Session-01.Fundamentals-Of-DA-PSO-06

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 81

Data Analytics For Business

Session 1:
Fundamentals of
Data Analytics
2023.07
Quang-Khai Tran, Ph.D
How do they know what I want to buy?
How do they know what I want to watch?
How do they know what I want to eat?
How…?

They know more about me than I do!!!

2
Contents
I. Course Introduction
II. Introduction to Data Analytics
III. Some Successful Cases
IV. Five Most Important and Basic Charts
(with some examples from the real world)
V. Discussion

3
Part I
1. Overview
Course Introduction 2. Lecturer
3. Students

4
Overview
5
How to harvest values from this data farm?

(Source: Internet)

6
I. Course Introduction
1. Overview

Turning Data into Actionable Insights

Source: https://towardsdatascience.com/turning-data-into-actionable-insights-c246969fa4c
7
I. Course Introduction
1. Overview

Exploiting values in data needs an analytical


mindset and some technical skills

Source:
https://vitalflux.com/what-are-actionable-insights-examples-concepts/ Source: https://www.softwaretestinghelp.com/data-analytics-companies/

8
I. Course Introduction
1. Overview

This course is aiming to guide


you become a data-driven
professional and enhance your
data literacy

Source: https://learn.g2.com/advanced-analytics

9
I. Course Introduction
1. Overview

You will be prepared with


most needed skills in Excel
and PowerBI to perform data
analytics by yourself.

Source: https://www.forbes.com/advisor/business/software/best-data-analytics-tools/

10
I. Course Introduction
1. Overview

Schedule
Session 1 07.15 Fundamentals of Data Analytics
Common Excel skills for data exploration
Session 2 07.22
Datasets Delivery (or self-suggestion)
Session 3 07.29 Exploratory Data Analysis (EDA) with Excel
Session 4 08.02/08.09 (ONLINE) Customer Segmentation
Session 5 08.12 (must attend) Mid-term Presentation (50%)
Session 6 08.19 (ONLINE) Data Visualization with PowerBI
Session 7 08.26 Problem Solving with PowerBI (dashboard)
Session 8 09.09 The techniques of persuasive data storytelling
- Expert sharing: Applications of DA in the reality
Session 9 09.16
- Extra presentations: students who want a bonus
Session 10 09.23 (must attend) Final Exam (50%): Group Presentation

11
I. Course Introduction
1. Overview

Some notices:

➤ Attendance: minimum 80% (6 sessions + 2 presentations)


➤ Before each class: read recommended readings and watch videos
➤ Break-time in each class:

Class time 55 min

Break time 05 min

Class time 55 min

Break time 10 min

Class time 55 min

12
I. Course Introduction
2. The lecturer

➤ Fullname: Trần Quang Khải


➤ Nickname: Ricardo
➤ Background:
● Big Data Science (Ph.D, 2019)
● Computer Science (Master)
● Software Engineering (BS)
● Others:
- Climate Change
- Sustainable Socio-Eco Development
- Food Traceability with Blockchain

Link: https://scholar.google.com/citations?hl=en&user=H61NLGMAAAAJ

13
I. Course Introduction
Tutor

➤ Mr. Thach Nguyen (Nguyễn Ngọc Thạch)


➤ Lead Data Analytics, Home Credit Vietnam
➤ Email: ngocthach.nguyen1102@gmail.com
➤ Phone: 0399 197 930
14
I. Course Introduction
3. Student Self-Intro

➤ Introducing yourself and sharing about your work with data

15
Part II
Introduction to Data 1. Overview (and basic concepts):
Data Analytics/Data Analysis/Data Science
Analytics 2. Four levels of Data Analytics:
(descriptive, diagnostic, predictive,
prescriptive)
3. Career in DA/DS

16
II. Introduction to DA
1. Overview

Data Science: the intersection of statistical


methodology, computational science, and a wide
range of application domains
(Harvard Kenneth C. Griffin Graduate School of Arts and Sciences)

(Source: https://towardsdatascience.com/q-a-common-questions-in-data-science-7cd7f9d82a8d)

17
II. Introduction to DA
1. Overview

What is Data Analytics?

What is Data Analysis?

What is Data Science?

Data Literacy?
Data-driven Business?
Business Analytics?
Data Mining?
Advanced Analytics?
Augmented Analytics?
Data Analytics Governance?
Source: https://www.naukri.com/learning/what-is-data-analyst-dg439#description
18
II. Introduction to DA
1. Overview

Data analytics is the collection, transformation, and organization of data in


order to draw conclusions, make predictions, and drive informed decision
What is Data making.
(Google Data Analytics, www.coursera.com:
https://www.coursera.org/professional-certificates/google-data-analytics)
Analytics?
Data analytics is the science of analyzing raw data to make conclusions
about that information
(Source: https://www.investopedia.com/terms/d/data-analytics.asp)

Data analytics is the process of analyzing raw data in order to draw out
meaningful, actionable insights
(Source: https://careerfoundry.com/en/blog/data-analytics/what-is-data-analytics/)

Data analytics is the process of examining data sets in order to find trends
and draw conclusions about the information they contain.
(Source: https://www.techtarget.com/searchdatamanagement/definition/data-analytics)

19
II. Introduction to DA
1. Overview

Data analytics is the collection, transformation, and organization of data in


order to draw conclusions, make predictions, and drive informed decision
making.
(Google Data Analytics, www.coursera.com:
https://www.coursera.org/professional-certificates/google-data-analytics)

Data analytics is the science of analyzing raw data to make conclusions


about that information
(Source: https://www.investopedia.com/terms/d/data-analytics.asp)

Data analytics is the process of analyzing raw data in order to draw out
My opinion ⇒ meaningful, actionable insights
(Source: https://careerfoundry.com/en/blog/data-analytics/what-is-data-analytics/)

Data analytics is the process of examining data sets in order to find trends
and draw conclusions about the information they contain.
(Source: https://www.techtarget.com/searchdatamanagement/definition/data-analytics)

20
II. Introduction to DA
1. Overview

In dictionaries: In dictionaries:
"analysis is the division of a "analytics is the science of
whole into small components" logical analysis"

Source: https://www.jigsawacademy.com/blogs/business-analytics/analysis-vs-analytics

21
II. Introduction to DA
1. Overview

2. Data Analytics is more comprehensive, refers to the complete management of


data, and may include one to several to many "Data Analyses"

1. Data analysis is the process consisting of cleaning,


transforming, modeling, and questioning data to find useful
information (in details)

● Each act of data analysis may have these activities,


from collection to storage to visualization

● Data analysis is usually limited to a single, already


prepared dataset, dividing it into small components

Source-1: https://www.bmc.com/blogs/data-analytics-vs-data-analysis/
Source-2: https://www.jigsawacademy.com/blogs/business-analytics/analysis-vs-analytics
22
II. Introduction to DA
1. Overview

Source: Google Data Analytics Course (https://www.coursera.org)


23
Typical Steps in a Data Analytics Project

Source: https://medium.com/codex/life-cycle-of-a-data-analytics-project-954d0e6926fe

24
Source: https://uxknowledgebase.com/data-literacy-quantitative-research-part-2-de07607f1127
25
"Gartner defines data literacy as the ability to
read, write and communicate data in context"
(Source: Data and Analytics: Everything You Need to Know | Gartner)

Source: https://uxknowledgebase.com/data-literacy-quantitative-research-part-2-de07607f1127 26
"... data literacy - the ability to analyze,
interpret, and even question data - is an
increasingly valuable skill."
— Janice Hammond, Professor at Harvard Business School —

27
II. Introduction to DA
1. Overview

So, why data analytics is important for business?

"... For business professionals, knowing


how to interpret and communicate
data is an indispensable skill that can
inform sound decision-making."

Source: Examples of Business Analytics in Action | HBS Online

28
II. Introduction to DA
1. Overview

Some examples

Make better decisions about where to allocate resources

Moving to data-driven How to price products or services appropriately


business model has
Discover trends and understand customers
been an emerging wave
in recent years
Optimize internal operations to reduce costs and waste

Especially when wars and pandemics happen

Source: Examples of Business Analytics in Action | HBS Online

29
Data has been used
widely in Marketing as a
key to success

Source: Incorporating Data And Analytics Into Your Marketing Plan


30
Tweet: https://twitter.com/kalebw/status/1633470979820093441
Source Mind-map: https://whimsical.com/ai-for-marketing-kaleb-willems-YBKnEtRUpsoBH8MBerctJU 31
We are living in a data culture where business decisions are
based on facts, not opinions === Microsoft ===
Ref: https://learn.microsoft.com/en-us/power-bi/consumer/end-user-consumer
Types of Data Analytics
33
II. Introduction to DA
2. Types of DA

Four Types Of Data Analytics

34
II. Introduction to DA
2. Types of DA

Four Types (Levels) Of Data Analytics - Is a simple, surface-level type of analysis


based on historical data to examine,
understand, and describe what happened
Descriptive Analytics
- Uses BI and visualization tools to summarize
(Phân tích mô tả)
the data, or discover trends and patterns
- E.g.: Have the number of customers gone
up? Are sales better this month than last?
- Tries to uncover causal relationships
- May involve seeking to identify anomalies
Diagnostic Analytics
within the data
(Phân tích chẩn đoán)
- E.g.: Did the latest marketing campaign
impact sales?
- Is based-on historical data, past trends, and
Predictive Analytics
assumptions to predict future outcomes
(Phân tích dự đoán)
- Uses machine learning models
- Tries to find out and suggest what individuals
or organizations should do to obtain future
Prescriptive Analytics
targets/goals
(Phân tích đề nghị)
- Uses predictive analytics to show results of
different scenarios

Others: cognitive analytics, behavioral analytics, risk analytics...


35
II. Introduction to DA
2. Types of DA

03 types of knowledge from data (not only insights):

➤ Hindsight: ability to learn from the past.


➤ Insight: ability to understand and respond to what is happening at the present
➤ Foresight: ability to predict/forecast and prepare for the future

Source:https://www.linkedin.com/pulse/hindsight-insight-foresight-key-ingredients-effective-woods
36
To succeed, we need all three!

Source: https://www.linkedin.com/pulse/hindsight-insight-foresight-patrick-mcdonald
37
II. Introduction to DA
2. Types of DA

In summary

Source: https://www.franklin.edu/blog/accounting-mvp/accounting-data-analytics
38
Career in Data Analytics/Science
39
II. Introduction to DA
3. Career in DA

Data Analyst Data Engineer Data Scientist


Examine data and help others Is involved in the data Have a more technical and broad-ranging
understand the story that the preparation process (collecting role, usually try to find patterns in the data
data is telling, make reports, and validating the information, and answer questions about the future
suggestions and support processing raw data, providing (uncover patterns, develop algorithms,
decision making… required data…) and make predictions)
Analytics Engineer (or Fullstack DA) Machine Learning Engineer

40
II. Introduction to DA
3. Career in DA

Source: https://towardsdatascience.com/the-data-science-process-a19eb7ebc41b
41
II. Introduction to DA
3. Career in DA

(Source: https://medium.com/indeed-engineering/where-do-data-scientists-come-from-fc526023ace)
42
II. Introduction to DA
3. Career in DA

(Source: https://medium.com/indeed-engineering/where-do-data-scientists-come-from-fc526023ace)
43
3. Career in DA

Advantages and Disadvantages


of people with business
background?

Source: 4 Reasons Why Economists Make Great Data


Scientists (And Why No One Tells Them) | Medium
44
II. Introduction to DA
3. Career in DA

Your advantages (with an MBA):

➤ Know about linear regression and logistic regression


(something that will lead to neural networks, deep learning)
➤ You understand the business and needed metrics better
➤ You know how to explain causality and have alternative perspective on data
➤ You communicate better in an economic way with presentational skills
➤ Python (the leading programming language in DS) is not too hard for everyone

Source: How to Get Into Data Science With an Economics Degree?


45
II. Introduction to DA
3. Career in DA

Your advantages (with an MBA):

Know the Right Communicate


Tools for the Job Persuasively

Find Connections Predict What’s


that Matter Next

Turn Ideas Develop Unique


into Execution Strategic Insights

Source: The Benefits of Learning Data Analytics From a Business School |


CMU

46
II. Introduction to DA
3. Career in DA

Your disadvantages?

Source: https://wol.iza.org/articles/big-data-in-economics/long

47
II. Introduction to DA
3. Career in DA

The most important skills of data scientists | Jose Miguel Cansado | TEDxIEMadrid
Source: https://www.youtube.com/watch?v=qrhRfPY4F4w
48
II. Introduction to DA
3. Career in DA

Do you want to change your career now?

Link: https://www.youtube.com/watch?v=3gssAG0agO8
49
Our challenges and chances?

50
Source: Total data volume worldwide 2010-2025 | Statista
51
Source: https://www.linkedin.com/pulse/know-top-10-data-science-trends-2022-learnbay?trk=organization-update-content_share-article 52
The Top 5 Data Science And Analytics Trends In 2023 (forbes.com)

Data Governance Data


and Regulation Democratization
Non-technical people can use DA
tools more easily and effectively

NLP: can understand and communicate with


us in human languages (e.g: Chat GPT)

Real-Time Artificial Computer vision: can understand and


Data Intelligence process visual information (just like our eyes)

Cloud and Generative AI: can create text, images,


sounds and video from scratch
Data-as-a-Service

Source: forbes.com

53
Part III
Some Successful
Cases

54
III. Some Successful Cases

Amazon’s recommendation system


● Understands each user’s preferences
⇒ Easily find the thing that we want
among 12 million products
● Contributes 35% of the annual sales

Source: 5 Interesting Case Studies of Companies Using Data 55


III. Some Successful Cases

What else in Amazon?


● Supply Chain Optimization: fulfill the orders
quickly, locates the closest warehouse to a
customer/vendor to reduce the shipping costs

● Price Optimization: prices change frequently


due to activity on the website, competitors’
pricing, product availability, item preferences,
order history, expected profit margin, …

● AI for fraud detection: screen purchases and


return requests for signs of fraud

● To change and modify physical stores


Source: How Amazon uses Big Data?
● Analyzing sentiments in smartphone reviews

56
III. Some Successful Cases

In Covid-19 lockdowns
● Operates only via apps and drive-thru
for takeout or delivery
● Creates dashboards for monitoring
real-time operation
● Starbucks Now, Starbucks Delivery,
Deep Brew: recommend menus based
on weather, seasons, and time
● Supports voice ordering

Source: 5 Interesting Case Studies of Companies Using Data 57


III. Some Successful Cases

● Makes A/B testing to give users the


best experience
● Recommendation system: the easiest
way to find the right accommodations
● Applies NLP to detect the true feeling
of customers
● Asia users often leave the site after
visiting the ‘Neighborhood’ page
⇒ Replaces with ‘Top destinations’,
resulting in a 10% booking increase
● Uses ML to find important factors that
affect customers' decision

Source: 5 Interesting Case Studies of Companies Using Data 58


III. Some Successful Cases

Recommendation system
● Is far more detailed than we expect
⇒ all activities when we watch movies
● 80% of the users follow Netflix
recommendations
● 74% Customer Retention Rate

Source: 5 Interesting Case Studies of Companies Using Data 59


III. Some Successful Cases

"one of the first global corporations, outside the IT industry, to join the big data conversation"

500 brands of
soft drinks

~ $84 b (> Pepsi,


200 countries Budweiser, Subway,
& KFC combined)

1.9 billion servings


everyday

"relies on solid data-driven strategies for business intelligence and to guide strategic decisions"

Source: How Coca Cola Company leverages data? 60


Mining social media data Uses AI to conduct inventory
⇒ consumption patterns checks & manage supply chain
⇒ feedback of customers
⇒ determine if retailers and
Uses AI to identify when an image vending machines have stocks
of the company’s product is posted of other brands

⇒ enhance targeted advertisement


and monitor the success of products
Source: OpScoop Chapter III Issue XXXXI: Coca Cola’s Taste on Data Analytics 61
III. Some Successful Cases

Coca-Cola's customer loyalty program:

1. Big data needs to start small


2. Listen carefully to consumers' opinions - by
phone, email or social networks
3. Collect, store, and use data in a privacy safe way
4. Data help create more relevant content for
different audiences (personalization: creating
advertising content that speaks differently to
different audiences)

Source:
5 Big Data Analytics Examples of Leading Brands - Mentionlytics

62
Part IV 1. Line (and area) chart
2. Bar (and column) chart
Five Basic Charts 3. Pie (and donut) chart
4. Scatter chart
5. Histogram chart

63
IV. Five Basic Charts

Data visualization: representation of data in graphical or pictorial format

Source: Matplotlib (https://matplotlib.org/)


64
IV. Five Basic Charts

"A Picture Is Worth A Thousand Words"


(since 1910s, see Wikipedia)

Source: https://www.youtube.com/watch?v=GpP0EbSMRpg&ab_channel=LeilaGharani

65
IV. Five Basic Charts

Example

Source: https://boostlabs.com/blog/10-types-of-data-visualization-tools/
66
IV. Five Basic Charts

Source: https://www.ataccama.com/platform/data-stories
67
IV. Five Basic Charts

Bar/Column Histogram Scatter Line/Area Pie

68
IV. Five Basic Charts
1. Line Chart & Area Chart

When to use line chart or area chart?

➤ Line charts show trends over time


➤ Area chart is similar to line chart, but the area under each line is colored (or shaded)
➤ Both are usually used when x-labels are date/time or texts

69
IV. Five Basic Charts
1. Line Chart & Area Chart

There are several types of area charts:

➤ Normal area chart


➤ Stacked area chart: elements are added at each label point
➤ 100% stacked area chart: elements are added up to 100%

Source: https://www.exceltip.com/tips/the-area-chart-in-excel.html

70
IV. Five Basic Charts
2. Bar Chart & Column Chart

When to use column chart or bar chart?

➤ To compare the values of different categories


➤ Column chart: uses vertical columns
➤ Bar chart: is a horizontal version of column chart
⇒ Usually used when the label is a long text
➤ Note: the order of categories is not important

71
IV. Five Basic Charts
2. Bar Chart & Column Chart

Group bar chart/column chart: example

72
IV. Five Basic Charts
2. Bar Chart & Column Chart

Stacked bar chart/column chart: example

73
IV. Five Basic Charts
3. Pie Chart & Donut Chart

➤ Pie chart shows proportion of categories (slices) to a total (the pie)


➤ Donut chart is just a version of pie chart

74
IV. Five Basic Charts
4. Scatter Chart

➤ Shows relationship between 02 variables X and Y


➤ Usually used to investigate the correlation or trend
➤ Or to discover groups

75
IV. Five Basic Charts
4. Scatter Chart

➤ Shows relationship between 02 variables X and Y


➤ Usually used to investigate the correlation or trend
➤ Or to discover groups

76
IV. Five Basic Charts
5. Histogram Chart

➤ Shows the distribution of (numeric) data


➤ Notes:
● Only ONE category
● The data will be divided into many
bins (sub-categories)
➤ Can show frequency or ratio (probability)

77
Thank you!

And please make sure to


install Microsoft Excel in your
laptop before the next class

78
Topics for Extra Presentations

1 PowerPivot

Predictive Analytics
2 (Linear Regression and Logistic
Regression using Solver in Excel)

3 Some AI-based visuals in PowerBI

79
Discussion
How did you analyze data?
Which charts did you usually use?
What is the most "headache" issue?

(Source: Internet)

80
81

You might also like