Semester 5 Modules: Specialization Artificial Intelligence

(Intelligence Artificielle – IA -)

Total Contact Private

code Title type Coefficients ECTS
workload hours study

AI.5.1 Multi-agent systems Compulsory 2 2 50 30 20

AI.5.2 Non-Classical Logics Compulsory 2 2 50 30 20
AI.5.3 Applied Deep Learning Compulsory 2 2 50 30 20
AI.5.4 Natural Language Processing Compulsory 2 2 50 30 20
AI.5.5 Robotics and soft computing Compulsory 2 2 50 30 20
ISA.5.1 Big Data Compulsory 2 2 50 30 20
ISA.5.3 Interactive Decision Support Systems optional 2 2 50 30 20
ISA.5.2 Data mining Compulsory 2 2 50 30 20
MAT.5.1 Combinatorial Optimization optional 2 2 50 30 20
MAT.5.2 Category theory and functional programming Compulsory 2 2 50 30 20
AI.5.6 Knowledge Representation & Reasoning Compulsory 1 1 25 15 10
ISA.5.6 Text Mining Compulsory 1 1 25 15 10
ISA.5.5 Recommender systems optional 2 2 50 30 20
ISA.5.4 Business Intelligence optional 2 2 50 30 20
AI.5.7 Deep Reinforcement Learning optional 1 1 25 15 10
ISA.5.7 Distributed databases optional 1 1 25 15 10
ISA.5.8 Advanced BI optional 1 1 25 15 10
AI.5.10 Semantic web & Linked Data optional 1 1 25 15 10
AI.5.8 Affective computing optional 1 1 25 15 10
AI.5.9 Intelligent Transport Systems optional 1 1 25 15 10
DOS.5.8 Introduction to Cloud computing Compulsory 1 1 25 15 10
DOS.5.2 Blockchain Compulsory 1 1 25 15 10
AI.5.11 AIOT optional 1 1 25 15 10
Complementary Module 1 2 2 50 30 20
Complementary Module 2 2 2 50 30 20
Complementary Module 3 2 2 50 30 20
Complementary Module 4 1 1 25 15 10
Complementary Module 5 1 1 25 15 10
Complementary Module 6 1 1 25 15 10

8 Ref ID.230:01
 AI.5.1 Multi-agent systems
Module designation AI.5.1 MultiAgent Systems

Semester(s) in which the module S5

is taught

Person responsible for the module Narjès Bellamine Ben Saoud

(coordinator)

Language French/English

Relation to curriculum Compulsory

Teaching methods lesson, lab works, project, seminar.

Workload (incl. contact hours, Total workload:50h

self-study hours) Contact hours :30h
Private study:20h

Credit points 2 ECTS

Required and recommended pre- SE.3.1 Software Engineering

requisites for joining the module AI.3.1

Module objectives/intended learn- - Master the concepts of agent and multi-agent systems
ing outcomes - Study and apply a design methodology for a multi-agent system.
- Learn the development of a multi-agent system
- Explore complementary research questions
Competencies: C1 and C7

Content Chapter 1: Introduction to MAS

Chapter 2: Intelligent Agents
Chapter 3: Agents architectures
Chapter 4: Methodologies for developing multi-agent systems
Chapter 4: MAS Development environments & case study

Examination forms 35% Continues evaluation + 65% Written exam

Study and examination require- 10/20

ments

Reading list Ferber, J. (1995). Les systèmes multi-agents. InterEditions.

Wooldridge, M. (2002). An Introduction to MultiAgent Systems. Wiley.
Russell, S., & Norvig, P. (2006). Intelligence Artificielle (2ème édition).
Pearson Education France. (Chapter 2).
Russell, S., & Norvig, P. (2020). Artificial Intelligence: A Modern Ap-
proach (4th ed.). Pearson.

AI.5.2 Non-Classical Logics

9 Ref ID.230:01
 Module designation AI.5.2 Non-Classical Logics

Semester(s) in which the module is S5

taught

Person responsible for the module Leila Ben Ayed

Language French

Relation to curriculum Compulsory

Teaching methods lesson, project (Prolog and Spin Tools)

Workload (incl. contact hours, self- Total workload: 50h

study hours) Contact hours: 30h (20h Lessons, 10h Exercises and Project)
Private study: 20h

Credit points 2 ECTS

Required and recommended pre- AI.1.1 Formal logic

requisites for joining the module AP.2.3 Automata theory and formal Languages
SE.4.1 Formal Development Methods

Module objectives/intended learning The objective of this lesson is to introduce the temporal logic used to
outcomes express and deduce system dynamic properties as well as fuzzy
logic for approached reasoning.

The practical aspect is implemented with Prolog and SPIN.

● Knowledge:
- Formal systems and formal inference for decision
- Theorem proving
- Fuzzification, defuzzificaion and decision
- Formal mapping rools
Competencies: C1, C4, C5, C9

Content Lesson I (06h). Introduction Theoretical aspects of Logic – Formal

proof and logical deduction
- Formal systems and deduction steps
- Inference rools and theorem proving
Lesson II (06h). Formal Systems for Temporal logics
- Formal system for Modal logic
- Formal system forTemporal logic
Lesson III (09h) . PLTL logic, Buchi models, automata composition
and semantic verification
- Buchi automata
- Modeling properties in Buchi
- Automata composition for the verification
Lesson III (09h) Fuzzy Logic- Fuzzification ans Defuzzification
- Modelling in Fuzzy logic
- Fuzzifcation Methods
- Defuzzification Methods

Examination forms 35% Continues evaluation + 65% Written exam

Study and examination require- 10/20

ments

10 Ref ID.230:01
 Reading list Manna, Z., & Pnueli, A. (1992). The Temporal Logic of Reactive and
Concurrent Systems: Specification. Springer.

Schnoebelen, Ph. (1999). Vérification de logiciels Techniques et outils

du model-checking. Vuibert.

Walper, P. (2001). Introduction à la Calculabilité. Dunod.

Gacogne, L. (1997). Eléments de logique floue. Hermès-Lavoisier.

AI.5.3 Applied Deep Learning

11 Ref ID.230:01
Module designation AI.5.3 Applied Deep learning

Semester(s) in which the module is S5

taught

Person responsible for the module (co- Nesrine Ben Yahia

ordinator)

Language French/English

Relation to curriculum Compulsory

Teaching methods lesson, lab works, project, seminar.

Workload (incl. contact hours, self- Total workload:50h

study hours) Contact hours :30h
Private study:20h

Credit points 2 ECTS

Required and recommended prerequi- MAT.1.1 Probability and Statistics

sites for joining the module MAT.4.1 Stochastic processes
AI.3.1 AI & Machine learning

Module objectives/intended learning - Students know the basics and the different types of deep
outcomes learning
- Students use and apply some supervised and unsupervised
deep learning algorithms

Content Chapter 0. Machine Learning Basics

1. Machine Learning types
2. Overfitting and Underfitting
3. Hyperparameters and Parameters
4. Estimators, Bias and Variance
5. Gradient-Based Learning
6. Stochastic Gradient Descent
7. Challenges Motivating Deep Learning
Chapter 1. Introduction to deep learning
1. What is deep learning?
2. Machine learning VS deep learning?
3. Artificial neural networks (ANN)
4. Deep neural networks (DNN)
5. Back-Propagation Algorithms
6. Regularization for deep learning
7. Optimization for training deep models
8. Typology
9. Applications
Chapter 2. Recurrent neural networks (RNN)
1. Definition
2. Algorithms
3. LSTM
4. GRU
5. Practical: LSTM with python
Chapter 3. Feed-Forward Neural Network (FFNN)
1. Definition
2. Example: Learning XOR
3. Algorithms
4. MLP
5. CNN
6. Practical: CNN with python
Chapter 4. Autoencoders (AE)
1. Definition

12 Ref ID.230:01
 2. Encoders & Decoders
3. Applications of Autoencoders
4. Practical: AE with python
Chapter 5. Deep Generative Models
1. Restricted Boltzmann Machines
2. Deep Belief Networks
3. Deep Boltzmann Machines
4. Convolutional Boltzmann Machines
Chapter 6. Introduction to Federated learning

Examination forms 35% Continues evaluation + 65% Written exam

Study and examination requirements 10/20

Reading list Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learn-
ing. MIT Press.

Chollet, F. (2017). Deep Learning with Python. (384 pages).

13 Ref ID.230:01
 AI.5.4 Natural Language Processing

14 Ref ID.230:01
 Module designation AI.5.4 Natural Language Processing (NLP)

Semester(s) in which the S5

module is taught

Person responsible for the Chiraz Ben Othmane Zribi

module (coordinator)

Language French

Relation to curriculum Compulsory

Teaching methods e.g. lecture, lab works, project

Workload (incl. contact Total workload:50h

hours, self-study hours) Contact hours : 30h (18h lecture + 6h exercise + 6h lab)
Private study:20h

Credit points 2 ECTS

Required and recom- AI.3.1 AI & Machine learning

mended prerequisites for
joining the module

Module objectives/intended This course introduces the fundamental concepts and techniques of NLP
learning outcomes by presenting the different levels of processing and giving a relatively
broad overview of the the commonly used algorithms and tools.
Objectives :
Students will gain an in-depth understanding of the computational prop-
erties of natural languages and the commonly used algorithms for pro-
cessing linguistic information.
Learning Outcomes:
The students will be able to:
- Understand key concepts from NLP those are used to describe and an-
alyze language
- Understand POS tagging and context free grammar for Natural lan-
guage
- Understand semantic representation of Englisfh Natural language for
processing
- Apply Machine learning /deep learning methods to process texts

Learning outcomes: C1, C8, C9 and C13

15 Ref ID.230:01
 Content 1. Introduction
- Overview of course content
- Motivations and challenges of TAL
- Examples of applications
- Levels of analysis (oral and written)
- Types of problems to solve
- Approaches and techniques for NLP
- Presentation of Python toolkit for NLP
2. Morphological analysis
- Presentation of the morphology (flexion, derivation)
- Representation of morphology by lexical lists
- Representation of morphology by finite state automata
- Representation of morphology by automata with transducers
3. Part-Of-Speech (POS) tagging
- Definitions (POS labeling, tag set,...)
- Rule-based taggers (Example: Brill's taggerà
- Probabilistic taggers (NGrams, based on hidden Markov chains)
4. Syntactic analysis and formal grammars
- General presentation of syntactic analysis
- Formal grammar: definition and types
- Context free grammar for NLP
- Syntax analysis with a context free grammar
- Common parsing algorithms (CYK, Earley)
5. Lexical semantics
- General presentation of the semantics
- Types of semantic relationships between words
- Semantic similarity between words based on a thesaurus
- Semantic similarity between words based on context
- Semantic vectors and construction methods (LSA, Word2Vec)
6. Machine Learning (ML) and deep learning (DL) for NLP
- Drawbacks of Statistics methods and linguistic methods
- Why machine learning / deep learning for NLP ?
- Construction of features
- Word embeddings
- Application 1: POS-tagging based en ML
- Application 2: Text classification based en DL
7. Dialog systems and Chatbots
- Rule based chatbots
- Corpus based chatbots
- Hybrid architectures
- Frame based dialog agents

- Voice XML
- Evaluating Dialog systems
7. Named Entity recognition (NER)
- Presentation of NER
- Frame based dialog agents
- Evaluating Dialog systems
8. Named Entity Recognition (NER)
- General presentation of NER
- Extraction of NE as sequence labelling based on super-
vised ML methods
- Extraction of relations based on pattern methods
- Extraction of relations based on semi-supervised meth-
ods
- Evaluating NER and relation extraction

Examination forms Project (35%)

Final exam (75%)

Study and examination re- 10/20

quirements
16 Ref ID.230:01
 Reading list Gunning, D. (2019). Natural Language Processing Fundamen-
tals. Packt Publishing.

Jurafsky, D., & Martin, J. (2020). Speech and Language Pro-

cessing.

Indurkhya, N., & Damerau, F. J. (Eds.). (2010). Handbook of

Natural Language Processing.

Gelbukh, A. (Ed.). (2008). Computational Linguistics and Intelli-

gent Text Processing.

AI.5.5 Robotics and soft computing

Module designation AI.5.5 Robotics & Soft Computing

Semester(s) in which the S5

module is taught

Person responsible for the Moncef TAGINA

module (coordinator)

Language French

Relation to curriculum Compulsory

Teaching methods Lesson, lab works.

Workload (incl. contact Total workload: 50h

hours, self-study hours) Contact hours:30h (18h lecture, 6h exercise, 6h lab)
Self-study:20h

Credit points 2 ECTS

Required and recom- MAT.1.1, MAT.1.2

mended prerequisites for
joining the module

Module objectives/intended In terms of Knowledge:

learning outcomes Soft computing is a set of so-called intelligent techniques that allow effi-
cient control of automated systems and particularly autonomous mobile
robots. The objective of this course is to present these techniques and
make them accessible to students for implementation in practical cases.
Competences: C2, C3, C5, C9

17 Ref ID.230:01
 Content
I. Introduaction to Robotics and Soft computing (1.5 h)
II. Classical commands in robotics (3 h)
- Introduction and definitions
- Position control
- Speed control
- Position and speed control
- Limits of classical methods
III. Fuzzy control (10.5 h)
- Introduction and definitions
- Fuzzy logic operator
- Fuzzy rules
- Inference
- Fuzzy control process
IV. Neural control (9h)
- Introduction and definitions
- Deterministic Perceptron
- Probabilistic Perceptron
- Multi Layer Perceptron and error backpropagation
- Implementation of neural control
V. Path optimization (6h)

- Introduction and definitions

- Genetic algorithm
- Application in robotics

Examination forms 100% Final exam

Study and examination re- 10/20

quirements

Reading list 1. Dieulot, J.-Y., Dubois, L., Borne, P., & Rozinoer, J. (1998). Intro-
duction à la commande floue. Editions Technip.
2. Faure, A. (2006). Classification et commande par réseaux de
neurones. Editions Hermès.
3. Santhosh, S. (2012). Soft-I-Robot: Soft Computing Techniques.
LAP Lambert Academic Publishing.

ISA.5.1 Big Data

18 Ref ID.230:01
 Module designation ISA.5.1 Big Data

Semester(s) in which the module is S5

taught

Person responsible for the module Raoudha Chebil

(coordinator)

Language French

Relation to curriculum Compulsory

Teaching methods lesson, lab works, presentations.

Workload (incl. contact hours, self- Total workload: 50 hours

study hours) Contact hours: 30 hours (20 hours lessons + 10 hours lab works)
Private study]: 20 hours

Credit points 2 ECTS

Required and recommended pre- DAT.1.1 Database and DBMS

requisites for joining the module

Module objectives/intended learn- Knowledge: Students:

ing outcomes -Master the basic building blocks of the Hadoop platform, namely
HDFS and MapReduce, and have an idea of the components of its
ecosystem;
-Master the MapReduce approach for problem solving;
-Understand the limits of the relational model and know the differ-
ent models of NOSQL databases.
Competences: C4, C7, C8

19 Ref ID.230:01
 Content Chapter I – Introduction to Big Data
1. Motivations
2. Definition
3. The 3Vs and the additional Vs
4. Benefits and challenges
5. Application examples
6. Stages of a Big Data project
7. New professions
8. Related fields
Chapter II – Hadoop: Building Blocks
1. Hadoop presentation
2. Hadoop history
3. Hadoop ecosystem
4. HDFS
5. MapReduce V1
6. MapReduce V2
7. Design Patterns MapReduce
Chapter III - Advanced Processing Tools
1. Data processing types
2. MapReduce review
3. Abstraction languages
a. Pig
b. Hive
4. Apache Spark
Chapter IV – NOSQL Databases
1. DBMS strengths
2. DBMS limits
3. BD NOSQL
4. BDR vs BD NOSQL
5. Study of BD NOSQL instances
a. Cassandra
b. MongoDB
Chapter V – Big Data Architectures
1. Motivations
2. Lambda architecture
3. Kappa architecture
4. Other architectures
5. Case study

Practical Works
1. Installation and testing of the working environment
2. HDFS
3. MapReduce
4. Pig and Hive
5. Spark
6. HBase

Examination forms 35% continuous evaluation (Lab works, presentations) ; 65% writ-
ten exam

Study and examination require- 10/20

ments

20 Ref ID.230:01
 Reading list Mooc
− “Fundamentals for Big Data”, Télécom ParisTech
− “Introduction to Hadoop and MapReduce”, Univer-
sity Nice Sophia Antipolis
Books
Bruchez, R. (2015). NoSQL Databases and Big Data: Un-
derstanding and Implementing. Editions Eyrolles.

Marr, B. (2015). Big Data: Using SMART Big Data, Analyt-

ics and Metrics to Make Better Decisions and Improve
Performance. John Wiley & Sons.

Zikopoulos, P., Eaton, C., et al. (2011). Understanding Big

Data: Analytics for Enterprise Class Hadoop and
Streaming Data. McGraw-Hill Osborne
Media.Classes
Nerzic, P. (2016). Hadoop Tools for Big Data. Lecture presented at
Rennes1 University, France.

ISA.5.3 Interactive Decision Support Systems

21 Ref ID.230:01
 Module designation ISA.5.3 Interactive Decision support systems (SIAD)

Semester(s) in which S5
the module is taught

Person responsible Imen BOUKHRIS

for the module (coor-
dinator)

Language French

Relation to curricu- optional

lum

Teaching methods lesson, lab works.

Workload (incl. con- Total workload:50h

tact hours, self-study Contact hours :30h
hours) Private study:20h

Credit points 2 ECTS

Required and rec- MAT.1.1, M.1.2, DAT 2.1

ommended prerequi-
sites for joining the
module

Module objec- In terms of Knowledge:

tives/intended learn- This course exposes students to the knowledge and skills needed to use computer
ing outcomes software to solve business problems, particularly to support decision making. It
involves the formalization of a decision problem and its resolution in order to assist
decision-makers in their decision-making process in semi-structured tasks. The
decision process is defined by the involvement of several variables which makes
it very complicated and difficult to manage. The objective of this module is to un-
derstand the decision-making process and investigate the different components of
a decision support system as well as the different models (e.g., multi-criteria, under
risk or uncertainty) in order to be able to apply them according to a given situation
(e.g., (individual, group, conflictual decision) as well as to the personality of the
decision maker (e.g., optimistic, pessimistic).

Competences: C1, C2, C4, C9

Content . I Introduction to decision theory (3 h)

II. What are decision support systems ? (3 h)
- DSS vs Information system
- Decision systems classifications
- Group decision making
- DSS Structures
- DSS Architectures
III. Multi-criteria decision making (6h)
- Pareto-optimality
- Analytical Hierarchical Process (AHP)
- Analytical Network Process (ANP)
IV. Decision making under uncertainty (9 h)
- Introduction and definitions
- Objective vs subjective probability
- Lottery and Utility
- Maximax, Wald, Hurwicz, Laplace, Savage criteria
V. Decision making under risk (9h)
- Introduction and definitions
- Raiffa Decision trees
- Influence diagrams

Examination forms 100% Final exam

22 Ref ID.230:01
 Study and examina- 10/20
tion requirements

Reading list 1. Alter, S. (1977). A taxonomy of decision support systems. Sloan Manage-
ment Review (pre-1986), 19(1), 39.
2. Sprague Jr, R. H. (1980). A framework for the development of decision sup-
port systems. MIS quarterly, 1-26.
3. Marakas, G. M. (2003). Decision Support Systems in the 21st Century (Vol.
134). Upper Saddle River: Prentice Hall.
4. Marchau, V. A., Walker, W. E., Bloemen, P. J., & Popper, S. W. (2019).
Decision Making Under Deep Uncertainty: From Theory to Practice (p.
405). Springer Nature.

ISA.5.2 Data mining

23 Ref ID.230:01
 Module designation ISA.5.2 Data Mining

Semester(s) in which the module S5

is taught

Person responsible for the mod- Aroua Hedhili

ule (coordinator)

Language French

Relation to curriculum Compulsory

Teaching methods Lesson, lab works, project.

Workload (incl. contact hours, Total workload: 50h

self-study hours) Contact hours :30h (20h lessons 10h labs)
Private study: 20h

Credit points 2 ECTS

Required and recommended MAT.1.1: Probability and Statistics

prerequisites for joining the mod- AI .3.1.: IA & Machine Learning
ule Students must be competent in python.

Module objectives/intended Key question: what learning outcomes should students attain in the
learning outcomes module?
E.g. in terms of:
- Knowledge: Knowing standard data mining methods and tech-
niques such as association rules, data clustering and classifica-
tion.
- Skills: Students are able to how apply standard and new tech-
niques on datasets of realistic sizes using modern data analysis
frameworks.
- Competences: Students know new and advanced techniques
for emerging applications (e.g. social network analysis, stream
data mining).
Competencies: C1, C2, C3, C4, C5

24 Ref ID.230:01
 Content Chapter 1: Introduction and Motivation
1. Why Data Mining?
2. What is Data Mining?
3. What is NOT Data Mining
4. What kind of data can be mined?
5. What kind of knowledge can be extracted?
6. When to use data mining?
7. Data Mining applications
8. Overview of data mining workflow
9. Data mining Project Methodology
Chapter 2: Data Understanding
1. Data Types
2. Numeric Attributes
3. Categorical Attributes
4. Data exploration
5. Data exploration tools
6. Statistical Measures
7. Basic Graphs
Chapter 3: Data cleaning
1. Goal of data cleaning
2. Common data problems
3. Data problems detection
4. Data cleaning methods
Chapter 4: Feature Engineering
1. Basic concepts
2. Common features transformation
3. Transformation for numeric variables
4. Transformation for categoric variables
5. Python functions and classes
Chapter 5: Feature Selection
1. Basic concepts
2. Features selection approaches
3. Dimensionality Reduction with PCA
4. Dimensionality Reduction with T-SNE
5. Python functions and classes
Chapter 6: Clustering
1. Introduction
2. Partitioning methods
3. Hierarchical methods
4. Density-based methods
5. Grid-based methods
Chapter 6: Learning Algorithms
1. Related concepts
2. Linear Regression
3. Decision Tree
4. Probabilistic approach
5. KNN Algorithm

Examination forms 35% in-class evaluation (project, labs) ; 65% written exam

Study and examination require- 10/20

ments

25 Ref ID.230:01
 Reading list Han, J., Pei, J., & Kambe, M. (2012). Data Mining: Concepts
and Techniques (3rd ed.). Elsevier. ISBN 978-0-12-
381479-1.

Zaki, M. J., & Meira Jr, W. (2020). Data Mining and Machine
Learning: Fundamental Concepts and Algorithms (2nd
ed.). Cambridge University Press. ISBN: 978-
1108473989.

Cornuéjols, A., Miclet, L., & Barra, V. (2021). Apprentissage

Artificiel, Concepts et Algorithmes - De Bayes et Hume
au Deep Learning. Collection Algorithmes.

MAT.5.1 Combinatorial Optimization

26 Ref ID.230:01
 Module designation MAT.5.1 Combinatorial optimization

Semester(s) in which the S5

module is taught

Person responsible for Olfa BELKAHLA DRISS

the module

Language French

Relation to curriculum Optional

Teaching methods lecture, lesson, lab works, project, exercises

Workload (incl. contact Total workload:50h

hours, self-study hours) Contact hours :30h (20h lecture, 10h exercices)
Private study:20h

Credit points 2 ECTS

Required and recom- AP.1.2 Algorithms and data structure

mended prerequisites for MAT.2.1 Graph theory
joining the module

Module objectives/in- - Students know that combinatorial optimization consists in solving

tended learning out- many complex problems and finding an optimal solution in a large but
comes finite search space.
- Students know that combinatorial optimization is concerned with the
study of efficient algorithms to solve these problems by efficiently ex-
ploring the space of solutions.
- Students know categorizing problems into complexity classes.
- Students are able to model and solve not only polynomial problems,
but also NP-complete and NP-hard problems.
- Students are able to apply exact methods, such as Branch and
Bound, as well as approximate methods such as TS, RS, ACO, GA,
etc. in order to solve many practical problems to avoid the combinato-
rial explosion and overcome the complexity and find solutions in a
reasonable time.
- Student are able to implement algorithms and perform experimenta-
tions on benchmark instances.
Competencies: C1, C4, C9, C13

27 Ref ID.230:01
 Content Chapter 1: Introduction to combinatorial optimization (4H)
1. Definitions
2. Applications
3. Problem Modeling
4. Examples of combinatorial optimization problems
5. Problem Complexity
Chapter 2: Resolution methods (4H)
1. Categories of resolution methods of CO problems
2. Exact Methods
3. Overview on Branch & Bound
4. Approximate methods: heuristics and metaheuristics
Chapter 3: Constructive methods (4H)
1. Constructive approaches: Greedy algorithms
2. Greedy heuristics on the Knapsack problem
3. Greedy heuristics on scheduling problems
4. Greedy heuristics on the TSP
5. Greedy heuristics on the Bin Packing Problem
6. Advantages and disadvantages of constructive methods
7. Evaluation of the performance of heuristics
Chapter 4: Local search based methods (6H)
1. Scientific history: Development of heuristics
2. Introduction to local search: Fable of the hikers
3. Gradient descent
4. Neighborhoods
5. Simulated Annealing
6. Taboo Search
Chapter 5: Evolutionary approaches (6H)
1. Definitions
2. Choice of a metaheuristic
3. Ant Colony Optimization
4. Genetic algorithms
Chapter 6 : Hybrid approaches (3H)
1. Advantages of hybridization
2. Examples of hybrid approaches
Chapter 7 : Multi-objective optimization (3H)
1. Problem Modeling with MOO
2. Dominance relationship
3. Pareto optimality
4. NSGA-II
5. Quality measures
6. Multi-objective optimization frameworks

Examination forms 35% continuous assessment + 65% written exam

Study and examination 10/20

requirements

28 Ref ID.230:01
 Reading list Martello, Silvano; Toth, Paolo (1990), "Bin-packing problem", Knapsack
Problems: Algorithms and Computer Implementations, Chichester, UK:
John Wiley and Sons, ISBN 0471924202
Bernhard Korte, Jens Vygen, Jean Fonlupt, Alexandre Skoda « Optimisa-
tion combinatoire Theorie et algorithmes », Collection IRIS, Springer,
2010.
Alexander Schrijver « Combinatorial Optimization Polyhedra and Efficiency
», Collection Algorithms and Combinatorics, 2003
Jerry Swan, Steven Adriaensen, Alexander E.I. Brownlee, Kevin Ham-
mond, Colin G. Johnson, Ahmed Kheiri, Faustyna Krawiec, J.J. Merelo,
Leandro L. Minku, Ender Özcan, Gisele L. Pappa, Pablo García-Sánchez,
Kenneth Sörensen, Stefan Voß, Markus Wagner, David R. White, “Me-
taheuristics “In the Large””, European Journal of Operational Research,
Volume 297, Issue 2, 2022, Pages 393-406, ISSN 0377-2217,
https://doi.org/10.1016/j.ejor.2021.05.042.
https://www.sciencedirect.com/science/article/pii/S0377221721004707
El-Ghazali Talbi, “Metaheuristics: From Design to Implementation”, Wiley,
ISBN: 978-0-470-27858-1 June 2009.

MAT.5.2 Category theory and functional programming

29 Ref ID.230:01
 Module designation MAT.5.2 Category theory and functional programming

Semester(s) in which the module is S5

taught

Person responsible for the module Fethi Kadhi

Language French

Relation to curriculum Compulsory

Teaching methods Lessons, Lab, Projects.

Workload (incl. contact hours, self- Total workload:50h

study hours) Contact hours :30h
Private study:20h

Credit points 2 ECTS

Required and recommended prerequi-

sites for joining the module

Module objectives/intended learning The objectives of this modules are:

outcomes - The Introduction of the category theory
- Application of the category theory to func-
tional programming
- The study of Haskell as a pure functional
language programming
Competencies:
C1, C9

Content 1. Categories
1.1 Definition
1.2 Examples
2. Functors and natural transformations
2.1 Definition
2.2 Examples
3. Yoneda lemma.
3.1 Representable functors
3.2 Yoneda lemma
4. Functional programming
4.1 Imperative programming
4.2 Functional programming
5. Hask category
5.1 Hask types
5.2 Hask functions
5.3 Hask functors
6. Hask Monads
6.1 Maybe monad
6.2 List monad

Examination forms 100% Written final exam

Study and examination requirements 10/20;

Reading list Leinster, T. (2014). Basic Category Theory. Cambridge

Studies in Advanced Mathematics. Cambridge Press.
Serrano Mena, A. (2022). Practical Haskell. Apress.

AI.5.6 Knowledge Representation & Reasoning

Module designation AI.5.6 Knowledge representation and Reasoning
30 Ref ID.230:01
 Semester(s) in which the module S5
is taught

Person responsible for the mod- Anja Habacha

ule (coordinator)

Language French

Relation to curriculum Compulsory

Teaching methods lesson, lab works, projects.

Workload (incl. contact hours, Total workload: 25h

self-study hours) Contact hours: 15h (9h lessons, 6h lab works)
Private study in hours: 10h

Credit points 1 ECTS

Required and recommended IA.1.1 Formol Logic

prerequisites for joining the SE.3.2 Object Oriented Analysis & Design
module

Module objectives/intended In terms of Knowledge:

learning outcomes One of the objectives of artificial intelligence is to endow computer sys-
tems with intellectual capacities comparable to those of humans to pro-
duce "thinking" machines capable of reasoning. It is a question of dealing
with problems requiring knowledge in a very specific field. This course ex-
poses the diverse nature of the knowledge acquired, the problems of their
representation in an incomplete, uncertain context, etc., the existing for-
malisms of representation and the means of interpretation allowing the im-
plementation of the reasoning.

Competencies: C1, C2, C4, C9

Content Chapter 1: Introduction (3h)

- History
- AI approaches
- AI Definitions
- Knowledge Definition
- Knowledge type and nature
Chapter 2: Relational representations (6h)
- Classical logic
- Non-classical logic
- Production rules
Chapter 3: Object representations (6h)
- Semantic networks
- Frames
- Ontologies
- Practical case of biomedical domain

Examination forms 35% continuous assessment + 65% written exam

Study and examination require- 10/20

ments

Reading list Russell, S., & Norvig, P. (2022). Artificial Intelligence: A Modern Ap-
proach (4th USA edition).
Barr, A., & Feigenbaum, E. A. (1986). Le manuel de l’Intelligence Artifi-
cielle. Tome I. Eyrolles.
Laurière, J.-L. (1988). Intelligence Artificielle - Représentation des Con-
naissances. Tome II. Eyrolles..

31 Ref ID.230:01
 ISA.5.6 Text Mining

32 Ref ID.230:01
 Module designation ISA.5.6 Text Mining

Semester(s) in which S5
the module is taught

Person responsible Imtiez FLISS

for the module (co-
ordinator)

Language French/English

Relation to curricu- Compulsory

lum

Teaching methods lectures, lab works and project.

Workload (incl. con- Total workload: 25h

tact hours, self-study Contact hours : 15h
hours) Private study: 10h

Credit points 1 ECTS

Required and recom- ISA.5.2 Data Mining

mended prerequi-
sites for joining the
module

Module objec- Key question: what learning outcomes should students attain in the module?
tives/intended learn- At the end of this module, students should:
ing outcomes – Understand the need for Text Mining
– Know the definition of Text Mining
– Know the difference between Text Mining and data mining
– Know the different fields of application and the challenges of Text Mining
– Understand the process of Text Mining applications
– Design and develop a Text Mining application

Learning outcomes: C1, C2, C3, C4, C5

Content Chapter 1: Introduction

1. Definition and importance of Text Mining
2. Applications of Text Mining
3. Text mining pipeline
4. Challenges
Chapter 2: Text Pre-processing and text feature extraction
1. Importance of Text Data Preprocessing
2. Text Cleaning and Pre-processing techniques
3. Text Cleaning and Pre-processing using Python
4. Feature extraction
a) Feature extraction models (Bag of words, TFIDF, Word2vec, GloVe, FastText, BERT)
b) Feature extraction models using Python
c) Advantages and Limitations of Feature extraction Models
Chapter 3: Text classification
1. Text Classification definition
2. Applications of Text classification
3. Different levels of scope
4. Text Classification Pipeline
a) Dataset Preparation
b) Feature extraction
c) Feature Selection

33 Ref ID.230:01
 d) Classification Techniques
e) Evaluation of Text Classifiers
5. Text classification using Python
Chapter 4: Sentiment analysis
1. Sentiment Analysis definition
2. Sentiment analysis applications
3. Different Levels of Sentiment Analysis
4. Supervised Sentiment analysis pipeline
5. Unsupervised Sentiment analysis pipeline
6. Sentiment Analysis Challenges
7. Sentiment analysis using Python
Chapter 5: Topic identification
1. Topic Modelling definition
2. Applications of Topic Modelling
3. Topic modelling algorithms (Latent Dirichlet Allocation, Non-Negative Matrix Fac-
torization)
4. Topic modelling algorithms using Python
5. Topic modelling evaluation
6. Topic modelling evaluation using Python

Examination forms 100% Lab works and project

Study and examina- 10/20

tion requirements

Reading list Konchady, M., & Sanger, J. (2006). Text Mining Application Programming. Boston:
Charles River Media.

Srivastava, A. N., & Sahami, M. (Eds.). (2009). Text Mining: Classification, Clustering,
and Applications. CRC Press.

Aggarwal, C. C., & Zhai, C. (Eds.). (2012). Mining Text Data. Springer Science & Busi-
ness Media.

Berry, M. W., & Kogan, J. (2010). Text Mining: Applications and Theory. West Sussex,
PO19 8SQ, UK: John Wiley & Sons.

Bengfort, B., Bilbro, R., & Ojeda, T. (2018). Applied Text Analysis with Python: Ena-
bling Language-Aware Data Products with Machine Learning. O'Reilly Media, Inc.

Sarkar, D. (2019). Text Analytics with Python: A Practitioner's Guide to Natural Lan-
guage Processing. Apress.

Silahtaroğlu, G. (2019). Concepts of Text Mining with Python and Real-Life Exercises.

Žižka, J., Dařena, F., & Svoboda, A. (2019). Text Mining with Machine Learning: Princi-
ples and Techniques. CRC Press.

34 Ref ID.230:01
 ISA.5.5 Recommender systems
Module designation ISA.5.5 Recommender Systems

Semester(s) in which S5
the module is taught

Person responsible for Sonia Ben Ticha

the module (coordina-
tor)

Language French

Relation to curriculum Optional

Teaching methods lesson, lab works, projects

Workload (incl. contact Total workload: 50h

hours, self-study hours) Contact hours: 30h (16h lessons, 14h lab works)
Private study: 20h

Credit points 2 ECTS

Required and recom- MAT.1.1, M.1.2,

mended prerequisites Linear algebra
for joining the module Algorithmics and data structures
python

Module objectives/in- In terms of Knowledge:

tended learning out- This course introduces the foundations of the recommender systems with a fo-
comes cus on personalised recommender systems.
Students Have a general overview of
- Personalized user modelling
- Collaborative recommender systems
- Content-based recommender systems
- Evaluation of a recommender systems
- Implementing a collaborative filtering recommender system in py-
thon
Competencies: C1, C2, C4, C9

Examination forms 35% continuous assessment + 65% written exam

Study and examination 10/20

requirements

Reading list Chevalier, M., Julien, C., & Soulé-Dupuy, C. (2009). Collaborative and Social In-
formation Retrieval and Access: Techniques for Improved User Modeling. In-
formation Science Reference.

Ricci, F., Rokach, L., Shapira, B., & Kantor, P. B. (Eds.). (2011). Recommender
Systems Handbook. Springer US. e-ISBN 978-0-387-85820-3.

ISA.5.4 Business Intelligence

35 Ref ID.230:01
 Module designation ISA.5.4 Business intelligence

Semester(s) in which the module S5

is taught

Person responsible for the mod- Manel BenSassi

ule (coordinator)

Language French

Relation to curriculum Optional

Teaching methods lesson, lab works.

Workload (incl. contact hours, Total workload: 50h

self-study hours) Contact hours:30h (21h lesson, 9h Lab works).
Self study: 20h

Credit points 2 ECTS

Required and recommended DAT.1.1 Database and DBMS

prerequisites for joining the
module

Module objectives/intended This course refers to technologies, applications and practices of hetero-
learning outcomes geneous data integration, storage, multidimensional analysis, and visual-
ization to support business decision making.
Thus, the student will be able to propose concrete conceptual and tech-
nological architecture for the integration od heterogenous data in the pro-
fessional environment as he will acquire many competencies such as:
- Become able to evaluate the technologies that make up
BI (data Warehousing, OLAP)
- Become able to plan the implementation of a BI architec-
ture.
Learning outcomes: C1, C2, C3, C4, C8, C9

Content Chapter 1 : Understanding Business intelligence

- The challenge of decision making
- What is business intelligence
- The BI value chain and value
Chapter 2: Data Integration
- Data integration motivation
- ETL Process
- ETL techniques
Chapter 3: Data Storage: Data Warehousing
- What is data warehousing?
- Data Marts and analytical Data
- Organization of DataWarehouse
- Data access
Chapter 4: Multi dimensional Analysis with OLAP
- Definitions
- OLAP vs OLTP
- Operational data stores
- Multi-Dimensions techniques
- OLAP architecture
Chapter 5 : MDX Language
- Problem presentation
- MDX Syntax and Request

Examination forms 35% Continues evaluation + 65% Written exam

Study and examination require- 10/20

ments

36 Ref ID.230:01
 Reading list Fernandez, A. (2013). Les nouveaux tableaux de bord des managers: le
projet Business Intelligence clés en main (6ème édition). Eyrolles.

Fernandez, A. (2013). L’essentiel du tableau de bord: Concevoir le ta-

bleau de bord de pilotage avec Microsoft Excel (4ème édition).

Galzy, C., Girona, P., Martin, B., Nicoloso, C., & Vandermoere, J. (May
2010). La Business Intelligence, Livre Blanc.

AI.5.7 Deep Reinforcement Learning

Module designation AI.5.7 Deep Reinforcement Learning

Semester(s) in which the module S5

is taught

Person responsible for the mod- Wided Chaari

ule (coordinator)

Teaching team Sabrine Krichen (Instadeep)

Language English

Relation to curriculum Optional

Teaching methods Lessons + lab works + project

Workload (incl. contact hours, Total workload: 25h

self-study hours) Contact hours: 15h (10h lessons, 5h lab session)
Private study: 10h

Credit points 1 ECTS

Required and recommended MAT.1.1 Probability and Statistics

prerequisites for joining the MAT.4.1 Stochastic processes
module AI.3.1 AI & Machine Learning
AI.5.3 Applied Deep Learning

Module objectives/intended Students have a good understanding of Reinforcement Learning.

learning outcomes Students know the difference between the different deep reinforcement
learning algorithms.
Students are able to implement a deep reinforcement learning algorithm
on an open ai gym environment.

Competencies: C1, C2, C4, C9

Content Introduction to reinforcement learning.

Value based methods in reinforcement learning.
Policy based methods in reinforcement learning.
Reinforcement learning tips and tricks.

Examination forms Deep reinforcement project on GitHub + mini report.

Questions answering session.

Study and examination require- Implement a converging deep reinforcement learning algo-
ments rithm on an Open AI gym environment.
Have a good understanding of the implemented work.
Have an organized GitHub repo of the project.
Write a very clear mini report.

Reading list Sutton, R., & Barto, A. Reinforcement Learning: An Introduction.

Lapan, M. Reinforcement Learning Hands-On: A Book with Tutorials.

37 Ref ID.230:01
 ISA.5.7 Distributed databases
Module designation ISA.5.7 Distributed Databases

Semester(s) in which the module is S5

taught

Person responsible for the module Raoudha KHCHERIF

(coordinator)

Language French

Relation to curriculum Optional

Teaching methods lecture, lesson, assignment, labs

Workload (incl. contact hours, self- Total workload: 25 h

study hours) Contact hours: 15 h
Private study: 10h

Credit points 1 ECTS

Required and recommended pre- DAT.2.1, DAT.2.2, NET3.1, NET3.2, NET4.1

requisites for joining the module

Module objectives/intended learn- This course will deal with the fundamental issues in large distributed
ing outcomes database systems which are motivated by the computer networking
and distribution of processors, and control. The theory, design, speci-
fication, implementation, and performance of large systems will be
discussed.

Competencies:C1, C2, C8, C13

Content I INTRODUCTION
II. BDR DESIGN AND IMPLEMENTATION
II. TRANSACTION AND COMPETITOR ACCESS
IV. OPTIMIZATION OF DISTRIBUTED QUERIES

Examination forms 100% written Exam

Study and examination require- 10/20

ments

Reading list Özsu, M. T., & Valduriez, P. (2011). Principles of Distributed Database
Systems. Springer.

Rahimi, S. K. (2010). Distributed Database Management Systems.

John Wiley & Sons Inc.

ISA.5.8 Advanced BI

38 Ref ID.230:01
 Module designation ISA.5.8 Advanced Business intelligence

Semester(s) in which the mod- S5

ule is taught

Person responsible for the Manel BenSassi

module (coordinator)

Language French

Relation to curriculum Optional

Teaching methods lesson, lab works.

Workload (incl. contact hours, Total workload: 25h

self-study hours) Contact hours: 15h (9h lesson, 6h Lab works).
Self study: 10h

Credit points 1 ECTS

Required and recommended ISA.5.4 Business intelligence

prerequisites for joining the
module

Module objectives/intended This course refers to new technologies, applications and practices of Big
learning outcomes and heterogeneous data integration, storage, multidimensional analysis,
and visualization to support business decision making in a distributed envi-
ronment.
Thus, the student will be able to propose concrete conceptual and techno-
logical architecture for the integration of big data in the professional envi-
ronment as he will acquire many competencies such as:
- Become able to evaluate the technologies that adapt BI
concept to new data challenges (datalake, real time dataware-
housing, ect)
- Become able to plan the implementation of a Big data based
BI architecture.

Learning outcomes: C1, C2, C3, C4, C8, C9

Content Chapter 1: Understanding Business intelligence

- The challenge of Business Intelligence
Chapter 2: Data Storage: DataLake and NoSQL datawarehousing
- Limits of classical storage
- Data lake and data Acess
- NoSQL DataBase: document, key-value, column and graph
database
Chapter 3: Data integration and analysis
- ETL vs ELT
- ELT Process and techniques
- OLAP in NoSQL dataBase
Chapter 4: Graph analytics techniques
- Presentation of Neo4j
Chapter 5 : Introduction to recommender system with graph
- Concept presentation
- Evaluation and metrics

Examination forms 30% Continues evaluation + 70% Written exam

Study and examination re- 10/20

quirements

Reading list  Fernandez, A. (2013). Les nouveaux tableaux de bord des mana-
gers: le projet Business Intelligence clés en main (6ème édition).
Eyrolles.

39 Ref ID.230:01
  Bruchez, R. (02/12/2021). Les bases de données NoSQL: Com-
prendre et mettre en œuvre (3ème édition).
 Snowflake & Talend. Construire un Data Lake dans le Cloud: Livre
blanc.

40 Ref ID.230:01
 AI.5.10 Semantic web & Linked Data

41 Ref ID.230:01
 Module designation AI.5.10 Semantic Web & Linked Data

Semester(s) in which S5
the module is taught

Person responsible for Anja Hbacha

the module (coordina-
tor)

Language French

Relation to curriculum Optional

Teaching methods lesson, lab works, projects.

Workload (incl. contact Total workload: 25h

hours, self-study Contact hours: 15h
hours) Private study: 10h

Credit points 1 ECTS

Required and recom- AI.3.1 AI & Machine Learning

mended prerequisites
for joining the module

Module objectives/in- This course aims to:

tended learning out- represent the standards of the Web of data and the Semantic Web.
comes represent and publish linked data on the Web (RDF);
to interrogate and select data remotely and through the Web (SPARQL);
Students are able to semantically link relationships between Web resources, real
world resources, and concepts through the use of Linked Data enabled by Re-
source Description Framework (RDF).

Learning outcomes: C1, C2, C3, C4, C5

Content Chapter1: Introduction to Semantic Web & Linked Data

1. Brief history of the web
2. Architectural Principles
3. Standards and deployment
4. Linked Data Principles
7. The Standardization Stack

Chapter2: RDF data model

1. Resources
2. Model of triples and graphs
3. Serialization Syntaxes
4. Values, types and languages

Chapter3: SPARQL query language

1. RDF graph matching
2. Filters, constraints and functions
3. Different forms of requests
4. Modification of RDF databases

Chapter4: Supporting RDFS ontologies and schemas

1. Introduction to ontologies
2. Schemas for and by RDF
3. Class hierarchies
4. Property Hierarchies
5. Property signatures
6. Schema Documentation

Chapter4: Formalization in OWL

42 Ref ID.230:01
 1. Class relations
2. Characterization of properties
3. Equivalences and alignments
4. Restriction of properties
5. Manage schemas
6. OWL Profiles

Examination forms 50% Continues evaluation + 50% Final project

Study and examination 10/20

requirements

Reading list Hitzler, P., Krötzsch, M., & Rudolph, S. (2009). Foundations of Semantic Web
Technologies. Chapman & Hall/CRC.

Patel, A., Debnath, N. C., & Bhushan, B. (2022). Semantic Web Technolo-
gies: Research and Applications. CRC Press.

AI.5.8 Affective computing

Module designation AI.5.8 Affective computing

Semester(s) in which the mod- S5

ule is taught

Person responsible for the Imtiez Fliss

module (coordinator)

Language French

Relation to curriculum Optional

Teaching methods lesson, lab works, projects.

Workload (incl. contact hours, Total workload: 25h

self-study hours) Contact hours: 15h
Private study:10h

Credit points 1 ECTS

Required and recommended AI.3.1 AI & Machine Learning

prerequisites for joining the
module

Module objectives/intended This course aims to:

learning outcomes 1. master the basic concepts of emotion and affective computing;
2. be able to design an emotion recognition system
3. have knowledge of some computational models of emotions for a robot and
for a virtual character;
4. learn about the presentation of scientific work.

Learning outcomes: C1, C2, C3, C4, C5

Content Chapter 1: Introduction

Section I: Importance of emotions
Section2: Emotional Intelligence
Section 3: Affective Computing

Chapter II: Emotions

Introduction
Section I: Emotion Definitions
Section II: Types of emotion
Section III: Components of an emotion

43 Ref ID.230:01
 Section IV: Theories of emotions
Section V: Representation of emotions

Chapter III: Automatic Emotion Recognition

Introduction
Section I: Recognition of emotions by facial expressions
Section II: Recognition of emotions by voice analysis
Section III: Recognition of emotions by gesture analysis
Section IV: Recognition of emotions from physiological signals
Section V: Multimodal Recognition of Emotions

Chapter IV: Computer Modeling of Emotions

Introduction
Section I: Computational models of emotions
Section II: Computational modeling of specific emotions for virtual agents
Section III: Examples of expressive virtual agents
Section IV: Modeling emotions in robotics
Section V: Examples of Expressive Companion Robots

Examination forms 35% Continues evaluation + 65% Written exam

Study and examination require- 10/20

ments

Reading list Picard, R. (1997). Affective Computing. MIT Press.

Salovey, P., Bedell, B., Detweiler, J., & Mayer, J. (2000). Current directions
in emotional intelligence research. In M. Lewis & J. Haviland-Jones
(Eds.), Handbook of Emotions (pp. 504–520). Guilford Press.

Gratch, J., & Marsella, S. (2007). Computational models of emotion: De-

signs and evaluation. Tutorial ACII’2007.

Baggia, P., Burkhardt, F., Pelachaud, C., Peter, C., & Zovato, E. (2011).
Emotion Markup Language.

Poria, S., Cambria, E., Bajpai, R., & Hussain, A. (2017). A review of affec-
tive computing: From unimodal analysis to multimodal fusion. Infor-
mation Fusion, 37, 98-125.

44 Ref ID.230:01
 AI.5.9 Intelligent transport Systems
Module designation AI.5.9 Intelligent Transport Systems

Semester(s) in which the mod- S5

ule is taught

Person responsible for the mod- Walid Chaker

ule (coordinator)

Language French

Relation to curriculum Optional

Teaching methods lesson, lab works, projects.

Workload (incl. contact hours, Total workload: 25h

self-study hours) Contact hours: 15h
Private study:10h

Credit points 1 ECTS

Required and recommended AI.3.1 AI & Machine Learning

prerequisites for joining the
module

Module objectives/intended The objective of the course is to offer students a panorama of modern
learning outcomes methodologies and technologies for Intelligent Transport Systems (ITS)
through the study of real projects organized by theme.
Following the course, the student will be able to:
- Propose an ITS project using the knowledge acquired in the other
training modules
- Estimate the technological and budgetary resources to be deployed in
an ITS project
- Get an idea of recent trends and national and international ITS pro-
grams

Learning outcomes: C2, C3, C4, C5

Content Chapter I: Introduction to ITS

− Brief history
− Objectives of ITS, architecture and orientations

Chapter II: Examples of projects

(one or more examples for each of the following headings)
− ATIS (Advanced Traveler Information Systems)
− ATMS (Advanced Transportation Management Systems)
− APTS (Advanced Public Transportation Systems)
− Connected Vehicles and Autonomous Vehicles
− ITS in crisis and emergency management

Chapter III: ITS in Tunisia

− Strategy of the Ministry of Transport in the field of ITS
− Examples of completed projects
− Presentation of student initiatives

Examination forms 50% Continues evaluation + 50% Final project

Study and examination require- 10/20

ments

Reading list Pagano, P. (2016). Intelligent Transportation Systems: From Good

Practices to Standards (1st ed.). CRC Press.

45 Ref ID.230:01
 DOS.5.8 Introduction to Cloud computing

46 Ref ID.230:01
 Module designation DOS.5.8 Introduction to cloud computing

emester(s) in which the module 5

is taught

Person responsible for the mod- Dr. Mehrez Essafi

ule (coordinator)

Teaching team -

Language French

Relation to curriculum Compulsory

Teaching methods • Lesson

• Lab work

Workload (incl. contact hours, Total workload: 25h

self-study hours) Contact hours: 15 (12h lessons, 3h lab work)
Private study: 10h

Credit points 1 ECTS

Required and recommended OS.2.1 Introduction to Operating systems and Unix environment
prerequisites for joining the NET.3.1 Local Networks
module NET.4.1 Computer Networks
SE.4.2 Software Architecture
SEC.4.1 Cybersecurity & cryptography

Module objectives/intended • Introduce cloud computing as the provision of computing re-

learning outcomes sources.
• Expose modern systems architectures and software development
kits that, together, provide cloud-computing frameworks.
• Learn about different aspects of the design, development, provi-
sioning and management of cloud-based applications.
• Gain a sound understanding of cloud-based computing and the op-
portunities that it provides for a diverse range of computing applica-
tions.
• Make a special attention to security of cloud-based applications
and the different strategies that are available in these deployments.
By the end of the course, students are expected to be able to:
• Apply services and architectures offered by virtualisation and
cloud platforms
• Explain how cloud platforms can be used by organisations to
achieve efficiencies in developing, operating and scaling modern
applications
• Propose new ways of modifying, extending or combining existing
methodologies and implementation techniques
• Apply international standards documents and application program-
ming interface documentation
• Examine the potential for, and practicality of, developing cloud
computing architectures for specific applications
Competencies: C2, C4, C5, C6, C7, C9, C10.

47 Ref ID.230:01
 Content Unit 1 – Cloud Computing: main concepts
• General introduction
• Historical overview
• Cloud characteristics
• Business model
• Advantages and limits
Unit 2 – Data-centres
• Definitions
• Main components
• Green Computing
• Security
• High Availability
Unit 3 – Cloud Services and deployment models
• IaaS (Infrastructure as a Service)
• PaaS (Platform as a Service)
• SaaS (Software as a Service)
• FaaS (Function as a Service)
• Other services
• Public Cloud
• Private Cloud
• Hybrid Cloud
• Community Cloud
• DevOps approach
Unit 4 – Virtualization
• Definitions
• Architectures
• Solutions
• Servers virtualization
• Containers
• Storage virtualization
Unit 5 – AI and Cloud Computing
• Machine learning cloud services
• Cloud AI platforms

Examination forms • 20% labs

• 80% written examination

Study and examination require- Student must achieve an overall minimum module mark of 10/20
ments

Reading list Mell, P., & Grance, T. (2011). The NIST Definition of Cloud Computing
(800-145). National Institute of Standards and Technology (NIST).
Duncan, C. H. (2017). Cloud Computing Gateway, Cloud Computing
Hypervisor, and Methods. International Conference on Cloud Compu-
ting.
Hennion, R., Tournier, H., & Bourgeois, E. (2014). Cloud Computing:
Décider - Concevoir - Piloter – Améliorer.
Plouin, G. (2014). Cloud Computing: Sécurité, Stratégie d'Entreprise et
Panorama du Marché. Collection InfoPro, Dunod.
Rapport Cigref. (2013). Fondamentaux du Cloud Computing: Le Point
de Vue des Grandes Entreprises.
Moyer, C. M. (2011). Building Applications in the Cloud: Concepts, Pat-
terns, and Projects. Addison-Wesley.
Marks, E. A., & Lozano, B. (2010). Executive’s Guide to Cloud Compu-
ting. Wiley.
Kshirsgar, P. (2021). Green Computing in Cloud Technology. Linguistica
Antverpiensia.
Hwang, K. (2017). Cloud Computing for Machine Learning and Cogni-
tive Applications. The MIT Press.

DOS.5.2 Blockchain
48 Ref ID.230:01
Module designation DOS.5.2 Blockchain

Semester(s) in which the S5

module is taught

Person responsible for the Hanen Idoudi

module

Teachers team Hanen Idoudi, Mohamed Houcine Hdhili,

Language French

Relation to curriculum Compulsory

Teaching methods Lesson, Lab works

Workload (incl. contact Total workload:25h

hours, self-study hours) Contact: 15h (9 lesson, 6h lab works)
Private study: 10h

Credit points 1 ECTS

Required and recom- SEC.4.1 Cybersecurity and Cryptography

mended prerequisites for
joining the module

Module objectives/in- Knowledge:

tended learning outcomes After completing this course, students should be able to:
• Explain blockchain and how it is applied across industries.
• Describe key principles of blockchain technology and the benefits and value that
they bring to enterprises.
• Explain the role of a shared ledger.
• Explain fundamental concepts in Hyperledger Fabric.
• Describe the elements of a business network, the role of channels, and how the
world state is maintained.
• Develop, test, debug, and deploy chaincode with IBM Blockchain Platform Ex-
tension for Visual Studio Code
• Apply concepts of blockchain security, identity and access control, and data pri-
vacy to blockchain solutions.
• Write applications that interact with a blockchain network.
• Describe patterns, best practices, and reference architectures for integration
from enterprise applications to blockchain networks.
Competencies: C6, C13

Content Unit 1. Blockchain overview

Unit 2. Introduction to chaincode development
Unit 3. Chaincode query methods
Unit 4. Best practices for writing, testing, and debugging chaincode
Unit 5. Identity and access control
Unit 6. Data privacy
Unit 7. Basics of application development
Unit 8. Blockchain integration and advanced application development

Examination forms 100% Continuous evaluation

Study and examination re- 10/20

quirements

Reading list IBM Blockchain Developer – Official course material

AI.5.11 Articial Intelligence of Things (AIoT)

49 Ref ID.230:01
 Module designation AI.5.11 AIoT

Semester(s) in which the mod- S5

ule is taught

Person responsible for the mod- Pr Wided Chaari

ule (coordinator)

Teaching team Dr Mansour Saber

Language French

Relation to curriculum Optional

Teaching methods lecture, lesson, project, presentation

Workload (incl. contact hours, Total workload: 25h

self-study hours) Contact hours: 15h (lessons)
Private study: 10h

Credit points 1 ECTS

Required and recommended AI.3.1 AI & Machine learning

prerequisites for joining the DAT.1.1 Databases and DBMS
module

Module objectives/intended Key question: what learning outcomes should students attain in the mod-
learning outcomes ule?
E.g. in terms of:
- Knowledge: familiarity with the IoT domain and relatives
technologies (protocols, requirements...)
Skills: Able to design an IoT solution linked to the field and able
to use the data to provide added services using AI.
- Competences: Students master the concepts of IoT and are
able to develop systems using the IA and the capabilities of
IoT (sensors and actuators)

Competencies: C1, C2, C3, C4, C5, C6

Content 1- Presentation of the IoT domain: History, evolution and perspectives

2- Impact and Challenges of installing IoT systems to the classic Infor-
mation systems (opening, external data, privacy…)
3- IoT and AI what is it and for which added value services
4- AIoT real live examples: Smart Building and Smart Cities as domain
example
5- Deep Learning algorithm examples applied to IoT systems in Energy
Management solution
6- A step by step design system solution for Verticla IoT solution (from
hardware to API and data retreivement).

Examination forms Mini Project of AiOT solution presented by pair of students

Study and examination require- Identify a real life issue to resolve using IoT solution. Define the different
ments Design Systems Components of the solution. Prensents in details a ma-
chine learning algorithm that answer a required service.

Reading list Springer. (2021). Intelligence of Things: AI-IoT Based Critical-Applica-

tions and Innovations (AIoT Innovation) (1st ed.).

AI.5.17 Quantum Artificial Intelligence

50 Ref ID.230:01
 Module designation AI.5.17

Semester(s) in which the S5

module is taught
Person responsible for the Nesrine Ben Yahia
module
Language English

Relation to curriculum Optional

Teaching methods lesson, seminar, projects, workshops

Workload (incl. contact (Estimated) Total workload: 30h

hours, self-study hours) Contact hours: 15h (10h lessons, 5h labs)
Private study including examination preparation, specified in hours1:
15h

Credit points 1.5

Required and recommended Python

prerequisites for joining the AI.1.1 Formal logic
module AI.3.1 Artificial Intelligence & Machine learning
AI.5.3 Applied deep learning

Module objectives/intended Students have a general overview of quantum computing and its key con-
learning outcomes cepts
Students identify opportunities in machine learning for using quantum ad-
vantages and resources.
Students implement quantum enhanced machine and deep learning mod-
els in Python

Students are able to:

C1. To master in depth the basic sciences, in particular computer science
and mathematics, essential for the design and production of computer ap-
plications.
C8. Master good practices in terms of software development as well as ap-
plicable standards and regulations.
C9. Make complex decisions based on incomplete or limited information.
C13. Become aware of the need to constantly update your knowledge and,
if necessary, to undertake additional studies.

1 When calculating contact time, each contact hour is counted as a full hour because the organisation of the
schedule, moving from room to room, and individual questions to lecturers after the class, all mean that about 60
minutes should be counted.
51 Ref ID.230:01
 Content CHAPTER 1: INTRODUCTION TO QUANTUM COMPUTING
1. What is Quantum computing (QC) and why?
2. QC principles
2.1 Superposition
2.2 Entanglement
2.3 Decoherence
3. Quantum computing fundamentals?
3.1 Qubit
3.2 Quantum sates
3.3 Quantum gates
3.4 Quantum algorithms
4. QC in industry
5. Practical Hands-on experiments for quantum circuits using Cirq

CHAPTER 2: QUANTUM MACHINE LEARNING (QML)

1. Quantum data & models
2. Hybrid quantum-classical models : Parameterized quantum gates, varia-
tional quantum algorithm
3. Quantum ML approches
4. Supervised learning with quantum classifiers
5. Unsuprevisedlearning using quantum resources
5. Parametrized Quantum Circuits for Reinforcement Learning

CHAPTER 3: QUANTUM MACHINE LEARNING (QML) TOOLS

1. GOOGLE Quantum AI
2. IBM Quantum
3. Amazon Braket
4. Microsoft Azure Quantum
5. Practical Hands-on session

CHAPTER 4: QUANTUM NEURAL NETWORKS

1. Quntum deep learning
2. Quantum CNN
3. Quantum generative adversarial networks
4. Practical Hands-on session using TensorFlow Quantum (TFQ)
Examination forms 100% continuous assessment (project)

Study and examination re- 10/20

quirements
Rietsche, R., Dremel, C., Bosch, S. et al. Quantum computing. Electron
Reading list
Markets 32, 2525–2536 (2022). https://doi.org/10.1007/s12525-022-00570-
y
Biamonte, J., Wittek, P., Pancotti, N. et al. Quantum machine learning. Na-
ture 549, 195–202 (2017). https://doi.org/10.1038/nature23474
Schuld, M., Sinayskiy, I., Petruccione, F. An introduction to quantum ma-
chine learning, Contemporary Physics, 56:2, 172-
185, (2015) DOI: 10.1080/00107514.2014.964942

52 Ref ID.230:01