E-Systems Applications

CPIS 483
Spring 2023 – Third Term
Dr. Ghada Amoudi
Outline
• Introduction
• e-System e-Commerce and e-Business
• Trends in e-Systems
• AI and Machine Learning
• Applications of Machine Learning in e-Systems
• Course tools
• Course information
• What’s Next?

2
Introduction
• Introduction to e-business systems CPIS 380
• Covered e-commerce from a business standpoint
• This course presents advanced materials in e-
systems
• From a technical – computer science – standpoint

3
e-System e-Commerce and
e-Business
• e-Systems: An umbrella term for any electronic system.
(Dictionary definition), such as e-Learning, e-Government,
e-Commerce and e-Business
• e-Commerce: digitally enabled commercial transactions
between and among organizations and individuals
• e-Business: Digital enabling of transactions and processes
within a firm, involving information systems under firm’s
control
• Does not include commercial transactions involving an exchange of
value across organizational boundaries
4
Trends in e-Systems
• AI
• Blockchain
• IoT
• Virtual Reality and Augmented Reality
• ….

5
AI and Machine Learning
• Artificial Intelligence: AI
• Machine learning: a branch of AI
• Deep learning

6
Applications of Machine Learning
in e-Systems
• Recommender system
• Dynamic pricing
• Predicting whether a user will make a purchase
• Predicting whether the user will be returning
• Customer churn prediction (Chapter 3)
• Prediction of demand for specific product categories
• Fraud detection
• Setiment analysis
Source: https://addepto.com/blog/best-machine-learning-use-cases-ecommerce/

7
Course tools
• Python
• Jupyter notebook
• Amazon AWS
• SageMaker

8
Python
• Mature, created since 1989 by Guido van Rossum
• Emphasizes on code readability
• Handles small and large datasets
• Support multiple programming paradigms, OOP,
procedural, functional
• Flexible, easy to adapt
• Dynamic typing, dynamic memory allocation
• Large echo system of scientific libraries

9
Python
Advantages
• Ease of programming
• Minimizes the time to develop and maintain code
• Modular and object-oriented
• Large community of users
• A large standard and user-contributed library
Disadvantages
• Interpreted and therefore slower than compiled
languages
• Decentralized with packages
10
Jupyter notebooks
• A new generation of IDEs for interactive languages
such as Python.
• Started in the academia and e-learning
communities
• Used in classrooms and to show results in
presentations or on business dashboards.
• IPython notebook: shows the Python execution
results very clearly and concisely by means of cells.
• Cells can contain content other than code

11
Amazon AWS
• AWS is Amazon’s cloud service.
• Allows companies set up servers and interact with
services in the cloud rather than building their own
data centres.
• AWS has dozens of services:
• computing services such as cloud-based servers (EC2),
• messaging and integration services
• domain-specific machine learning services such as
Amazon Transcribe (for converting voice to text) and
AWS DeepLens (for machine learning from video feeds).

12
SageMaker
• SageMaker is Amazon’s environment for building and
deploying ML applications.
• SageMaker provides the following functionalities:
• Serves as your development environment in the cloud
• Uses a preconfigured ML application on your data
• Uses inbuilt tools to validate the results from ML application
• Hosts your machine learning application
• Automatically sets up an endpoint that takes in new data
and returns predictions
13
SageMaker

• SageMaker development environment uses:

• Jupyter notebook
• Jupyter: based on Python programming
language
• Jupyter is the most popular tools for data
scientists/ML
• Python is the fastest growing programming
language for data scientists.
14
Course information
CLOs, Assessment, Textbook etc.
Teaching strategy: Project-Based
Learning
• A student-centred teaching method in which students learn by actively
engaging in real-world projects.
• Give students the opportunity to develop knowledge and skills through
engaging projects that resembles challenges and problems they may face
in the real world
• Involves a dynamic classroom where students acquire a deeper
knowledge through active exploration of real-world challenges and
problems.
• Students learn about a subject by working for an extended period of time
to investigate and respond to a complex question, challenge, or problem
• It is a style of active learning and inquiry-based learning.

16
Textbook

17
Course Learning Outcomes
1. Apply the fundamental principles of Project Management
to develop a project plan for the group project.
2. Recognize how machine learning applies to businesses to
achieve productivity and automation.
3. Complete the process of creating an AWS account, setting
up an S3 file storage system and SageMaker to use for
building machine learning models.
4. Identify machine learning opportunities and data
requirements.
5. Build a machine learning system to make decisions
6. Explain how the XGBoost algorithm works. 18
Course Learning Outcomes
7. Solve the problem of imbalanced data in the machine learning process
8. Examine datasets resources and how to collect and organize a dataset
or use an existing one.
9. Describe natural language processing (NLP) and how it is applied in a
machine learning scenario
10. How to prepare data for an NLP scenario using SageMaker’s text
analytics engine, BlazingText.
11. Describe recommender systems, their goals, types and some well-
known examples of recommendation applications.
12. Develop a machine learning project involving composing a dataset,
then processing, analyzing and modelling this dataset.
19
The Course Flow
• You will learn by doing…
• Select a topic
• Learning and working/practicing in parallel

20
Tasks, schedule and assessment

Task % Due week

Lab activities (attendance/solving exercises) 10 2-10
Project plan (CPIS 334) 5 3
Phase 1: Idea, introduction, literature & 10 5
presentation.
Midterm 20 7
Phase 2: dataset 10 9
Phase 3: results, presentation submit final notebook 15 10
Comprehensive final exam 30 12-13
Total 100

21
What’s Next?
• Form a group of 3-4 students
• Choose a topic from the following – you have to
sign up your names in one project
Project Max No.of
Groups
Sentiment analysis 2
Fake reviews detection 2
Helpful reviews detection 2
Collaborative recommender system 2
Content-based recommender system 2
Fraud detection 2
Dynamic pricing 2
22