance supply
and demand in a manner that achieves the financial and service
objectives of the enterprise.
The supply planning process commences with an approved demand
plan. The demand plan is a sum of all the sales reviewed and
approved channel, regional and/or customer forecasts.
The approved demand plan goes through a demand translation
step. If you are using a supply chain planning software system, such
as Demand Caster, the demand translation step will automatically
allocate the demand plan to the location where the demand for a
specific customer/item relationship is typically fulfilled. 44
Procurement and Strategic Sourcing
Strategic sourcing is a procurement process that connects data
collection, spend analysis, market research, negotiation, and
contracting. It stops short of the actual purchase of and payment for
goods and services.
Strategic sourcing can be customized to meet a customer’s specific
needs, but its main goal is to leverage a single, integrated system to
enhance profitability.
Strategic sourcing best practices include: digitizing documents,
participating in a digital business network, and automating
workflows

45
 Procurement and Strategic Sourcing
The process can be broken down into four steps:

• Data collection and spend analysis: Spend analysis concentrates supplier data into
one source, letting organizations know exactly what’s being spent where and
presenting the opportunity to streamline vendors.
• Supplier discovery and RFx: Sourcing becomes a strategic advantage when
organizations can access supplier data through a digital business network, allowing
them to request RFPs and have suppliers compete for their business.
• Negotiations and contracting: Automated tools can speed workflows, simplify the digital
signature process, and create an electronic repository of contracts where organizations
can set renewal alerts.
• Implementation and optimization: When sourcing is automated and digitized,
organizations can move faster, build in feedback loops for continual optimization, and
constantly evaluate suppliers to make sure they're getting the best sourcing
agreements possible.

46
Inventory Modeling
Inventory management is a systematic approach to sourcing, storing,
and selling inventory—both raw materials (components) and finished
goods (products). In business terms, inventory management means the
right stock, at the right levels, in the right place, at the right time, and at
the right cost as well as price. The stock might be kept “in-house”, which
means in the vicinity or close by for sure-fire use; or it could be held in a
removed stockroom or dispersion place for some time later. Except for
firms using in the nick of time strategies, usually, the expression “stock”
infers a put-away amount of merchandise that surpasses what is
required for the firm to work at the current time. The significance of
inventory management models is in the exactness it gives.
47
Aggregate Planning and Resource
Allocation decisions
An organization can finalize its business plans on the
recommendation of demand forecast. Once business plans are
ready, an organization can do backward working from the final sales
unit to raw materials required. Thus annual and quarterly plans are
broken down into labor, raw material, working capital, etc.
requirements over a medium-range period (6 months to 18 months).
This process of working out production requirements for a medium
range is called aggregate planning.

48
Make / Buy decisions
A make-or-buy decision is an act of choosing between
manufacturing a product in-house or purchasing it from an external
supplier.
Also referred to as an outsourcing decision, a make-or-buy decision
compares the costs and benefits associated with producing a
necessary good or service internally to the costs and benefits
involved in hiring an outside supplier for the resources in question.
To compare costs accurately, a company must consider all aspects
regarding the acquisition and storage of the items versus creating
the items in-house, which may require the purchase of new
equipment, as well as storage costs.
49
Human resource analytics (HR analytics)

It is defined as the analysis of human resources

(employees), which embodies the entire life-cycle of an
employee, such as recruitment, managing performance,
incentives, and employee engagement. Using the right
metrics can improve policies and procedures, increase
team members' satisfaction and retention, focus employee
training and support, improve morale, reduce costs, and
increase productivity. The activities impacted by human
capital involve recruitment, training, employee
relationships, employee satisfaction, and turnover.
50
HR analytics
Human resource analytics is an area in the field of analytics that
refers to applying analytic processes to the human resource
department of an organization in the hope of improving employee
performance and therefore getting a better return on investment. HR
analytics does not just deal with gathering data on employee
efficiency. Instead, it aims to provide insight into each process by
gathering data and then using it to make relevant decisions about
how to improve these processes. The goal of human resource
analytics is to provide an organization with insights for effectively
managing employees so that business goals can be reached quickly
and efficiently.

51
Levels of HR Analytics

There are three levels of HR analytics:

1) Descriptive Analytics: The use of data to record a
particular aspect of HR and provide information on what
has been happening to, e.g., labor turnover or absence
rates. Descriptive analytics cover the following types of
data:
ii) Basic Workforce Data: Demographic data, working
arrangements, absence and sickness, turnover, health and
safety, pay.
iii) People Development and Performance Data 52
Methods for HR Analytics
1) Clustering: HR analytics team will investigate hidden group
patterns with the help of clustering methods. For example, which
common people characteristics can predict better sales
performance? Which cluster of recruitment sources can predict
better people retention?
2) Driver Analysis: HR analytics team will try to understand hidden
relationships. Most of the time, they use the word 'impact' to explain
relationships between events or people/business characteristics.
For example, what is the impact of poor engagement on client
satisfaction? What is the impact of sales training on business
revenue development?
53
Methods for HR Analytics

3) Risk Analysis: HR analytics team will try to understand

probabilities or the likelihood of occurring events. For example,
which high performers are at risk to leave in the next 24 months?
Will the reduction of training investments increase the risk of
employee turnover?
4) Forecasting: HR analytics team will try to understand future trend
lines, based on historical patterns. For example, what will be the
employee turnover in the coming 3, 6 or 12 months? What will be
the typical first six months' time-to productivity trend line of a call
center newcomer without call center experience?

54
Metrics Used for HR Analytics for human
resource analytics
Monthly Turnover Rate: Metrics used are: (number separations
during month / average number of of employees during month) x
100.
Revenue per Employee = total revenue / total number of
employees.
Yield Ratio = percentage of applicants for recruitment source that
make it to a determined stage of the application process. a
Human Capital Cost = Pay + Benefits + Contingent Labor
Equivalents Cost / Full Time
HR to Staff Ratio= Employees / Human Resources Team Members.
Cost to Hire= Training + Advertising + Lost Productivity + Time /
55
employee salary (ranges from 50-250%).
Intuition Vs. Analytical Thinking
The analytical style of thinking is step-wise and logical. It usually
attempts to break a problem or issue into its constituent parts both
to understand and to address or solve it. ... On the other hand, the
intuitive style of thinking is driven more by gut-feel and confidence
derived from experience.
The analytical style of thinking is step-wise and logical. It usually
attempts to break a problem or issue into its constituent parts both
to understand and to address or solve it. It is usually very methods-
driven, following thought-through (and sometime research-derived)
models and frameworks. Examples of these are the real options
analysis for planning technology investments, and business model
canvas, used during new business development). 56
People Analytics
People analytics is defined as the deeply data-driven and goal-
focused method of studying all people processes, functions,
challenges, and opportunities at work to elevate these systems and
achieve sustainable business success.
Advanced analytics can dramatically improve the way organizations
identify, attract, develop, and retain talent. However, many
organizations still make those decisions based on instinct and
intuition. People Analytics helps managers and senior talent leaders
unlock the power of data—increasing rigor, reducing bias, and
improving performance.

57
RECRUITMENT ANALYTICS
Recruitment analytics refer to a systematic design of
tracking, measuring, correlating, and collating candidate
and employee data points in order to improve business
performance. Recruitment analytics also enables you to
provide unique opportunities to the existing employees
and attractive packages for interested candidates in order
to optimize talent acquisition.

58
Steps of Recruitment Analytics
Step 1: Finding the Right Candidate: Companies can use
data and analytics to spot the best prospective employees
from the starting stage of the hiring process. Especially,
soft skills are easier to measure using data backed hiring
algorithms tailored to be used around a company's work
culture. This would evaluate the candidate for skills
required to excel and grow within an organization.

59
Steps of Recruitment Analytics
Step 2: Refining Sourcing: A company involved into
recruitment for some time, would already know how to do
best sourcing for required talent. Predictive analytics takes
this to a further level by helping the company to gain
insights through each resource which is being used. Data
can be mined from social media sites, popular job
aggregators, etc., to find the key elements need to be
focused on.

60
Steps of Recruitment Analytics
Step 3: Gathering Performance Data: With the data mining
technologies of present era, an organization can start
evaluating the satisfaction levels of new hires from day
one. It provides a live view to see if the new crew. from
managers to CEO, are meeting the expected standards in
their respective jobs. This can even be used to measure
the overall employee engagement, which would provide
signals if there is a risk of turnover

61
Steps of Recruitment Analytics
Step 4: Analyze Overall Recruitment Experience: Analytics
can be used communicating flaws in the boarding process
to the HR team, thus fixing it up at the earliest by
addressing the issues from the beginning future issues
can be avoided
Step5: Help with profile Tuning
Step 6: identity best candidate sources
Step 7: Analyze Market Trends.

62
Finding out selection bias
Selection bias is the bias introduced by the selection of individuals,
groups, or data for analysis in such a way that proper randomization
is not achieved, thereby failing to ensure that the sample obtained is
representative of the population intended to be analyzed. It is
sometimes referred to as the selection effect. The phrase "selection
bias" most often refers to the distortion of a statistical analysis,
resulting from the method of collecting samples. If the selection bias
is not taken into account, then some conclusions of the study may
be false.

63
Finding out selection bias
Sample selection bias is a type of bias caused by
choosing non-random data for statistical analysis. The bias
exists due to a flaw in the sample selection process, where
a subset of the data is systematically excluded due to a
particular attribute. The exclusion of the subset can
influence the statistical significance of the test, and it can
bias the estimates of parameters of the statistical model.

64
Predicting the performance and turnover
Employee turnover refers to the percentage of workers
who leave an organization and are replaced by new
employees. It is very costly for organizations, where costs
include but not limited to: separation, vacancy, recruitment,
training and replacement. On average, organizations
invest between four weeks and three months training new
employees. This investment would be a loss for the
company if the new employee decided to leave the first
year.

65
ON-BOARDING ANALYTICS

Once the right candidates have been hired, they

need to be properly on-boarded to ensure they are
primary business goals and the overall mission of
the company. New hires need to have the best first
impression of the HR professional, as well as their
manager, and of the company.

66
ON-BOARDING ANALYTICS
4 C's of On-Boarding Analytics 4C's of on-boarding analytics are as
follows:
1 Compliance: It is the lowest level and includes teaching
employees basic legal and policy-related rules and regulations.
2) Clarification: It refers to ensuring that employees understand their
expectations. new jobs and all related
3) Culture: It is a broad category that includes providing employees
with a sense of organizational norms - both formal and informal.
4) Connection: It refers to the vital inter-personal relationships and
information networks that new employees must establish.

67
STAFFING ANALYTICS

Staffing is the process of acquiring, deploying, and

retaining a workforce of sufficient quantity and quality to
create positive impacts on the organization’s
effectiveness. Digital information and analytics
programmes allow staffing firms to use big data and
statistics to gain insight into future trends and learn from
past practices.

68
STAFFING ANALYTICS
Analytics enable staffing firms to measure ROI, track the
effectiveness of marketing campaigns, and make
important business decisions. Integrating the Applicant
Tracking System (ATS) and Customer Relationship
Management (CRM) systems yields integrated information
covering different business processes and provides a
better overall view of the business as a whole.

69
Role of Predictive Analytics in Staffing
Traditionally, predictive analytics has helped companies to address the basic
business questions of who, when and why. However, when applied to the staffing
industry, predictive analytics helps to anticipate and optimize
1)Talent Acquisition: Helps to identify who is the top talent? When should they be
contacted? Why is this requisition/ job opportunity attractive to this top talent?
2) Talent Pipeline Planning: Predictive analytics can optimize a talent pipeline by
leveraging macroeconomic and talent data to ascertain key factors that can lead
to better resource allocation. For instance, identifying the best locations to invest
in recruitment campaigns for certain skills
3) Job-response the recruitment process, predictive analytics helps organizations
optimize their job-postings response. Data analysis can provide companies with
custom recommendations and tailored best practices to help firms achieve better
responses to their jobs postings based upon factors such as duration, location,
occupation, and industry.

70
PERFORMANCE SKILL GAP ANALYTICS

Performance Analytics Performance analytics are used to

visualize data that is collected over time. This data reveals
trends, which can be used to make real-time adjustments
and improve the business functions. Performance
Analytics provide secure, simple access to Key
Performance Indicators (KPIs) and metrics that enable
taking initiative about improving business services. Rich,
interactive visualizations empower accurate data needed
to drive service delivery quality and efficiency.

71
PERFORMANCE SKILL GAP ANALYTICS

Performance analytics are also used to align resources,

systems, and employees to strategic objectives and
priorities. By its nature, performance analytics is backward
looking. It is assessing past performance and how an
employee has performed throughout the previous year. As
such it is not that useful for predicting future performance.
If done well, performance analytics provide a positive
experience that contributes to the overall employment and
career development experience and helps to strengthen
the relationship between line managers and their reports.
72
Skill Gap Analytics
Skill Gap is the difference in the skills required on the job and the
actual skills possessed by the employees. Skill gap presents an
opportunity for the company and the employee to identify the
missing skills and try to gain them. Employees are recruited by
companies to work on fulfilling company objectives. Hence, people
with the correct skill sets are recruited by companies. However,
often it happens that employees lack certain knowledge & training
which creates a skill gap. Because of this the employee is unable to
perform the complete job.
Skills Needed - Skills Possessed = Skills Gap

73
COMPENSATION BENEFIT ANALYTICS

Compensation analytics gives management and the Board

of Directors the tools to exercise effective oversight of
benefit programmes and the ability to evaluate the
adequacy and appropriateness of these benefits, both
individually and in the aggregate. Organizations that
integrate compensation analytics into the decision-making
process generally have a balanced framework that reflects
short-term and long-term planning needs.

74
COMPENSATION BENEFIT ANALYTICS

To support business planning cycles, organizations use

compensation structures and labour cost estimates based
on data from a single time period benchmark with a
smaller sample size, as well as an analytics benchmark
with a larger sample size that reflects data collected from
thousands of available salary surveys.

75
TRAINING LEARNING ANALYTICS

Analytics refers to information, so it is quite simply

the study of data, learning (or training) analytics is
the study of learning data. It is important because it
allows the company to determine its effectiveness
and impact. There is new development in training
technology over the last few years is the advent of
training analytics systems.

76
Purposes of Training Analytics
The most important purposes of training analytics are as follows:
1) Its focus is on determining the extent to which the goals of
training programmes have been achieved.
2) It helps in spotting the strength and weaknesses of the training
programme.
3) It is conducted to collect data and help in assisting the future
programmes related to marketing field.
4) It ensures whether or not the training programme was specific as
per the requirements.
5) It helps in selecting the specific participant or members for future
programmes.
77
Purposes of Training Analytics
6) It also helps in spotting the most appropriate members
among all the participants.
7) It helps in preparing database that can provide support
to the management in making decisions.
8) It helps to determine the validity of tests, cases and
exercises used in the training programme.
9) It facilitates the comparison of cost and benefits of
training programme.
10) It determines whether or not the training programmes
were the right solution for the training need identified. 78
Learning Analytics
One of the crucial psychological processes is learning,
through which human behavior is determined. It is a never-
ending procedure which is everlasting. So, learning can be
defined the summation of as behavioral transformations
which are the outcome of knowledge attained during the
training. An individual attains knowledge and practicality
from the training process which acts as feedback to the
individual and a reference for future responses.

79
Learning Analytics
Learning analytics is the science and art of gathering,
processing, interpreting and reporting data related to the
efficiency, effectiveness and business impact of
development programmes designed to improve individual
and organizational performance and inform stakeholders.
It is the process of analyzing and decoding large amounts
of data to develop a better sense of performance and
interaction, for the purposes of enhancing HR practices
and techniques.

80
Promotion Analytics
Promotion is a term which is very commonly used in every
organization. When a person moves to a higher position
because of his/her good performance, this movement from
a lower to a higher-level position is called promotion.
Promotion basically occurs when the internal employees
are entitled to a higher post in the same organization and
are selected for the same. Besides that, a company can
also hire candidates externally for the vacancies available
in the organization.

81
Promotion Analytics
A promotion analytics is used to determine whether a
given employee will be successful or not if promoted to
senior level position. Promotion analytics can be done
through predictive analytic and graph theory analytics.
Those companies that are using such predictive analytic
and graph theory analytics recorded 33 percent more
success from their newly promoted employee than those
who are leveraging these insights.

82
Succession Planning Analytics
The vital area of HR department is succession planning.
As the day’s past, every human being gets retired or they
situations, leave the organization. For such organization
has to prepare themselves to fill up that vacant position as
soon as possible so that productivity will continue.
Succession planning plays a vital role in this aspect.
Predicting who is capable of succeeding the important
position is very difficult.

83
Succession Planning Analytics
The performance of the selected candidates has to be analyzed. It
is not possible to track the employee's performance over the years.
But HR analytics assist the Senior HR management to track the
performance of the star performers who are capable to succeeding
the said vacancy. A series of analysis are done through predictive
analytics, quantitative modelling, and reviewing the performance of
the employees for years then the outcome helps the HR
management to decide who is capable of succeeding such an
important position. Thus HR analytics again becoming the key
element for strategic planning as succession planning is the
provision of in-house replacements and retention of key talents.
84
 Steps of Succession Planning
• Pre-Planning
(Analyzing top jobs, future jobs and critical success factors)
• Assessment
Interview, test and seek feedback about candidate and consult
his superior)
• One-on-One and Group Meetings
Facilitation of advancement plan meeting
• CEO Discussion
Validation and acceptance
• Ongoing Review
By diagnostic feedback and linking to employee development,
performance development, self-development 85
COMPLIANCE ANALYTICS
Compliance is about more than monitoring adherence
rules and regulations outside the organization. True
compliance starts on the inside - with the attitudes are
actions of the people who conduct business every da and
those who supervise and manage them. The inherent in
giving employees decision-making autonomy is
inseparable from the performance company demands from
them.

86
Attrition Analytics
Employee attrition employees leaving Quantization is a very
common phenomenon. What is pertinent is to know how many
employees (as a percentage of the total workforce) are leaving and
the reasons for their leaving. The percentage of employees leaving
to the percentage of the existing workforce is known as the
employee turnover.
Every industry has a benchmark for employee attrition and any
deviation from that should be of concern to the management. If
many employees are leaving for better prospects, it means that the
organization is unable to retain skilled people. The reasons could be
many, from poor pay to poor working conditions to poor prospects
for growth. Every organization invests in its human resources and a
drain of employees is a loss, especially if skilled workers are leaving
87
in large numbers.
Retention Analytics
Retention refers to how many current employees stick
around over a given period of time. Retention of human
resources means creating a big picture of organizations as
Great Place to Work' and facilitating opportunities for total
learning, growth and wholesome development of people in
organization in it strive towards excellence and value
creation through human capital. This is accomplished by
ensuring appropriate policies, tools and techniques for
maximum utilization and retention of competent personnel.

88
Retention Analytics
Employee retention is about proactively identifying and
understanding which valuable employees are employees
at risk of leaving, and when and why they would leave.
Analytics can help to marry employee data, company data,
and market data to predict and interpret top performing
employees behavior, giving competitive insight for the
retention strategies.

89
 Steps for Employee Retention
1) Churn Prediction: Create analytics models to identify employees at risk of
leaving, so managers can rapidly change work conditions and behavior to keep
top people from leaving.
2) Attrition Root Cause: Correlate resignation with factors such as promotion wait
time, pay increase, commute, performance, attendance, employee development,
while cross referencing attrition variables with historic data and past resignations
to see which indicators are the most relevant. Based on the findings, one can
begin effectively targeting and fine-tuning the retention strategies.
3) Resignation Segments: Compare how the resignation rate varies across
locations, functions, tenure, age groups, diversity groups, and other variables, to
ensure programme investments are targeted where they will deliver the most
significant results.

90
HRMS / HIRS
Human Resources Management System (HRMS) is the flagship
project of Center for Modernizing Government Initiative – CMGI, a
society under the General Administration Department, Government
of Odisha. HRMS is a database – and - application software to carry
out personnel transaction of government employees online through
Internet. This aims to be the primary vehicle of transactions of
government employees. HRMS is the repository of all the service
records of the employees of Government of Odisha; through HRMS,
an employee may apply for leave, loan, or send their reports,
requests or grievances. They can receive the sanction or reply
online from their office.
91
HRMS
HRMS software automatically prepares all accounts and registers of
an employee, like Service Book, Leave Account, Loan Account,
Salary Account, Incumbency Chart etc., retrieving relevant data from
transactions. It also helps the superannuating employees to prepare
their pension papers on the click of buttons and help authorities to
process pension papers easily and quickly. Past transactions are
captured as legacy data and incorporated into the database, and
subsequent transactions are recorded in real time. The Service
Book of each employee is the most complete repository of such
transactions. Hence, service data of each employee from the
service book is the backbone of HRMS database.
92
HIRS
A human resource information system (HRIS) is software that
provides a centralized repository of employee master data that the
human resource management (HRM) group needs for completing
core human resource (core HR) processes.
An HRIS stores, processes and manages employee data, such as
names, addresses, national IDs or Social Security numbers, visa or
work permit information, and information about dependents. It
typically also provides HR functions such as recruiting, applicant
tracking, time and attendance management, performance appraisals
and benefits administration. It may also feature employee self-
service functions, and perhaps even accounting functions.

93
94
95
96
97
98
HR Dash Boards
The HR Dashboard is the visual representation of the
metrics that an HR manager needs to keep a track of to
judge the performance of different organizational
departments. Apart from the records of personnel, it
includes the business performance dashboard, marketing
dashboard, sales dashboard, finance dashboard, etc.

99
HR Dashboard Metrics
1. Employee Headcount: The number of employees
working in different departments at different positions
along with the number of years of experience is recorded.
2. Recruitment and Staffing: The number of positions lying
vacant in different departments for different levels in the
organization along with the number of vacancies already
filled in by the personnel is maintained.
3. Payroll: The payroll cost is recorded Viz. Payroll cost
per team, payroll cost by different levels, payroll cost by
different departments in the organization.
100
HR Dashboard Metrics
4. Employee Time: The employee’s arrival time, departure time,
working time, absenteeism is recorded.
5. Attrition: Keeping a track of the number of layoffs, either forcefully
or voluntary from different departments in the organization.
6. Incentives and Commissions: The amount of incentives to be paid
with respect to the cost incurred on an employee, Incentives to be
paid out of a given budget is maintained by the HR team.
7. Exit Interviews: The record of the number of employees leaving
the organization and keeping a proper data about the reason behind
their departure.

101
HR Dashboard Metrics
8.Skills: The individual competencies along with the possession of
unique skills is recorded.
9.Compensation: The amount of compensation to be paid to each
employee performance wise and category wise is computed.
10.Termination: The data about the number of employees
terminated from their positions and the reason for the same is
recorded.
11. Training metrics: The training programmes to be carried out to
incorporate essential skills and abilities in an individual is
summarized.

102
Power BI
Power BI is an interactive data visualization software
developed by Microsoft with primary focus on
business intelligence.
Power BI provides cloud-based BI (business
intelligence) services, known as "Power BI
Services", along with a desktop based interface,
called "Power BI Desktop". It offers data warehouse
capabilities including data preparation, data
discovery and interactive dashboards.
103
Key components of the Power BI ecosystem
comprises
Power BI Desktop
The Windows-desktop-based application for PCs and desktops, primarily
for designing and publishing reports to the Service.
Power BI Service
The SaaS-based (software as a service) online service. This was formerly
known as Power BI for Office 365, now referred to as PowerBI.com, or
simply Power BI.
Power BI Mobile Apps
The Power BI Mobile apps for Android and iOS devices, as well as for
Windows phones and tablets.
Power BI Gateway
Gateways used to sync external data in and out of Power BI and are
required for automated refreshes. In Enterprise mode, can also be used by
Flows and PowerApps in Office 365.
104
Key components of the Power BI ecosystem
comprises
Power BI Embedded
Power BI REST API can be used to build dashboards and reports into the custom
applications that serves Power BI users, as well as non-Power BI users.
Power BI Report Server
An on-premises Power BI reporting solution for companies that won't or can't
store data in the cloud-based Power BI Service.
Power BI Premium
Capacity-based offering that includes flexibility to publish reports broadly across
an enterprise, without requiring recipients to be licensed individually per user.
Greater scale and performance than shared capacity in the Power BI service
Power BI Visuals Marketplace
A marketplace of custom visuals and R-powered visuals

105
106
BASICS OF MARKETING ANALYTICS

The practice of measuring, managing, and analyzing the

marketing performance of a firm so that the return on
investment (ROI) can be optimized and increased is called
"marketing analytics". Marketers can achieve a greater
level of efficiency and reduce wastage of web marketing
funds by understanding the concept of marketing
analytics.

107
BASICS OF MARKETING ANALYTICS
A lot of vital information related to the preference and
requirements of the customer can be obtained with the
help of marketing analytics apart from the most obvious
reason of marketing activities, i.e., increasing sales and
generating leads. Most of the organizations are still not
able to understand the potential of marketing analytics,
despite having several advantages from it

108
The following reasons induce the use of
marketing analytics
1) Getting information related with new marketing trends;
2) Identifying successful programmes and evaluating their
reasons of success;
3) Analyzing trends over time
4) Completely analyzing the ROI of each programme
5) Forecasting the outcomes.

109
Components of Marketing Analytics
1) People: The marketing analytics process is created, executed,
and managed by people who own it. In most marketing
organizations, the process owner is the Chief Marketing Officer
(CMO) or the marketing director.
2) Steps: The marketing analytics process consists of a sequence of
steps.
3) Tools and Technology: While the marketing analytics process is
not necessarily complex, tools and technology help marketing
organizations deliver greater value faster than they ordinarily might.
4) Input and Output: Data feeds the process, with insights and
decisions as the output of the process.
110
 3-Step Methodology for Marketing Analytics

At the heart of every data-driven decision lie three essential

elements that contextualize data and allow real business decisions
to surface from the sea of numbers that are produced from
marketing activity. These are:
1) Understand: Analysis for the sake of analysis, o provides little
value and consumes precious time in a marketer's busy day.
Develop a basic understanding of what you plan to accomplish with
marketing analytics, i.e., improved customer knowledge, new
segments, channel effectiveness, etc. It might even be appropriate
to back into data requirements by defining a core business
challenge or key question that the organization would like to answer
with the analysis. 111
3-Step Methodology for Marketing Analytics

2) Execute: After developing a core understanding of the data itself

and translating that into business strategy, it is time to execute act.
Often involving some additional refinement of the information
produced in the Understand Phase, execution may be done through
reporting engines, Analytical models and even spreadsheets.
3) Monitor: It is critical for marketers to close the loop on analytical
exercises through ongoing monitoring of a standard set of
Knowledge Performance Indicators (KPIS). This might be
accomplished via a dashboard or periodic report, but the goal
should be to standardize metrics over time, so you can establish a
benchmark for significant changes in the data that may alert you to
opportunities to optimize marketing results. 112
 MARKETING DECISION MODELS

Models are tools which allow management to structure a problem,

identify and evaluate its determinants and solution options, and
select the best solution. Models vary in their objectives, structures,
assumptions, complexities, required inputs, algorithms (such as
estimation or optimization procedures), and outputs.
A marketing decision model provides a way to visualize the
sequences of events that can occur following alternative decisions
(or actions) in a logical framework, as well as the outcomes
associated with each possible pathway.

113
Characteristics of Marketing
Decision Models are the characteristics of marketing
decision models:
1) comprise generalized set processes which undertaken
the marketers, so make particular marketing decision.
These processes are followed marketers for their
respective business decisions.
2) purpose, which defines the construction and scope
reason for applicability. example, "adbudg' helps
managers in arriving at advertising budgets. The clustering
model is useful in identifying attractive market segments.
114
Characteristics of Marketing
3) Assumptions: All models contain assumptions, explicit implicit and
unlike models, marketing decision models require assumptions be
made One of the assumptions marketing decision models make
market forces same over the period of study or that market changes
are predictable and measurable.
4) Relationship between Variables: Mathematical functions are used
by the marketers to represent the relationship between the
variables. Variables are those aspects of a marketing phenomenon
that are not fixed. In a marketing system many things can vary, i.e.,
firm's sales, intensity of competition, inflation, etc.

115
Types of Marketing Models
On the basis of structural characteristics, it is being
classified into …
Verbal models
Graphical models
Mathematical models

116
117
118
Text analysis and Search Analysis

119
SEA

120
Customer Profiling

121
122
Basics of R
R is a software environment which is used to analyze
statistical information and graphical representation. R
allows us to do modular programming using functions.
Our R tutorial includes all topics of R such as introduction,
features, installation, r studio ide, variables, datatypes,
operators, if statement, vector, data handing, graphics,
statistical modelling, etc. This programming language was
named R, based on the first name letter of the two authors
(Robert Gentleman and Ross
123
Why R Programming ?
"R is an interpreted computer programming language
which was created by Ross Ihaka and Robert Gentleman
at the University of Auckland, New Zealand." The R
Development Core Team currently develops R. It is also a
software environment used to analyze statistical
information, graphical representation, reporting, and data
modeling. R is the implementation of the S programming
language, which is combined with lexical scoping
semantics.

124
125
History of R Programming

The history of R goes back about 20-30 years ago. R was

developed by Ross lhaka and Robert Gentleman in the
University of Auckland, New Zealand, and the R
Development Core Team currently develops it. This
programming language name is taken from the name of
both the developers. The first project was considered in
1992. The initial version was released in 1995, and in
2000, a stable beta version was released.

126
127
Features of R programming
R is a domain-specific programming language which aims to do data analysis. It has some unique
features which make it very powerful. The most important arguably being the notation of vectors. These
vectors allow us to perform a complex operation on a set of values in a single command. There are the
following features of R programming:

1. It is a simple and effective programming language which has been well developed.
2. It is data analysis software.
3. It is a well-designed, easy, and effective language which has the concepts of user-defined,
looping, conditional, and various I/O facilities.
4. It has a consistent and incorporated set of tools which are used for data analysis.
5. For different types of calculation on arrays, lists and vectors, R contains a suite of operators.
6. It provides effective data handling and storage facility.
7. It is an open-source, powerful, and highly extensible software.
8. It provides highly extensible graphical techniques.
9. It allows us to perform multiple calculations using vectors.
10. R is an interpreted language.
128
Response Model Concepts
Market response models are intended to help scholars and
managers understand how consumers individually and collectively
respond to marketing activities, and how competitors interact.
Appropriately estimated effects constitute a basis for improved
decision making in marketing. We review the demand and supply of
market response models and we highlight areas of future growth.
We discuss two characteristics that favor model use in practice, viz.
the supply of standardized models and the availability of empirical
generalizations.

129
 Clustering Algorithms

Clustering or cluster analysis is an unsupervised learning

problem.
It is often used as a data analysis technique for
discovering interesting patterns in data, such as groups of
customers based on their behavior.
There are many clustering algorithms to choose from and
no single best clustering algorithm for all cases. Instead, it
is a good idea to explore a range of clustering algorithms
and different configurations for each algorithm.
130
Perceptual Maps
A perceptual map is a visual representation of the
perceptions of customers or potential customers about
specific attributes of an organization, brand, product,
service, or idea.
This diagrammatic technique (perceptual mapping) asks
participants to place products relative to one another along
2 or more axis.
The resulting map shows how consumers see the
strengths of competing products in a particular market.
131
Why create a Perceptual Map?
Understanding what your customers think about you and
your competitors is crucial to your success. A perceptual
map helps organizations to:
Understand the thoughts and behaviors of consumers.
Gain insight into their competition.
Track market trends.
Identify gaps in the market.

132
How to create a Perceptual Map
Assemble a group of consumers or across a range of relevant demographics. The
quality of the outcomes is dependent on the insight of the participants, and a
diverse group helps to gain a better insight into the market.
Gathering a large group in one location can be expensive and difficult to
coordinate. Using an online collaborative tool such as Group map enables
facilitators to engage consumers in different places at different times and
effortlessly combine the results to get an overview.
There are 3 general steps in perceptual mapping. The time required to complete
the map will vary on the size of the group, and the focus of the session. However,
there is no reason why the map can’t be completed in less than 20 minutes.
Development of a comprehensive plan in response to the map will require further
effort.

133
How to create a Perceptual Map
• Set Dimensions
Define the attribute dimensions on the perceptual map template.
• Brainstorm
Gather products from a group of consumers.
• Position
Position products on the perceptual map template.
• Share
Report on the outcomes and monitor as part of your strategy.

134
Google Analytics
Google Analytics is a web analytics service offered by Google that
tracks and reports website traffic, currently as a platform inside the
Google Marketing Platform brand.
Google Analytics is used to track website activity such as session
duration, pages per session, bounce rate etc. of individuals using
the site, along with the information on the source of the traffic. It can
be integrated with Google Ads, with which users can create and
review online campaigns by tracking landing page quality and
conversions (goals). Goals might include sales, lead generation,
viewing a specific page, or downloading a particular file.

135
WEB ANALYTICS
Web analytics is the application of Business Analytics (BA)
activities to web-based processes, including commerce.
This term is used to describe the application of BA to Web
sites. The tools and methods are highly e visual in nature.
Web analytics (or Web Intelligence) refers to analysis of
web data (known as clickstream data). Such analyses are
useful in market research and competitive intelligence.

136
WEB ANALYTICS
Web analytics is the process of analyzing the behavior of
visitors to a Web site. The use of Web analytics is said to
enable a business to attract more visitors, retain or attract
new customers for goods or services, or to increase the
dollar volume each customer spends.
Web analytics provides information about the number of
visitors to a website and the number of page views. It
helps gauge traffic and popularity trends which is useful for
market research.
137
Most web analytics processes down to four
essential stages
1) Collection of Data: This stage is the collection of the basic, elementary data.
Usually, this data is counts of things. The objective of this stage is to gather the
data.
2) Processing of Data into Information: These stages usually take counts and
make them ratios. although there still may be some counts. The objective of this
stage is to take the data and conform it into information, specifically metrics.
3) Developing KPI: This stage focuses on using the ratios (and counts) and
infusing them with business strategies, referred to as Key Performance Indicators
(KPI). Many times, KPIs deal with conversion aspects, but not always. It depends
on the organization.
4) Formulating Online Strategy: This stage is concerned with the online goals,
objectives, and standards for the organization or business.

138
CLICK STREAM ANALYTICS

Clickstream analytics is the analysis of user's Web activity

through the items they click on a page. Clickstream
analytics is a special type of web analytics that gives
special attention to clicks.
A clickstream is the recording of the parts of the screen a
computer user clicks on while web browsing or using
another software application. As the user clicks anywhere
in the webpage or application, the action is logged on a
client or inside the web server, as well as possibly the web
browser, router, proxy server or ad server.
139
SOCIAL MEDIA ANALYTICS
Social media analytics is the practice of gathering data
from blogs and social media websites and analyzing that
data to make business decisions. The most common use
of social media analytics is to mine customer sentiment in
order to support marketing and customer service activities.
Social media analytics describes the activities of
monitoring and recording, analyzing, and interpreting the
results of interactions and associations among people,
topics, and ideas. It is the process of gathering data from
blogs and social media sites and analyzing that data to
make business decisions.
140
 Sentiment Analysis
Sentiment analysis (sometimes known as opinion mining or emotion
AI) refers to the use of natural language processing, text analysis,
computational linguistics, and biometrics to systematically identify,
extract, quantify, and study affective states and subjective
information.
Sentiment analysis is widely applied to voice of the customer
materials such as reviews and survey responses, online and social
media, and healthcare materials for applications that range from
marketing to customer service to clinical medicine.
Sentiment analysis is important for businesses because it identifies
ways for them to improve their operations and the products they
offer.
141
Importance of in Social Media Analytics
Social media sentiment analysis can be an excellent
source of information and can provide insights that can:
1) Determine marketing strategy
2) Improve campaign success
3) Improve product messaging
4) Improve customer service
5) Test business KPIs
6) Generate leads

142
Types of Sentiment Analysis
1) Manual Processing: Human interpretation of sentiment
is definitely the most mature and accurate judge of
sentiment. However, it still is not 100% accurate. Very few
vendors still use this process without the additional use of
a tool. This is due to the prolific growth of social media.
2) Keyword Processing: Keyword processing algorithms
assign a degree of positivity or negativity to an individual
word, then it gives and overall percentage score to the
post. For example, positive words, great, like, love or
negative words: terrible, dislike.
143
Types of Sentiment Analysis

3) Natural Language Processing (NLP): NLP refers to

computer systems that process human language term its
meaning. NLP understands that several words make
phrase, several make sentence ultimately, sentences
convey NLP works by analyzing language for meaning.
NLP systems are for in number areas such converting
speech language It likened interpret English language
rules one was taught English class.

144
Sentiment Analysis Methods
The existing sentiment analysis can classify the problem of
text classification in different forms such as
1.Document-Level Sentiment Analysis: This technique
every document and unsupervised learning.
2. Sentence-Level Sentiment Analysis: This technique
based on specialization of documents which ca sentences
and sentence contains opinion behind it.

145
Sentiment Analysis Methods
3.Aspect-Based Sentiment Analysis: Attribute(aspect)
based classification is performed customer a processor,
every attribute opinion product is considered for better
understanding sentiments.
4)Here identification of sentences which contain
comparative opinions are filter out and opinions are
extracted.
5) Sentiment Lexicon Acquisition: This most popular
crucial resource for the sentiment algorithms. sentiment
lexicon performed by manually, diction approaches and 146
corpus-based approaches.
ANALYTICS IN DIGITAL DECODING CONSUMER INTENT

To maximize your marketing investment, you have to

connect with your consumers on their terms. You have to
be hyper relevant. Social Code delivers intelligence from
real-time audience data on platforms like Facebook,
Twitter, Instagram and Pinterest. We then take immediate
action with paid media to drive brand perception and
business results. With a combination of automation and a
strategic services group, we enable marketers to make
better overall marketing decisions and drive growth.

147
Signals of Intent
As marketers, our mission is to discover consumer's
intent. Consumer's intent means the reason our why the
individual will purchase a product or service now or in the
future. In search marketing, we learn about the consumer's
intent through his or her keywords entered into the search
engine.
"Search data, captured across e-commerce, pricing
comparison and product review sites, is one of the
strongest signals of intent and best sources for new
customer acquisition."
148
DECODING CUSTOMER SENTIMENTS

Customer sentiment analysis is a complex, multi-step

process. The various tools work in somewhat different w
ways. Typically, they rely on natural language processing,
which turns text like tweets and blogs into a format that
computers can understand then artificial intelligence
applies algorithms against that data. A basic task in
sentiment analysis is classifying the polarity of a given
piece of material, which means determining whether the
expressed opinion is positive, negative or neutral.

149
Text Mining

Text mining, also referred to as text data mining,

roughly equivalent to tear high-quality information
from text. High-quality information is typically
derived through the devising of pattern trends
through means such as statistical pattern learning.
Text Mining refers to the whole process of
extracting interesting patterns from text data in
order to discover knowledge.
150
Text Mining
Text mining usually involves the process of structuring the
input text, deriving patterns within the structured data, and
finally evaluation and interpretation of the output. 'High
quality in text mining usually refers to some combination of
relevance, novelty, and interestingness.

151
Text Mining from Opinion Platform
Generally, individuals and companies are always
interested in other's opinion like if someone new product,
then firstly, he/she tries to know the reviews, ie. what other
people think about the product and based? Similarly;
companies also excavate deep for consumer reviews.
Digital ecosystem has a plethora for same in the form on
those reviews, he/she takes the decision of blogs, reviews,
etc.

152
 PREDICTIVE ANALYTICS
Predictive analytics is an area of data mining that deals
with extracting information from data and using it to predict
trends and behavior patterns. Often the unknown event of
interest is in the future, but predictive analytics can be
applied to any type of unknown, whether it be in the past,
present or future. For example, identifying suspects after a
crime has been committed, or credit card fraud as it
occurs.

153
PREDICTIVE ANALYTICS
The core of predictive analytics relies on capturing
relationships between explanatory variables and the
predicted variables from past occurrences, and exploiting
them to predict the unknown outcome. It is important to
note, however, that the accuracy and usability of results
will depend greatly on the level of data analysis and the
quality of assumptions.

154
Predictive Analytics Process
1) Define Project: It define the project outcomes, deliverables,
scoping of the effort, business objectives, and identify the data sets
which are going to be used.
2) Data Collection: Data mining for predictive analytics prepares
data from multiple sources for analysis. This provides a complete
view of the customer interactions.
3) Data Analysis: It is the process of inspecting, cleaning,
transforming, and modelling data with the objective or discovering
useful information, and arriving at conclusions.
4) Statistics: Statistical analysis enables to validate the
assumptions, hypotheses and test them with using standard
statistical models. 155
Predictive Analytics Process
5) Modelling: Predictive modelling provides the ability to
automatically create accurate predictive models about
future. There are also options to choose the best solution
with multi-model evaluation.
6) Deployment: Predictive model deployment provides the
option to deploy the analytical results into the everyday
decision-making process to get results, reports and output
by automating the decisions based on the modelling.
7) Model Monitoring: Models are managed and monitored
to review the model performance to ensure that it 156 is
providing the results expected.
Application of Predictive Analytics

1) Banking and Financial Services: The financial industry,

with huge amounts of data and money at stake, has long
embraced predictive analytics to detect and reduce fraud,
measure credit risk, maximize cross-sell/up-sell
opportunities and retain valuable customers.
Commonwealth Bank uses analytics to predict the
likelihood of fraud activity for any given transaction before
it is authorized, i.e., within 40 milliseconds of the
transaction initiation.

157
Application of Predictive Analytics

2) Oil, Gas and Utilities: Whether it is predicting

equipment failures and future resource needs,
mitigating safety and reliability risks, or improving
overall performance, the energy industry has
embraced predictive analytics with vigor.

158
PREDICTIVE MODELS
A short definition of a predictive model is "Using data to
make decisions". A longer definition might be "Using data
to make decisions and to take actions using models that
are empirically derived and statistically valid". Predictive
modelling is a commonly used statistical technique to
predict future behavior.

159
PREDICTIVE MODELS
Predictive modelling solutions are a form of data mining
technology that works by analyzing historical and current
data and generating a model to help predict future
outcomes.
Predictive models are created whenever data is used to
train a predictive modelling technique. It can be
mathematically expressed as:
Data + Predictive Modelling Technique = Predictive Model.

160
Types of Predictive Models
Predictive modelling means developing models that can
be used to forecast or predict future events. In business
analytics, models can be developed based on logic or
data. Basically, classified into two categories
 Logic Driven Predictive Models
 Data Driven Predictive Models

161
Logic Driven Predictive Models
A logic-driven model is one based on experience,
knowledge, and logical relationships of variables and
constants connected to the desired business performance
outcome situation. The question here is how to put
variables and constants together to create a model that
can predict the future. Doing this requires business
experience. Model building requires an understanding of
business systems and the relationships of variables and
constants that seek to generate a desirable business
performance outcome.
162
Data Driven Predictive Models
Logic-driven modelling is often used as a first step to establish
relationships through data-driven models (using data collected from
many sources quantitatively establish model relationships).
These models enable one to make predictions of what might
happen in the future based on past observations. Combining this
with domain knowledge, expertise, and business logic enables
analysts to make data-driven decisions using these predictions,
which is the ultimate outcome of predictive analytics. The data here
is data which has already been observed in the past and has been
collected over a period of time for analysis.

163
Data Driven Predictive Models- can be
categorized into

1.Retail Pricing Markdowns Model

2.Modelling Relationships and Trends in Data

164
Retail Pricing Markdowns Model
Most department stores and fashion retailers clear their
seasonal inventory by reducing prices. The key question
they face is what prices should they set-and when should
they set them to meet inventory goals and maximize
revenue? For example, suppose that a store has 1000
summer tees of a certain style that go on sale April 1 and
wants to sell all of them by the end of June. Over each
week of the 12-week selling season, they can make a
decision to discount the price.

165
Modelling Relationships and Trends in Data

Many applications of business analytics involve modelling

relationships between variables or predicting the future
value of a single variable over time. Understanding both
the mathematics and the descriptive properties of
relationships is important in using predictive analytical
models. One often begins by creating a chart the data to
understand it and chooses the appropriate type of
functional relationship to incorporate into an analytical

166
MODELS INVOLVING UNCERTAINTY

Because predictive analytical models are based on

assumptions about the future and incorporate variables
that most likely will not be known with certainty, it is usually
important to investigate how these assumptions and
uncertainty affect the model outputs. This is one of the
most important and valuable activities for using predictive
models to gain insights and make good decisions.

167
What-If Analysis
"If an input variable, assumption or parameter value is
changed, then what will happen to the solution?" This is
structured in what-if analysis. Some examples are as
follows:
1) If the cost of carrying inventories increases by 20 per
cent, then what will be the total inventory cost?
2) If the advertising budget increases by 10 per cent, then
what will be the market share?

168
Tools for What-If Analysis

There are three different tools available in Excel

perform what-if analysis:
1.Scenarios
2.Data tables
3)Goal seek

169
Scenario Manager and Scenarios
A Scenario is a set of values that Excel saves and can
substitute automatically on worksheet. One can create and
save different groups of values as scenarios and then
switch between these scenarios to view the different
results.
If several people have specific information that one wants
to use in scenarios, he/she can collect the information in
separate workbooks, and then merge scenarios from the
different workbooks into one.
170
Goal-Seeking Analysis
Goal-seeking analysis determines the values of the inputs in order
to find out the expected output (goal). It is also known as backward
solution approach.
There are many examples of goal-seeking analysis. Some of them
are as follows:
1) In order to touch 20 per cent growth rate, what will be the annual
budget of a company in year 2016?
2) For reducing the waiting time less than 10 minutes in the
emergency room, how many nurses are required?

171
Goal-Seeking Analysis

Goal-seeking analysis reverses the direction of the

analysis done in what-if and sensitivity analysis.
Instead of observing how the changes in a variable
affect other variables, goal-seeking analysis (also
called how can analysis) sets a target value (a
goal) for a variable and then repeatedly changes
other variables until the target value is achieved.

172
APPLICATIONS OF ANALYTICS BUSINESS AREAS

The telecommunications industry within the sector of

information and communication technology is made up of
all Telecommunications/Telephone companies and internet
service providers and plays the crucial role in the evolution
of mobile communications and the information society. The
telecom sector continues to be at the epicenter for growth,
innovation, and disruption for virtually in any industry.

173
ANALYTICS IN TELECOME

Mobile devices and related broadband connectivity

continue to be more and more embedded in the fabric of
society today and they are key in driving the momentum
around some key trends such as video streaming, Internet
of Things (IoT), and mobile payments.
Telecom companies that want to be innovative and
maximize their revenue potential must have the right
solution in place so that they can harness the volume,
variety and velocity of data coming into their organization
and leverage on actionable insights from that data.
174
ANALYTICS IN TELECOME

Telecom companies are sitting on terabytes of data

that are stored in silos and scattered across the
organization. For simpler and faster processing of
only relevant data, telcos need advanced analytics
driven data solution that will help them to achieve
timely and accurate insights using data mining and
predictive analytics.

175
Analytics in Location based Intelligence Marketing

Location-based intelligence marketing is a direct marketing

strategy that uses a mobile device's location to alert the
device's owner about an offering from a near-by business.
Leveraging data analytics to drive location-based
marketing is picking up strong momentum across the
globe and for solid reasons like:
1) It allows retailers to provide the right offer, at the right
time, in the right location, to the right customer a win-win
for both the retailers and their shoppers.
176
Analytics in Location based Intelligence
Marketing
2) Analytics in location-based marketing helps retailers to cost-
effectively, and more accurately, target their customers with offers
that resonate with them.
3) This type of data-driven marketing works well for retailers,
especially if they have thousands of retail outlets throughout certain
geographical locations, making it challenging and costly.
4) The power of data analytics for location-based marketing can be
used to market to all the audiences in specific locations.
5) It helps to analyze data from many channels ,online, social, web,
in-store to detect new market trends and change in consumer
demand.
177
Analytics Packaged Goods (CPG)
Consumer Packaged Goods (CPG) or Fast-Moving
Consumer Goods (FMCG) are products that are sold
quickly and at relatively low cost. For example, non-
durable goods such as soft drinks, toiletries, over-the
counter drugs, processed foods and many other
consumables. In contrast, durable goods or major
appliances such as kitchen appliances are generally
replaced over a period of several years.

178
Analytics in Utilities

The utilities sector is a category of stocks for

utilities such as gas and power. The sector
contains companies such as electric, gas and
water firms, and integrated providers. Because
utilities require significant infrastructure, these firms
often carry large amounts of debt with a high debt
load.

179
Analytics helps organizations in utilities sector to

1) Identify data quality requirements and performance across key

areas such as customer, meter, product, tariff and industry flows.
2) Identify opportunities for revenue optimization and reduce
leakage including contract and billing alignment.
3) Validate the operation of estimation and billing routines to ensure
customers are billed accurately, estimation routines generate
realistic bills and unusual tariffs and billed volumes are identified.
4) Respond to issues that affect their business using significant
volumes of data to understand what happened, when, why, who did
it and potentially what might happen in the future.

180
Analytics in Health Care
The healthcare industry is being transformed continually
by the biological and medical sciences, which hold
considerable potential to drive change and improve health
outcomes. However, healthcare in industrialized
economies is now poised on the edge of an analytics
driven transformation. The field of analytics involves "the
extensive use of data, statistical and quantitative analysis,
explanatory and predictive models, and fact-based
management to drive decisions and options."

181
Analytics in Health Care

1.Comparative effectiveness
2.Disaster planning
3.Patient flow
4.Radio frequency identification
5.Genetics

182
Analytics in Online Retail
Online retailing is often referred to web retailing, internet
retailing, electronic retailing or e-tailing. In this retail
classification, the retailer and customers communicate
with each other in a typically non personal way via some
type of electronic, interactive system, generally a computer
facilitated by the Internet. Online retailing has many
dimensions such as Business to-Business (B2B),
Business-to-Commerce (B2C),C2C along with e-
payments, e procurement etc.

183