PowerPoint Presentation
What is analytics
Analytics is the use of:
data,
information technology,
statistical analysis,
quantitative methods, and
mathematical or computer-based models
to help managers gain improved insights about their business operations and make better, fact-based decisions.
What if you could…
. . . predict the buying behavior and decision criteria of your potential customers weeks before your competition?
. . . gain first-mover advantage by introducing new products and services to micro-segments that haven’t been identified by competitors?
. . . evaluate the impact of your marketing campaigns hourly and make adjustments in real-time?
. . . improve customer experience scores that grow products per customer, reduce attrition, and leverage the power of customer recommendations for new business?
. . . predict likely failures of critical equipment and processes?
Why Business Analytics?
There is a strong relationship of BA with:
– profitability of businesses
– revenue of businesses
– shareholder return
BA enhances understanding of data
BA is vital for businesses to remain competitive
BA enables creation of informative reports
Why Business Analytics?
Actional insights for Business Value Creation
Why Business Analytics?
https://sloanreview.mit.edu/projects/analytics-mandate/
Management of customer relationships
Financial and marketing activities
Supply chain management
Human resource planning
Pricing decisions
Sport team game strategies
Applications
From Business Analytics to Business Value
Business
Question Analytic Application Business
Value
What market segments do my customers fall into, and their characteristics? Customer segmentation Personalize customer relationship for higher satisfaction and retention.
Which customer are most likely to respond to my promotion? Propensity to buy Increase campaign profitability by focusing on the most likely to buy
How can I tell which transactions are likely to be fraudulent? Fraud detection Quickly determine fraud and take immediate actions to minimize cost.
Overview of Business Analytics
Lecture Topics – Refer to Course Map on Moodle
Week Topic
1 Introduction to Data and Machine Learning
2 Statistics
3 Data Exploration
4 Clustering
5 Linear Regression
6 Mid-term Break
7 Logistic Regression
8 Decision Trees
9 Data Robot I
10 Data Robot II
Introduction to Data
DATA
– collected facts and figures
DATABASE
– collection of computer files containing data
INFORMATION
– comes from a specific context
KNOWLEDGE
– comes from analyzing data
Data for Business Analytics
From Data to Wisdom
Data
Information
Knowledge
Wisdom
Given insights become
Given meaning become
Given context become
Data file contains code 802981
Code 802981 is a customer code
Customer with code 802981 is valued but has a high likelihood of churn
We need to be prepared to take action to retain customer
Examples of using DATA in business:
Annual reports
Accounting audits
Economic trends
Marketing research
Operations management performance
Human resource measurements
Customer interactions
Internet of Things (IoT)
Sources of data
Metrics are used to quantify performance.
Measures are numerical values of metrics.
Discrete metrics involve counting
– on time or not on time
– number or proportion of on time deliveries
Continuous metrics are measured on a continuum
– delivery time
– package weight
– purchase price
Data for Business Analytics
Four Types Data Based on Measurement Scale:
Categorical (nominal) data
Ordinal data
Interval data
Ratio data
Data for Business Analytics
Categorical (nominal) Data
Data placed in categories according to a specified characteristic
Categories bear no quantitative relationship to one another
Examples:
– customer’s location (America, Europe, Asia)
– employee classification (manager, supervisor, associate)
Data for Business Analytics
1-
Ordinal Data
Data that is ranked or ordered according to some relationship with one another
No fixed units of measurement
Examples:
– college football rankings
– survey responses
(poor, average, good, very good, excellent)
Data for Business Analytics
1-
Interval Data
Ordinal data but with constant differences between observations
No true zero point
Ratios are not meaningful
Examples:
– date
Data for Business Analytics
1-
Ratio Data
Continuous values and have a natural zero point
Ratios are meaningful
Examples:
– monthly sales
Data for Business Analytics
1-
Classifying Data Elements in a Purchasing Database
Data for Business Analytics
Figure 1.2
1-
Classifying Data Elements in a Purchasing Database
Data for Business Analytics
Categorical
Categorical
Categorical
Ratio
Categorical
Ratio
Ratio
Ratio
Interval
Interval
Figure 1.2
1-
Model:
An abstraction or representation of a real system, idea, or object
Captures the most important features – usually the relationships between factors
Can be a written or verbal description, a visual display, a mathematical formula, or a spreadsheet representation
Decision Models
Decision Models
Figure 1.3
A decision model is a model used to understand, analyze, or facilitate decision making.
Types of model input
– data
– uncontrollable variables
– decision variables (controllable) –how the measure changes when we modify these variables
Types of model output
– performance measures
– behavioral measures
Decision Models
Decision Models
observable
Observable, changeable
Non-observable, have to predict
Dimensions of Data Quality
FIGURE 3.6, Vidgen et al. 2019
Data Quality | Accuracy
To what extent does the data represent the ‘real-world’ aspect described?
Example
Manual data entry is prone to human error – this can range from a simple typo to the misplacement of a decimal place.
Automatic data feeds are more reliable, however still need to be checked in early stages of implementation. There could be a glitch in the system causing data type conversion errors.
Potential Measure
Cross-checking a sample of the data and determine the percentage which are deemed accurate based on a set of rules.
Data Quality | Completeness
How comprehensive is the data?
Does it meet the set expectations?
Example
The necessary information is provided, whereas optional fields can be missing. Such as the middle name of a customer, alternative contact details etc.
Potential Measure
Determine the percentage of required fields which are missing compared to the total number of fields.
Data Quality | Timeliness
To what extent is the data available when it is expected and needed?
Example
Real-time data may be required in critical scenarios. Such as trading data for investors or feeds from IoT sensors on critical equipment or high monitoring equipment.
Less time critical data can be available with a lag and still considered timely. Such as sales data, website traffic, company financials etc.
Potential Measure
The time interval between when the data are available compared to when it is expected/needed.
Data Quality | Validity
To what degree is the data sensible?
Example
Valid age for underage children included in voting demographic surveys.
Potential Measure
Determining the percentage of data items which are deemed to be valid or invalid based on a set of defined rules.
Data Quality | Integrity
The validity of data across the relationships. Can all the data be traced and appropriately joined to the other data points?
Example
An online sales database that connects a customer to an order, however if the order cannot be matched to a customer then we have what is known as an orphaned record.
Or we might have the opposite with a single order corresponding to multiple customers, this also calls into question the integrity of the data.
Potential Measure
The total count or percentage of orphaned records present in a dataset.
Data Quality | Consistency
Are the same data represented identically across the systems?
Example
There should be a single source of truth, however in many organisations data is duplicated across various system. E.g., HR and Finance teams may each have a set of employee details.
Potential Measure
The percentage of the same data that are correctly matched across different systems.
Introduction to Machine Learning
1-
What is Machine Learning?
Machine learning is a field of computer science that aims to teach computers how to learn and act without being explicitly programmed.
More specifically, machine learning is an approach to data analysis that involves building and adapting models, which allow programs to “learn” through experience.
Machine learning involves the construction of algorithms that adapt their models to improve their ability to make predictions.
– DeepAI
Types of Machine Learning
Image sourced from https://www.mathworks.com/help/stats/machine-learning-in-matlab.html
e.g., Logistic Regression
Decision Tree
e.g., Linear Regression
Most department stores clear seasonal inventory by reducing prices.
The question is:
When to reduce the price and by how much? evaluating trade-offs
Descriptive analytics: examine historical data for similar products (prices, units sold, advertising, …)
Predictive analytics: predict sales based on price
Prescriptive analytics: find the best sets of pricing and advertising to maximize sales revenue
Example: Retail Markdown Decisions
Descriptive Analytics refers to knowing what is happening in the organization and understanding some underlying trends and causes of such occurrences.
Clustering (e.g., customer segmentation)
Visualization (e.g., bar chart; line chart; etc.)
Business Analytics Techniques
Predictive Analytics refers to determining what is likely to happen in the future.
Linear regression (e.g., house pricing in 5 years)
Logistic regression (e.g., customer churn rate)
Decision tree (e.g., generate decision rules)
Business Analytics Techniques (Cont.)
Testing Prediction Accuracy
Machine learning algorithms aim to continuously improve their prediction accuracy as more and more data is fed through the model.
To test the accuracy of a model, data is generally split between train and test data.
The model is fitted to the training data and then the model is run on the test data to determine the accuracy of the model at predicting the target variable.
Testing Prediction Accuracy
.MsftOfcThm_Accent1_Fill_v2 {
fill:#0B3BD3;
}
/docProps/thumbnail.jpeg