Introduction to Data Analytics

Data Analytics, Data Science, and

Machine Learning
 Learning Objectives

By the end of this lesson, you will be able to:

Define data science and machine learning

Differentiate between data science, machine learning, and

data analytics
Introduction to Data Science
 Data Science

Data science is the study of data, which involves gathering, storing, analyzing, and
plotting data, to effectively extract useful information.

Aim: Gain meaningful insights from both structured and unstructured data.
 Data Science

Preparation and
Data cleansing analysis

Trend forecast Machine learning

and data analytics
 Types of Data Science

Data Science

Data Analytics Machine Learning Data Mining

 Data Analytics

Data analytics is the process of examining and analyzing raw data sets to:

Draw conclusion Derive information

Derive insights from raw data

sources
 Machine Learning

Learns from patterns in the past Predicts outcomes

using a set of algorithms accurately
Data Mining

● Data mining is the process of analyzing data from

different perspectives.

● It summarizes data into useful information.

● It helps increase revenue and cut costs.

Data Science, Data Analytics, and Machine Learning
 Data Science and Data Analytics

Forecasts the future

based on past patterns

Data Scientist

Extracts meaningful insights

from various data sources

Data Analyst
 Machine Learning

Machine learning creates systems that can learn from the data.

It is the ability of machines to predict outcomes based on patterns in the past.

 Machine Learning

Leverages various algorithms to

train the machine

ML Engineer
 Data Science and Machine Learning

Extracts useful information from

collected data sets Understands data from a
business point of view

Gathers data from various

sources
Provides accurate predictions to
improve key business decisions

Data Scientist
Understanding Data Science
 Understanding Data Science

A data scientist combines both domain and technology perspectives.

 Understanding Data Science

Works with data from video and

Knows multiple analytical
social media sources
functions

Data Scientist

Has a sound knowledge of technologies such as Python,

SAS, R, Scala, visualization libraries, SQL database, and
machine learning
 Data Science: Process Flow

How car insurance costs less if

you pay bills on time?

Data scientists found that the people who

pay bill promptly are less prone to the
accidents
 Data Science: Process Flow

Step 1: Data acquisition

Data scientists work with existing data

sets or gather them from various
sources.

Data acquisition

The most important part of the whole process is to have the correct data.
 Data Science: Process Flow

Step 2: Data wrangling

● Choose the right tools from

Python, R, and SQL

● Derive a clean data set

Data acquisition
● Apply pick-and-shovel
algorithms

● Obtain meaningful data

Data wrangling
 Data Science: Process Flow

Step 3: Machine learning

● Validate the model
● Perform necessary statistical analysis
Data acquisition ● Apply machine learning or recursive
analysis
● Run regression testing
● Compare results against other
techniques or sources

Machine learning
Data wrangling
 Challenge of a Data Scientist

The most challenging part of being a data scientist is taking the results and presenting them to the
stakeholders in an easy and consumable manner.
Data Science and Business Strategy
 Data Science and Business Strategy

Business owners used to measure their success based only on the Profit and Loss Statement.

Current era of technology leverages data science for efficient prediction on what will work.
 Data Science and Business Strategy

The process flow of a data-driven decision-making process:

Define business Build a team of data

goals scientists
1 2

4 3
Identify data sources and
Design business
dashboards to track goals enable new sources of data
capture
 Data Scientist: Asset to the Business

Empowers management Identifies and refines

the target audience
to make better decisions

Provides insights on
Identifies areas of
various KPIs and
improvement
parameters

Enables strategic changes Identifies opportunities

for better results

Data Scientist
Companies Using Data Science
Successful Companies Using Data Science

Few successful companies that use data science

Google Search Engine
 Google Search Engine

Google uses data science to provide relevant search recommendations.

The influencing factors include:

● Query volume: unique and verifiable users
● Geographical locations
● Keyword or phrase matches on the web
● Scrubbing for inappropriate content
Facebook Tags
 Facebook Tags

Facebook uses machine learning in every aspect including:

Scrolling the news feed Browsing images or videos

 Facebook Tags

Uses clustering algorithm to:

Find mutual friends Send friend

suggestions
Alibaba
 Alibaba’s Aliloan

Aliloan is an automated online system that provides flexible microloans to entrepreneurial

online vendors.
 Alibaba’s Aliloan

Analyzes trading records Uses predictive models to

and evaluates risk analyze transaction records

Aliloan

Collects data from e- Determines merchants’

commerce platforms creditworthiness
Travel Industry
 Travel Industry

Travel companies use datasets from social media, itineraries, predictive analytics, and location tracking
to arrive at the 360-degree view.

The sensors from different modes of transport provide real-time data on various parameters to predict
and prevent problems.
 Travel Industry

Integrates historical data to Offers deals based on the user’s

ensure maximum yield preferences or recommended
local attractions

Predictive algorithms help drivers predict fuel needs, ETAs, and delays.
Retail
 Retail

RFM analysis is a marketing technique that leverages data to determine the target customer.

Recency Frequency Monetary

Retailers use data science to segment customers into RFM groups and target marketing and
promotions.
E-Commerce
 E-Commerce

Amazon is an e-commerce giant that leverages data science to the fullest extent.

Amazon prefers an everything under one roof model.

 E-Commerce

E-commerce companies use data science to upsell through their websites.

Amazon’s People who viewed that product, also liked this functionality uses
sophisticated mining techniques and boosts business.
Crime Agencies
Crime Agencies

Analytics keeps crime in check by:

● Using identified patterns to derive

prediction techniques

● Analyzing previous data to prevent future

burglaries
 Crime Agencies

● Data mining can help identify pattern in from

domestic violence to terrorism.

● Advanced analytics helps prevent crime by using

information from social media.
 Crime Agencies

Crime prevention agencies use data science in

deciding:

● Where to deploy police manpower?

● Who to search at a border crossing?

● Which intelligence to consider in

counter-terrorism activities?
Analytical Platforms across Industries
 Analytical Platforms across Industries

Data storage Tools

Algorithms Architectures
platforms
 Analytical Platforms across Industries

Machine
Architectures Data storage Tools
learning
platforms
algorithms

Forecasting

Regression

Bayesian network

Vector autoregression
 Analytical Platforms across Industries

Machine learning Deep learning Data storage Tools

algorithms architectures platforms

Deep Belief Network (DBN)

Convolutional Neural Network (CNN)

Recurrent Neural Network (RNN)

 Analytical Platforms across Industries

Machine learning Deep learning Cloud storage Tools

algorithms architectures platforms

Amazon AWS

Microsoft Azure

Lambda
 Analytical Platforms across Industries

Machine learning Deep learning Cloud storage

architectures platforms Tools
algorithms

Analytics tools
● Spark
● Python
● R
Reporting tools ● Apache Pig
● Tableau
● Splunk
● Power BI
● Kibana
 Key Takeaways

Data science is the study of data, which involves gathering,

storing, analyzing, and plotting data, to effectively extract
useful information.

Data science is an umbrella that contains data analytics,

data mining, and machine learning.

Data science is used by many successful companies such as

Google, Facebook, and Alibaba.

Analytical platforms across industries include algorithms,

architecture, data storage platforms, and tools.