Introduction to

Intelligence Systems
By: SEK SOCHEAT
Advisor to DCS and Lecturer
Intelligence System Mobile: 017 879 967
Development
Email: [email protected]
2024 – 2025
Y4E1 – DCS – NU
Table of Contents
1. Overview of Intelligent Systems
• What are Intelligent Systems?
• Types of Intelligent Systems
• Applications of Intelligent Systems
• Key Components of Intelligent Systems
2. Software Development Tools and Frameworks for IS
• Programming Languages for Intelligent Systems
• AI and Machine Learning Frameworks
• Data Processing Tools
• Integrated Development Environments (IDEs)
• Version Control and Collaboration Tools
• Cloud-based AI Development Platforms
• Other Tools for Specific AI Applications
3. Practical Activity
Overview of Intelligent Systems
Types, Applications, and Key Components

What are Intelligent Systems?

• Intelligent systems are computational systems that use
techniques from artificial intelligence (AI) to perceive
their environment, learn from data, and act or make
decisions autonomously.

• They can be designed to solve complex problems by

simulating human-like intelligence, such as reasoning,
decision-making, pattern recognition, and adapting to
changing conditions.
Types of Intelligent Systems
Rule-based Systems: These systems rely on Neural Networks and Deep Learning: Inspired by
predefined rules (if-then logic) to make decisions. the structure of the human brain, these systems can
They were some of the earliest forms of intelligent process vast amounts of data and solve complex
systems. problems like image recognition, natural language
processing, and game playing.

Expert Systems: Designed to replicate the decision- Cognitive Systems: Mimic human thought
making ability of a human expert in a specific processes to understand context, reason, and suggest
domain, using knowledge bases and inference solutions (e.g., IBM Watson).
engines.

Machine Learning Systems: These systems use Robotic Systems: Combine AI with physical
data to learn and improve from experience over components to perform tasks in the real world, like
time. Examples include recommendation systems, autonomous drones or self-driving cars.
speech recognition, and image classification.
Applications of Intelligent Systems
• Healthcare: Diagnosis systems, predictive models for patient
outcomes, and robotic surgery.

• Finance: Fraud detection, algorithmic trading, and credit risk

evaluation.

• Manufacturing: Predictive maintenance, automated quality

control, and robotics for assembly lines.

• Transportation: Self-driving vehicles, traffic management

systems, and route optimization.

• Customer Service: Chatbots and virtual assistants for

answering queries and resolving issues.

• Smart Homes/IoT: Home automation systems (smart

thermostats, lighting, etc.) and predictive systems for energy
management.
Key Components of Intelligent Systems
1. Perception: Systems must first perceive their environment using sensors, cameras, or other
input devices (e.g., visual sensors for a self-driving car).

2. Data Processing: Raw data collected by sensors needs to be cleaned, structured, and
processed before it can be used (e.g., feature extraction, normalization in machine learning).

3. Learning/Inference Engine: Core AI algorithms enable the system to learn from data and
make decisions (e.g., neural networks in deep learning, decision trees in machine learning).

4. Knowledge Base: For systems like expert systems, the knowledge base contains domain-
specific information and rules for decision-making.

5. Actuation: Systems act upon their environment (e.g., a robot moving its arms) based on the
decisions or predictions they make.

6. Feedback Loop: Intelligent systems often improve over time by incorporating feedback from
the environment or user interactions (reinforcement learning, adaptive systems).
Software Development Tools and
Frameworks for Intelligence Systems
Programming Languages for Intelligent Systems
Python Programming:

• Python is the most widely used language for AI and

machine learning due to its simplicity and extensive
ecosystem of libraries.

Advantages: Easy to learn, strong community support, and

rich libraries for scientific computing (e.g., NumPy, SciPy,
Pandas).

Python's role in AI development: Its flexibility allows it

to be used in everything from data preprocessing to model
training and deployment.
AI and Machine Learning Frameworks
1. TensorFlow

• Developed by Google, TensorFlow is one of the most

popular deep learning frameworks.

• Provides tools for building and training machine learning

models, especially for neural networks.

• TensorFlow supports both low-level operations (for

research purposes) and high-level APIs like Keras for
quick prototyping.

• Use cases: Image recognition, NLP, time series analysis.

AI and Machine Learning Frameworks
2. PyTorch

• Developed by Facebook, PyTorch is another popular

framework known for its flexibility and ease of use,
particularly favored in research.

• Provides dynamic computational graphs, which makes it

easier to experiment with model architectures.

• PyTorch is well-suited for rapid development, and is

increasingly used in production environments.

• Use cases: Deep learning research, NLP, reinforcement

learning.
AI and Machine Learning Frameworks
3. Keras
• An open-source library that provides a
high-level API to build and train deep
learning models with minimal coding.
• Keras can run on top of TensorFlow,
simplifying complex machine learning
workflows.
Data Processing Tools
1. NumPy

• A fundamental package for numerical

computing in Python, providing support
for arrays, matrices, and high-level
mathematical functions.
• Essential for handling large datasets and
performing matrix operations, which are
common in machine learning.
Data Processing Tools
2. Pandas
• A powerful data manipulation
library used to handle structured
data in tables (e.g., data frames).
• Pandas makes it easy to
preprocess data, clean datasets,
and perform data analysis.
Data Processing Tools
3. Matplotlib/Seaborn

• Visualization libraries that

help create graphs, charts,
and plots to better
understand datasets and
model performance.
Integrated Development Environments (IDEs)
Jupyter Notebook
• Jupyter Notebook is an interactive web-based
development environment that supports
Python.
• Popular in data science and AI development
for its ability to combine code, text, and
visualizations in a single document.
• Useful for experimenting with machine
learning models, prototyping, and sharing
research.
Integrated Development Environments (IDEs)
VS Code

• A lightweight, flexible IDE with

strong support for Python
development and machine learning
extensions (e.g., TensorFlow and
PyTorch plugins).
Version Control and Collaboration Tools
Git and GitHub

• Git is a version control system that helps track code

changes, enabling collaboration in teams and across
projects.

• GitHub (or GitLab/Bitbucket) is an online platform

for hosting and sharing code, with features like pull
requests, issue tracking, and project management.

• Essential for collaboration in AI development, where

multiple team members may work on the same model
or datasets.
Cloud-based AI Development Platforms

• A cloud-based platform that allows

you to run Jupyter Notebooks in the
cloud with free access to GPUs.
• Ideal for students and small projects
that require powerful computation
without needing dedicated hardware.
Cloud-based AI Development Platforms

AWS SageMaker

Amazon’s machine learning

service that provides an
environment to build, train,
and deploy machine
learning models at scale.
Cloud-based AI Development Platforms
Microsoft Azure Machine Learning
Studio
Azure's machine learning
platform, providing pre-
built models, drag-and-drop
interfaces, and support for
Python and R.
Cloud-based AI Development Platforms
OpenCV
• A powerful library used for computer vision applications
such as object detection, image recognition, and facial
recognition.

• Provides tools to capture, process, and analyze visual

data.

NLTK (Native Language Tool Kit)

A toolkit for building natural language processing

systems, including tools for text processing,
tokenization, and sentiment analysis.
Practical Activity
Installation and Setup
• Install Python, TensorFlow, and PyTorch in the
development environment
(e.g., Jupyter Notebook, VS Code).

• Demonstrate how to set up a project

structure and manage dependencies using
tools like pip or Anaconda.

Hands-On Example
• Implement a simple machine learning model (e.g., linear
regression or image classification) using TensorFlow or PyTorch.

• Visualize the data using Matplotlib/Seaborn, and display training

accuracy and loss using Jupyter Notebook.
Thank
you