CloudTech’20

The 5th International Conference on Cloud

Computing and Artificial Intelligence:

Technologies and Applications

CloudTech’20

November 24-25, 2020, Morocco www.macc.ma/cloudtech20

IEEE Catalog number: CFP20B66-ART ISBN: 978-1-7281-6175-4

© 2020 IEEE

No part of this book may be reproduced or transmitted in any form or any means, electronic,

recording or otherwise, without consent from IEEE.

Conference venue: Morocco

Co - Editors

Mohamed Essaaidi / Mostapha Zbakh

ENSIAS, Mohamed V University, Morocco

This conference is co-sponsored by the University Mohammed V of Rabat and ENSIAS in partnership with IEEE.

http://www.macc.ma/cloudtech20

1

PREFACE

This is one of the most successful IEEE technically sponsored events on Cloud computing and Artificial Intelligence: technologies and applications, organized remotely from Rabat, Morocco.

Cloud computing has gained great and increasing attention from academia, government and industry as a new IT infrastructure requiring smaller investments in hardware platform, staff training, or licensing new software tools. Cloud computing can be seen as any subscription-based or pay-per-use service that, extends the Internet existing capabilities.

CloudTech´20 is organized by Higher National School of Computer Science and Systems Analysis - ENSIAS, Mohamed V University of Rabat and Moroccan Association of Cloud Computing in partnership with IEEE Morocco Section.

CloudTech´20 will address topics related to cloud technologies and Artificial Intelligence; architecture, applications and services including distributed computing and data centers, cloud infrastructure and its security, end-user services. This conference will provide a high-level international forum for scientists, researchers, professionals and students who will have the opportunity to present and to attend talks and papers on the state-of-the-art research results addresing new challenges, and discussing the new trends in Cloud Technology and Artificial Intelligence, technologies and applications.

CloudTech’20 received 160 submissions from 22 countries, which is really a great challenge we have won with such a level of submissions and international participations. From these, the Technical Program Committee selected 64 papers based on their originality, innovative relevance, and clarity of presentation, to be presented and included in this volume of the conference proceedings. CloudTech´20 technical program includes besides the accepted papers, 8 keynote talks covering different hot topics related with Cloud Computing and Artificial Intelligence: technologies, applications and services. These talks’ abstracts and short Biographies of their speakers are published in this proceeding.

We seize this opportunity to express our deepest thanks and gratitude to all the TPC chairs and members for the wonderful reviewing process which enabled us to come up with a high quality technical program.

M. Essaaidi and M. Zbakh

2

HONORARY CHAIRS

Mohammed RHACHI, President of Mohammed V University, Rabat, Morocco

GENERAL CHAIRS

General Co-Chairs

An Breaken, Sidi Ahmed Mahmoudi VUB, Brussels, Belgium UMONS, Mons, Belgium

KEYNOTES

Chunming Rong, University of Stavanger, Norway

Maria Fasli, University of Essex, UK

Panda Dhabaleswar, The Ohio State University, USA

Helen Karatza, Aristotle University of Thessaloniki, Greece

Othmane Bouhali, Texas A&M University, Qatar

Abdelmounaam Rezgui, Illinois State University, USA

TPC CHAIRS

Pierre Manneback, University of Mons, Mons, Belgium

Claude Tadonki, Mines ParisTech Paris, France

Mohamed ESSAAIDI National Higher School of IT (ENSIAS), Professor

IEEE Morocco Section Chair Rabat, Morocco essaaidi@ieee.org

+ 212 (0) 661 725 992

Pierre Manneback UMONS, Mons, Belgium

Abdellah Touhafi

VUB, Brussels, Belgium

Claude TADONKI MINES ParisTech, France

Mostafa Daoudi,

Mohammed I University, Oujda, Morocco

Mostapha ZBAKH National Higher School of IT (ENSIAS),

Professor Rabat, Morocco

m.zbakh@um5s.net.ma + 212 (0) 666 904 940

http://www.chunming.net/http://www.mines-paristech.fr/Accueil/mailto:essaaidi@ieee.org

3

ORGANIZING COMMITTEE

Abdellatif Afia

Ali Choukri,

Mohamed El Ghmary

Brahim Flilih

Fatima-Zohra Mhada

Hasnae Alaoui Lamrani

Rim Doukha

Abderrahmane Ez-zahout

Mahmoud Hamlaoui

Ali Ouacha

Laila Bouhouch

Abdellatif El Ghazi

Bouchaib Ferrahi

Najlae Kasmi

Hicham Touil

TECHNICAL PROGRAM COMMITTEE Karthik Gomadam, Accenture Technology Labs - Silicon Valley, USA

Marcel Kunze, Karlsruhe Institute of Technology (KIT), Germany

Pierre Manneback, Faculty of Engineering, University of Mons, Belgium

Abdulrahman A. Azab, University of Stavanger, Norway

Alexander Lazovik, The University of Groningen (RuG), Netherlands

Amac M. Guvensan, Yildiz Technical University, Turkey

Andy Ju An Wang, Southern Illinois University, USA

Bilal Kashif, COMSATS Institute of Information Technology, Pakistan

Bouchra Asri, ENSIAS, Rabat, Morocco

Chen Gao, Tufts University, USA

Ching-Hsien Hsu, Chung Hua University, Taiwan

Claude Tadonki, Mines ParisTech, France

Claudio Fiandrino, University of Luxembourg, Luxemburg

Claudio Rossi, Universidad Politécnica de Madrid - CSIC, Madrid, Spain

Dan Grigoras, UCC, Ireland

Dana Petcu, West University of Timisoara, Romania

Daniel Dïaz-Sanchez, Universidad Carlos III de Madrid, Spain

Deepal Jayasinghe, Microsoft, USA

Dongfang Zhao, Illinois Institute of Technology, Chicago IL, USA

Dzmitry Kliazovich, University of Luxembourg, Luxembourg

Eddy Caron, ENS Lyon, Inria CNRS, France

Fawaz Saleem Bokhari, PUCIT - University of the Punjab, Pakistan

Frederico Alvares, University of Nantes, France

Gianluigi Zavattaro, University of Bologna, Italy

Grégoire Danoy, University of Luxembourg, Luxembourg

Hai Jin, School of Computer Science and Technology, Huazhong University of Science and

Technology, Wuhan, China

Hajar Mousannif, Cadi Ayyad University, Marrakesh, Morocco

Hamid Harroud, AUI, Ifrane, Morocco

Helen Karatza, Aristotle University of Thessaloniki, Greece

4

Na Gong, University of South Alabama, USA

Abdelhak Lakhouaja, Faculty of Sciences, Mohammed First University, Oujda, Morocco

Abdelmounaam Rezgui, Illinois State University, Normal, Illinois, USA

Alberto Cano, Virginia Commonwealth University, Richmond, USA

Mawloud Omar, Research Unity LaMOS Faculty of Exact Sciences, University of Bejaia, Algeria

Hassina Aliane, University of Sciences and Technology, Bejaia, Algeria

Héctor Menéndez, University College, London, United Kingdom

Ismael Bouassida Rodriguez, Faculty of Computer Science, Complutense University, Madrid, Spain

Jung-Chun Liu, Tunghai university, Taiwan

Juraj Machaj, University of Zilina, Slovakia

Kun Ma, Institute of Technological Sciences, Wuhan University, Wuhan, China

Hanan El Bakkali, ENSIAS, Mohammed V University, Rabat, Morocco

Bouchaib Ferrahi, Faculty of Sciences, Abdelmalek Essaadi University, Tetouan, Morocco

Ningfang Mi, Northeastern University, Boston, USA

Imed Kacem, University of Lorraine, France

Irena Bojanova, University of Maryland University College, USA

Ismail Khalil, Institute of Telecooperation Johannes Kepler University Linz, Austria

Iyer Vasanth, New College of Florida, USA

Jaime Lloret, Polytechnic University of Valencia, Spain

Javier Garcia Blas, Carlos III University of Madrid, Spain

Jérémie LEGUAY, Thales Group, Colombes, France

Jiannong Cao, Hong Kong Polytechnic University, Hong Kong

Jose Maria Alcaraz Calero, University of the West of Scotland, UK

Juan Durillo, University of Innsbruck, Austria

Abdelkrim Haqiq, Hassan 1st University, Settat, Morocco

Kashif Munir, KFUPM, Dhahran, KSA

Ke Wang, Illinois Institute of Technology, Chicago, USA

Hanan El Bakkali, ENSIAS, Mohammed V University, Rabat, Morocco

Khalid Osman, COMSATS Institute of Information Technology, Pakistan

Laurent Lefevre, INRIA, France

Omar Alfandi, University of Goettingen, Germany

Leonel Sousa, IST, Technical University of Lisbon, Portugal

Lukas Kencl, Czech Technical University in Prague, Czech Republic

Abdellatif Afia, ENSIAS, Mohammed V University, Rabat, Morocco

Marco Aiello, University of Groningen, Netherlands

Masaru Kitsuregawa, University of Tokyo, Japan

Mohammed Boulmalf, IUR, Rabat, Morocco

Mohammed Essaaidi, IEEE, ENSIAS, Mohammed V University, Rabat, Morocco

Mohand Mezmaz, Faculty of Engineering, University of Mons, Belgium

Mohsine Eleuldj, EMI, Mohammed V University, Rabat, Morocco

5

Mostafa Daoudi, Mohammed I University, Oujda, Morocco

Mahdi Zargayouna, Université Paris Est, Ifsttar, France

Mahmoud Nassar, ENSIAS, Mohammed V University, Rabat, Morocco

Ali Ouacha, Faculty of Sciences, Mohammed V University, Rabat, Morocco

Amal Alhosban, University of Michigan-Flint, Flint, Michigan, USA

An Braeken, Vrije Universiteit Brussel (VUB), Brussels, Belgium

Meriem Thabet, University of Abdelhamid Mehri Constantine 2, Constantine, Algeria

Muhammad Younas, Oxford Brookes University

Mustafa Rafique, Rochester Institute of Technology, New York, USA

Okba Kazar, University of Biskra, Biskra, Algeria

Pardeep Kumar, University of Oxford

Paulo Quaresma, University of Evora, Portugal

Qanita Bani Baker, University of science and technology, jordan

Saqib Munawwar, Nazeer Hussain University, Karachi, Pakistan

Driss Olfa Belkahla, University of Manouba, Tunisia

Mostafa Bellafkih, INPT, Mohammed V University, Rabat, Morocco

Mostafa Ezziyyani, Faculty of Sciences and Techniques, Abdelmalek Essaâdi University, Tangier,

Morocco

Amine Roukh, University of Mons, Belgium

Nouredine Melab, LIFL, Lille, France

Paolo Bellavista, University of Bologna, Italy

Pascal Bouvry, University of Luxembourg, Luxembourg

Pascal Lorenz, University of Haute Alsace, France

Abdellatif El Ghazi, IUR, Rabat, Morocco

Rahal Romadi, ENSIAS, Mohammed V University, Rabat, Morocco

Rajdeep Bhowmik, Cisco Systems, Inc., USA

Saif Malik, COMSATS Institute of Information Technology, Pakistan

Abdellah Touhafi, VUB, Brussels, Belgium

Sajjad Ahmad Madani, COMSATS Institute of Information Technology, Pakistan

Samee U. Khan, North Dakota State University, Fargo, ND, USA

Samir Youcef, University of Lorraine LORIA-INRIA-UMR, Vandoeuvre-les-Nancy, France

Santiago Iturriaga, Universidad de la República, Uruguay

Sherif Khattab, Faculty of Computers and Information, Cairo University, Egypt

Sidi Ahmed Mahmoudi, Faculty of Engineering, University of Mons, Belgium

Song Fu, University of North Texas, USA

Stefano Giordano, University of Pisa, Italy

Steve Crago, Southern California University, USA

Tarek El-Ghazawi, The George Washington University, USA

Thomas Magedanz, TU Berlin

Fraunhofer FOKUS, Germany

6

Tomas Margalef, Autonomous University of Barcelona Barcelona, Spain

Karim Baina, ENSIAS, Mohammed V University, Rabat, Morocco

Tonglin Li, Illinois Institute of Technology, Chicago, USA

Toyotaro Suzumura, IBM Tokyo Research Laboratory, Japan

Veena B. Mendiratta, Bell Labs, Alcatel-Lucent, USA

Wei Jie, West London University, UK

Weiwei Chen, University of Southern California, USA

Yeh-Ching Chung, National Tsing Hua University, Taiwan

Atilla Elci, Aksaray University, Isparta, Turkey

Azahara Camacho, Universidad Complutense de Madrid, Spain

Chau Yuen, Singapore University of Technology and Design, Singapore

Raluca Maria Aileni, Politehnica University of Bucharest, Faculty of Electronics Telecommunication

and Information Technology, Romania

Saïd Mahmoudi, University of Mons - FPMs Mons, Belgium

Said Tazi, University of Pau et Pays de l’Adour, Anglet, FRANCE

Stanimir Stoyanov, University of Plovdiv "Paisii Hilendarski", Plovdiv, Bulgaria

Tzung-Pei Hong, National University of Kaohsiung, Kaohsiung, Taiwan

Mohammed Benjelloun, Faculte Polytehnique Mons, UMONS, Mons, Belgium

Zafeirios Papazachos, Aristotle University of Thessaloniki, Thessaloniki, Greece

Meriem Belguidoum, University of Constantine 2, Algeria

Ahmed Zerouali, University of Mons, Mons, Belgium

Nicola Capuano, University of Salerno, Fisciano, Italy

7

Contents

Keynotes

Secure Blockchain Ecosystem – The Network is my Computer _____________________________ 13

Chunming Rong, University of Stavanger, Norway

Cloud and Fog Computing Resource Allocation and Scheduling of Complex Real-Time

Applications______________________________________________________________________14

Helen Karatza, Aristotle University of Thessaloniki, Greece

Cloud-based Intelligent Remote Patient Monitoring_____________________________________ 15

Abdelmounaam Rezgui, Illinois State University, USA

Data Science and AI: Trends, Myths and Challenges______________________________________16

Maria Fasli, University of ESSEX, UK

Artificial intelligence and modeling activities in imaging and medical physics application_______ 17

Othmane Bouhali, Texas A&M University, Qatar

HPC and AI Meet Cloud: Opportunities and Challenges in Designing High-Performance MPI and Deep

Learning Libraries for Public Cloud____________________________________________________18

Panda Dhabaleswar, The Ohio State University, USA

8

Session 1

Security and Privacy

Efficient Mobile User Authentication Service Based on User's Memories with Privacy Preservation

and User Untraceability (ID-14) ______________________________________________________19

An Braeken and Abdellah Touhafi

Secure, Efficient and Dynamic Data Search using Searchable Symmetric Encryption (ID-74) ______20

Mohammed Saad Shaikh, Jatna Bavishi and Reema Patel

Cloud computing security classifications and taxonomies: a comprehensive study and comparison

(ID-78) __________________________________________________________________________21

Najat Tissir, Said El Kafhali and Nourddine Aboutabit

ESSMAR: Edge Supportive Secure Mobile Augmented Reality Architecture for Healthcare (ID-15)

________________________________________________________________________________22

An Braeken, Pawani Porambage, Amirthan Puvaneswaran and Madhusanka Liyanage

Towards Smart Blockchain-Based Framework for Privacy Preserving Data sharing in Smart Cities (ID-

39) _____________________________________________________________________________23

Driss El Majdoubi, Hanan El Bakkali and Souad Sadki

Session 2

Smart cities/Farming and Mobile Technologies

Big Data Storage and Analysis for Smart Farming (ID-58) _________________________________24

Fabrice Nolack Fote, Saïd Mahmoudi, Sidi Ahmed Mahmoudi and Amine Roukh

Performance Evaluation of Newly Implemented Resource Blocks (RB) Allocation Schemes on NS-3

simulator for mMTC 5G NR (New Radio) Femtocells (ID-70) ______________________________ 25

Ismail Angri, Abdellah Najid and Mohammed Mahfoudi

Improvements of Centroid Localization Algorithm for Wireless Sensor Networks (ID-96) ______ 26

Abdelali Hadir, Khalid Zine-Dine and Mohamed Bakhouya

An overview of recommender systems in the context of smart cities (ID-64) ________________ 27

Rabie Madani, Abderrahmane Ez-Zahout and Abdellah Idrissi

A re-engineering approach for extension of the Tourist Guide Knowledge Base (ID-40) _______ 28

Asya Stoyanova-Doycheva, Todorka Glushkova, Nevena Moraliyska and Emil Doychev

9

Session 3

Cloud services

NFaaS: A new cloud service for water distribution network monitoring (ID-11)________________29

Rayane El Sibai, Jacques Bou Abdo, Chady Abou Jaoude, Jacques Demerjian, Yousra Chabchoub and

Raja Chiky

An overall statistical analysis of AI tools deployed in Cloud computing and networking systems (ID-

10)______________________________________________________________________________30

Ikhlasse Hamzaoui, Benjamin Duthil, Vincent Courboulay and Hicham Medromi

Multi-task Offloading and Computational Resources Management in a Mobile Edge Computing

Environment (ID-102) ______________________________________________________________31

Mohamed El Ghmary, Youssef Hmimz, Tarik Chanyour and Mohamed Ouçamah Cherkaoui Malki

Deployment of Containerized Deep Learning Applications in the cloud (ID-51) ________________32

Rim Doukha, Sidi Ahmed Mahmoudi, Mostapha Zbakh and Pierre Manneback

Autonomous Provisioning of Preemptive Instances in Google Cloud for Maximum Performance Per

Dollar (ID-87) _____________________________________________________________________33

Jonas Kruger Svensson, Hårek Haugerud and Anis Yazidi

Session 4

Macine Learning

A Review of Credit Card Fraud Detection Using Machine Learning Techniques (ID-53) __________34

Nadia Boutaher, Amina Elomri, Noreddine Abghour, Khalid Moussaid and Mohamed Rida

Machine Learning for Anomaly Detection. Performance Study considering Anomaly Distribution in

an Unbalanced Dataset (ID-37) ______________________________________________________35

Salma El Hajjami, Jamal Malki, Mohammed Berrada and Bouziane Fourka

Novel Convex Polyhedron Classifier for Sentiment Analysis (ID-73) _________________________36

Soufiane El Mrabti, Mohamed Lazaar, Mohammed Al Achhab and Hicham Omara

A Recommendation approach based on Correlation and Co-occurrence within social learning

network (ID-7) ___________________________________________________________________37

Sonia Souabi, Asmaâ Retbi, Mohammed Khalidi Idrissi and Samir Bennani

Graph-based Model for Negative e-WOM Influence in Social Media (ID-66) _________________ 38

Abderraouf Dembri and Mohamed Gharzouli

10

Session 5

Big data

Big Data Architectures Benchmark for Forecasting Electricity Consumption (ID-52) ____________39

Houda Daki, Asmaa El Hannani and Hassan Ouahmane

A Big Data Placement Strategy in Geographically Distributed Datacenters (ID-59) _____________40

Laila Bouhouch, Mostapha Zbakh and Claude Tadonki

New Approach for implementing big Datamart using NoSQL Key-Value stores (ID-41) __________41

Abdelhak Khalil and Mustapha Belaissaoui

Big data for sustainability: A qualitative analysis (ID-6) ___________________________________42

Wail El Hilali, Abdellah El Manouar and Mohammed Abdou Janati Idrissi

Bloom filter and its variants for the optimization of MapReduce’s algorithms: A review (ID-55) __43

Fatimazahra Ezzaki, Noreddine Abghour, Amina Elomri, Khalid Moussaid and Mohamed Rida

Session 6

Deep Learning

Deep Learning and Tensorflow for Tracking People’s Movements in a Video (ID-97) ___________44

Bornia Jemai, Ali Frihida and Olivier Debauche

SAHAR-LSTM: An enhanced Model for Sentiment Analysis of Hotels’Arabic Reviews based on LSTM

(ID-21) __________________________________________________________________________45

Manal Nejjari and Abdelouafi Meziane

Medical Image Registration via Similarity Measure based on Convolutional Neural Network

(ID-2)___________________________________________________________________________ 46

Li Dong, Yongzheng Lin and Yishen Pang

Survey of Automated Deep Neural Networks Implementation on FPGAs (ID-42) ______________47

El Hadrami Cheikh Tourad

Machine learning and datamining methods for hybrid Iot intrusion detection (ID-27) _________ 48

Rachid Ait Moulay, Abdellatif El Ghazi and Zine El Abidine Abdelali

11

Session 7

High-performance computing

High-performance computing under availability constraints to solve dense triangular system

(ID-61)___________________________________________________________________________49

Mounira Belmabrouk and Mounir Marrakchi

Run Time Optimization using a novel implementation of Parallel-PSO for real-word applications (ID-

54) _____________________________________________________________________________50

Amine Chraibi, Said Ben Alla, Abdellah Touhafi and Abdellah Ezzati

Evaluation of NDVI and NDWI parameters in CPU-GPU Heterogeneous Platforms based CUDA (ID-

22) _____________________________________________________________________________51

Fatima Zahra Guerrouj, Rachid Latif and Amine Saddik

Better Edges not Bigger Graphs: An Interaction-Driven Friending Algorithm for the Next-Generation

Social Networks (ID-95) ____________________________________________________________52

Aadil Alshammari and Abdelmounaam Rezgui

A Cross-Layered Interference in Multichannel MAC of VANET (ID-77) _______________________53

Fadlallah Chbib, Khoukhi, Fahs, Haydar and Khatoun

Session 8

Image processing using Artificial Intelligence

Towards Breast Cancer Response Prediction using Artificial Intelligence and Radiomics

(ID-75) __________________________________________________________________________54

Yassine Amkrane, Mohammed El Adoui and Mohammed Benjelloun

An Image Processing Player Acquisition and Tracking System for E-sports (ID-88) _____________55

Nicolaas Luwes and Joaquim Vieira

Density Based Support Vector Machine (DBSVM) (ID-100) ________________________________56

Karim Elmoutaouakil, Abdellatif El Ouissari, Abdellah Touhafi and Nabil Aharrane

A Business Process Modeling Notation Extension for Real Time Handling - Application to Novel

Coronavirus (2019-nCoV) management process (ID-29) __________________________________57

Asma Ouarhim, Jihane Lakhrouit and Karim Baina

A Model of a Biometric Recognition System Based On the Hough Transform of Libor Masek and 1-D

Log-Gabor Filter (ID-81) ___________________________________________________________58

Djalal Rafik Hammou, Sidi Ahmed Mahmoudi, Reda Adjoudj and Boubaker Mechab

12

Session 9

IoT

Internet of Things learning: a practical case for smart building automation (ID-33) ____________59

Olivier Debauche, Saïd Mahmoudi and Yahya Moussaoui

Open Phytotron: A New IoT Device for Home Gardening (ID-34) ___________________________60

Rachida Ait Abdelouahid, Olivier Debauche, Saïd Mahmoudi, Abdelaziz Marzak, Pierre Manneback

and Frédéric Lebeau

An IoT data logging instrument for monitor and early efficiency loss detection at a photovoltaic

generation plant (ID-90) ____________________________________________________________61

Nicolaas Luwes and Burger Lubbe

Formal Modeling and Validation of Micro Smart Grids Based on IoT ReDy Architecture

(ID-80) __________________________________________________________________________62

Kaoutar Hafdi and Abderahman Kriouile

The key layers of IoT architecture (ID-98) ______________________________________________63

Alae-Eddine Bouaouad, Adil Cherradi, Saliha Assoul and Nissrine Souissi

Modelling and simulation of photovoltaic modules using combined analytical - numerical methods

(ID-72) (Poster)___________________________________________________________________64

Khalid Chennofi and Mohammed Ferfra

Session 10

Computational Intelligence

A New Recurrent Neural Network Fuzzy Mean Square Clustering Method (ID-99) _____________65 Abdellah Touhafi and Karim El Moutaouakil

Learning Analytics based on Bayesian Optimization of Support Vector Machines with Application to Student Success Prediction in Mathematics Course (ID-13) ________________________________66 Salim Lahmiri, Raafat G. Saade, Danielle Morin and Fassil Nebebe

Block Sizes Control for an Efficient Real Time Record Linkage (ID-48) ________________________67

Hamid Naceur Benkhaled, Djamel Berrabah and Faouzi Boufares

An Artificial Neural Networks Based Ensemble System to Forecast Bitcoin Daily Trading Volume (ID-

12) _____________________________________________________________________________68

Salim Lahmiri, Raafat G. Saade, Danielle Morin and Fassil Nebebe

A distributed large graph coloring algorithm on Giraph (ID-18) ____________________________69

Assia Brighen, Hachem Slimani Slimani, Abdelmounaam Rezgui Rezgui, Hamamache Kheddouci

Kheddouci and Assia Brighen

Quality Approach to Analyze the Causes of Failures in MOOC (ID-67) (Poster)________________70

Soukaina Sraidi, Smaili El Miloud, Salma Azzouzi and My El Hassan Charaf

13

Secure Blockchain Ecosystem – The Network is my Computer

Chunming Rong

University of Stavanger, Norway

Abstract:

Blockchain and other distributed ledger technologies (DLTs), through recent development, have not

only enabled simple transactions, but also complex computation on a network where parties are

geographically distant or have no particular trust in each other to interact and exchange value and

information on a fully distributed basis with fewer to non-existent central intermediaries. There are

obvious advantages associated with blockchain in areas such as, decentralization, disintermediation,

replication, timestamping, immutability, digital signatures, automation and smart contracts,

transparency and censorship resistance, especially in relation with the financial sector. However, there

are also noticeable challenges, for instances, scalability and performance, energy consumption,

security and privacy. Many popular applications are cloud related. These advances are now not just

limited to the financial sector, but also new internet applications can harness these building blocks to

empower users to take control of their online footprint, such as in healthcare, social media and other

digital services. We are building an open source hub for a secure and source-controlled blockchain

ecosystem, as the gathering place for academic researchers, practitioners and business innovators

alike, where they may meet and work together online to embrace, promote and enhance blockchain

technologies and their applications. This will bring us to new platform, new business model and new

ecosystem.

Biography:

Prof. Chunming Rong is the chair of IEEE CS STC on Blockchain, and served as co-chair of IEEE

Blockchain in 2018. He is also chair of IEEE Cloud Computing. He works as the head of the Center for

IP-based Service Innovation (CIPSI) at the University of Stavanger (UiS) and also as adjunct Senior

Scientist leading Big-Data Initiative at NORCE. He is also co-founder and CTO of two start-ups bitYoga

and Dataunitor in Norway, both received EU Seal of Excellence Award in 2018. He was the vice

president of CSA Norway Chapter (2016-2017). His research work focuses on data science, cloud

computing, security and privacy. He is an IEEE senior member and is honoured as member of the

Norwegian Academy of Technological Sciences (NTVA) since 2011. He has extensive contact network

and projects in both the industry and academic. He is also founder and Steering Chair of IEEE CloudCom

conference and workshop series. He is the steering chair and associate editor of the IEEE Transactions

on Cloud Computing (TCC), and co-Editors-in-Chief of the Journal of Cloud Computing (ISSN: 2192-

113X) by Springer. Prof. Rong has extensive experience in managing large-scale R&D projects funded

by both industry and funding agencies, both in Norway and EU.

14

Cloud and Fog Computing Resource Allocation and Scheduling of Complex Real-Time

Applications

Helen Karatza

Aristotle University of Thessaloniki, Greece

Abstract:

Cloud computing provides a virtually unlimited pool of resources to the end-users.Due to the popularity of the cloud computing paradigm, new complex computationally intensive applications have emerged. However, many important issues such as resource allocation, job scheduling, cost, and quality of service must be addressed in order to exploit cloud’s full potential. Complex application scheduling in the cloud is a challenging problem. Furthermore, it is more difficult to schedule complex real-time applications, considering deadlines along with energy conservation. Therefore, energy efficient scheduling strategies are required ensuring timeliness. Due to the increasing number of Internet of Things (IoT) applications, fog computing has emerged as a new paradigm, beyond cloud computing. The fog extends the cloud to the network edge in order to reduce data transmission latency. Real-time applications should be appropriately assigned to resources in both the fog and cloud layers, taking into account the communication and computational characteristics of each application. In this keynote talk we will focus towards challenges on resource allocation and scheduling complex real-time applications in cloud and fog systems and we will provide future research directions in this emerging research area. Biography:

Prof. Helen Karatza is a Professor Emeritus in the Department of Informatics at the Aristotle University of Thessaloniki, Greece, where she teaches courses in the postgraduate and undergraduate level. Dr. Karatza's research interests include Fog and Cloud Computing, Energy Efficiency in Large Scale Distributed Systems, Resource Allocation and Scheduling and Real-time Distributed Systems. Dr. Karatza has authored or co-authored 225 technical papers and book chapters including five papers that earned best paper awards at international conferences. She is senior member of IEEE, ACM and SCS, and she served as an elected member of the Board of Directors at Large of the Society for Modeling and Simulation International. She served as Chair and Keynote Speaker in International Conferences. Dr. Karatza is the Editor-in-Chief of the Elsevier Journal “Simulation Modeling Practice and Theory” and Senior Associate Editor of the “Journal of Systems and Software” of Elsevier. She was Editor-in-Chief of “Simulation Transactions of The Society for Modeling and Simulation International” and Associate Editor of “ACM Transactions on Modeling and Computer Simulation”. She served as Guest Editor of Special Issues in International Journals. More info about her activities/publications can be found in http://agent.csd.auth.gr/~karatza/

15

Cloud-based Intelligent Remote Patient Monitoring

Abdelmounaam Rezgui

Illinois State University, USA

Abstract:

The worldwide population growth has led to an unprecedented need for new sustainable approaches and technologies to provide health care. A health care system has three major stakeholders: (i) people, (ii) medical staff, and (iii) health administrators. People may want to monitor their health status while they conduct their daily activities. For this, they need solutions that perform continuous health monitoring. Health professionals (physicians, nurses, ICU specialists, etc.) need advanced accurate predictive capabilities to help make better decisions regarding their patients. Administrators at hospitals look for solutions that enable them to reduce waste and wait time, and increase service efficiency and productivity. In particular, administrators face significant challenges in terms of bed management, patient flow, and effective provision of medical supplies. In this talk, we present our recent research aimed at addressing the aforementioned challenges. To validate our research, we developed a cloud-based, intelligent remote patient monitoring system (called CIRPM) that can be used by people, medical professionals, and health administrators to improve the quality of health services.

Biography:

Prof. Abdelmounaam Rezgui is a faculty member in the School of Information Technology at Illinois State University. He received his PhD in Computer Science from Virginia Tech. His research interests include: networking, cloud computing, social networks, and big data. Abdelmounaam authored or coauthored over 90 papers in top journals and conferences including IEEE TBD, IEEE TKDE, ACM TOIT, IEEE TPDS, IEEE Internet Computing, IEEE Security and Privacy, IEEE ICDE, IEEE IC2E, and IEEE CLOUD. His research has been funded by NASA and Microsoft. He regularly serves on the program committees of several major conferences including IEEE Big Data, IEEE BigDataSE, IEEE Globecom, IEEE CloudNet, IEEE LCN, and IoTBDS. He has been a track chair, keynote speaker, or tutorial presenter at several international conferences. He also is on the editorial board of several journals including Springer’s Big Data Analytics. Research conducted by Abdelmounaam and his students has been invited for presentation at very selective events organized by major organizations including Google, NSF, and Siemens.

16

Data Science and AI: Trends, Myths and Challenges

Maria Fasli

University of ESSEX, UK

Abstract:

The availability of cheap computational devices and the growth of the WWW and applications has led to the proliferation in the amount of data - both structured and unstructured. In turn, increased computational power, cheap storage and data proliferation have led to the development of new techniques in Data Science and unprecedented progress in Artificial Intelligence (AI). From powering recommendation engines to autonomous vehicles, AI has seen a reinvigoration in interest while industry and academia have been investing heavily to remain competitive and take advantage of the power of AI. But is AI a panacea and how much progress has really been made in the field? This talk will explore the trends, myths but also challenges that AI poses providing a critical perspective of the field and where the pitfalls may lie. Biography:

Prof. Maria Fasli is a Professor, Director of the ESRC Business and Local Government Data Research

Centre and Director of the Institute for Analytics and Data Science at the University of Essex. Educated

in Greece (BSc Informatics 1996) and the UK (PhD Computer Science 2000), she has held positions at

Essex since 1999. In 2005, she was awarded a National Teaching Fellowship by the HEA UK for her

innovations in education. Between 2009-2014, she was Head of School of Computer Science and

Electronic Engineering. In 2016, she was awarded a UNESCO Chair in Analytics and Data Science. Her

research interests lie in artificial intelligence techniques for complex systems and analysing and

modelling structured/unstructured data. Her research has been funded by Research Councils and

other organisations and she has worked with a range of companies in data analytics related projects.

She has published over 130 papers in the field of AI and data science and has delivered keynote talks

at conferences.

17

Artificial intelligence and modeling activities in imaging and medical physics application

Othmane Bouhali

Texas A&M University, Qatar

Abstract

Artificial intelligence (AI) has captured the attention of the scientific and business communities for

decades. With its associated technologies, such as machine learning (ML) and deep Learning (DL), AI

has provided very efficient tools and platforms that help reducing cost, improving efficiency, and

increasing competitiveness. In this talk I will give an overview of our recent work on applying AI

techniques in several fields of health sectors including imaging and medical physics applications. Our I

will also review our activities in modeling medical and imaging devices.

Biography:

Prof. Othmane Bouhali is research professor and director of Research Computing at Texas A&M

University at Qatar. He is the founder and director of the TAMUQ Advanced Scientific Computing group

(TASC), a specialized group in computational modeling and simulation. His main work is on

computational physics, Monte Carlo simulation and high performance computing. Before he joined

TAMUQ, he was given the responsibility of the computing group at the High Energy Physics Institute

of Brussels. His research focused on the scientific grid computing, in particular the optimization and

implementation of the grid infrastructure for intensive scientific applications. Currently he is involved

in several projects on medical physics, 3D scientific visualization, distributed computing and image

analysis techniques.

18

HPC and AI Meet Cloud: Opportunities and Challenges in Designing High-Performance MPI

and Deep Learning Libraries for Public Cloud

Panda Dhabaleswar

Ohio State University, USA

Abstract:

Significant growth has been witnessed during the last decade in dedicated HPC clusters with multi-/many-core processors, accelerators, and high-performance interconnects (such as InfiniBand, Omni-Path, iWARP, RoCE and EFA). Public cloud providers like Amazon-AWS and Microsoft-Azure are gradually aiming to provide solutions for High-Performance Computing (HPC) and Deep Learning (DL) users and applications. Microsoft-Azure has recently moved to bare-metal environment with InfiniBand networking technology. Amazon-AWS has moved to using a new network adapter, Elastic Fabric Adapter (EFA), for its HPC instances. These new networks and platforms provide a new set of opportunities and challenges to design high-performance MPI and DL libraries for these emerging public cloud platforms. The MVAPICH MPI library (http://mvapich.cse.ohio-state.edu) has been a significant middleware component in the HPC domain for the last 18 years. Recently, the MVAPICH library is also enabling high-performance and scalable DL applications. The MVAPICH team members are currently working closely with Amazon-AWS and Microsoft-Azure to design and deploy high-performance and scalable MPI library for their respective could infrastructures. This talk will provide details on these designs, their performance (micro-benchmarks and applications) and comparison with other MPI libraries. A set of results comparing the performance numbers of HPC and AI applications on cloud environments with dedicated (non-virtualized) clusters will be presented.

Biography:

Prof. Panda Dhabaleswar is a Professor and University Distinguished Scholar of Computer Science and Engineering at the Ohio State University. He has published over 450 papers in the area of high-end computing and networking. The MVAPICH2 (High Performance MPI and PGAS over InfiniBand, Omni-Path, iWARP and RoCE) libraries, designed and developed by his research group (http://mvapich.cse.ohio-state.edu), are currently being used by more than 3,025 organizations worldwide (in 89 countries). More than 589,000 downloads of this software have taken place from the project's site. This software is empowering several InfiniBand clusters (including the 3rd, 5th, 8th, 15th, 16th, 19th, and 31st ranked ones) in the TOP500 list. The RDMA packages for Apache Spark, Apache Hadoop and Memcached together with OSU HiBD benchmarks from his group (http://hibd.cse.ohio-state.edu) are also publicly available. These libraries are currently being used by more than 315 organizations in 35 countries. More than 31,100 downloads of these libraries have taken place. High-performance and scalable versions of the Caffe and TensorFlow framework are available from https://hidl.cse.ohio-state.edu. Prof. Panda is an IEEE Fellow. More details about Prof. Panda are available at http://www.cse.ohio-state.edu/~panda.

19

Efficient Mobile User Authentication Service Based on User's Memories with Privacy

Preservation and User Untraceability

An Braeken, Abdellah Touhafi

INDI, Vrije Universiteit Brussel (VUB), Brussels, Belgium

Abstract— Security questions and answers for authentication are a common approach to enable the user to reset forgotten passwords. Moreover, they are also sometimes used as alternative for the classical username-password system, which fails in offering a good balance between user friendliness and security as long and complex passwords are required. However, in order to guarantee the privacy of the user as imposed by the new General Data Protection Regulation (GDPR), it should be impossible to derive the answer of the user by any other entity, including the server provider or the server managing the authentication. In this paper, we present an efficient mobile based security mechanism to realise this goal. The proposed scheme can be applied on top of any type of question-answer based authentication system. In addition, our solution also offers anonymity and untraceability of the user, such that no activity patterns can be drawn by simply eavesdropping on the communication channel to the service provider or the authentication server. We show that our proposed mechanism not only offers more security features compared to related work, but it is also significantly faster, in particular at the side of the user. Keywords— Authentication, Anonymity, Physical Unclonable Function.

ID-14

Security and Privacy

20

Secure, Efficient and Dynamic Data Search using Searchable Symmetric Encryption

Mohammed Saad Shaikh, Jatna Bavishi and Reema Patel

Department of Computer Science and Engineering

Pandit Deendayal Petroleum

University Gandhinagar, India

shaikhsaad133@gmail.com; jatnabavishi@gmail.com; reema.mtech@gmail.com

Abstract— Fog computing, which works complementary to cloud computing, is being developed to overcome the issue of cloud computing such as latency in the cases where the data is to be retrieved immediately. But, along with solving the problem of latency, fog computing brings along with it different set of security issues than cloud computing. The storage and processing capabilities of fog computing is limited and hence, the security issues must be solved with these constrained resources. One of the problems faced when the data is stored outside the internal network is loss of confidentiality. For this, the data must be encrypted. But, whenever document needs to be searched, all the related documents must be decrypted first and later the required document is to be fetched. Within this time frame, the document data can be accessed by an unauthorized person. So, in this paper, a searchable symmetric encryption scheme is proposed wherein the authorized members of an organization can search over the encrypted data and retrieve the required document in order to preserve the security and privacy of the data. Also, the searching complexity of the algorithm is much less so that it suitable to fog computing environment.

Keywords— Cloud Computing, Fog Computing, Secure Data Computation, Searchable Encryption, Searchable Symmetric Encryption.

ID-74 Security and Privacy

mailto:shaikhsaad133@gmail.commailto:jatnabavishi@gmail.commailto:reema.mtech@gmail.com

21

Cloud computing security classifications and taxonomies: a comprehensive study and comparison

Najat Tissir

Process Engineering, Computer Science and Mathematics laboratory

National School of Applied Sciences Sultan Moulay Slimane University Khouribga, Morocco

tissir.najat@gmail.com

Said El Kafhali

Hassan First University of Settat,

Faculty of Sciences and Techniques,

Computer, Networks, Mobility and Modeling laboratory: IR2M, 26000-Settat, Morocco

said.elkafhali@uhp.ac.ma

Nourddine Aboutabit

Process Engineering, Computer Science and Mathematics laboratory

National School of Applied Sciences Sultan Moulay Slimane University Khouribga, Morocco

n.aboutabit@usms.ma

Abstract— Cloud Computing is an emerging paradigm that is based on the concept of distributed computing. As with any novel technology, Cloud Computing is subject to security threats, vulnerabilities, and attacks. Recently, various classifications and taxonomies have been proposed to describe and organize cloud security issues. Some of them are based on general security factors, such as the CIA triad (confidentiality, integrity, and availability), while others specify cloud security classes. Most of these classes are determined based on the cloud’s characteristics, such as Cloud service models, cloud deployment models, and cloud actors. In this paper, we explore the already existing criteria and dimensions considered in the development of cloud computing security classification/taxonomy. Then, we study and compare their strengths and characteristics. Thereafter, our objective is to provide and develop a comprehensive view of cloud security taxonomy and help researchers to better understand this evolving field.

Keywords— Cloud Computing, Taxonomy, Classification, Threat, Attacks.

ID-78 Security and Privacy

mailto:tissir.najat@gmail.commailto:said.elkafhali@uhp.ac.ma

22

ESSMAR: Edge Supportive Secure Mobile Augmented Reality Architecture for Healthcare

An Braeken

Industrial Engineering INDI, Vrije Universiteit Brussel, Belgium

an.braeken@vub.be

Pawani Porambage, Amirthan Puvaneswaran

Centre for Wireless Communications, University of Oulu, Finland

pawani.porambage@oulu.fi, amirthan.puvaneswaran@student.oulu.fi

Madhusanka Liyanage

School of Computer Science, University College Dublin, Ireland

madhusanka@ucd.ie

Abstract— The recent advances in mobile devices and wireless communication sector transformed Mobile Augmented Reality (MAR) from science fiction to reality. Among the other MAR use cases, the incorporation of this MAR technology in the healthcare sector can elevate the quality of diagnosis and treatment for the patients. However, due to the highly sensitive nature of the data being circulated in this process, it is also highly vulnerable to security threats. In this paper, an edge-based secure architecture is presented for a MAR healthcare application. Based on the ESSMAR architecture, a secure key management scheme is proposed for registration and authentication phases. Then the security of the proposed scheme is validated using formal and informal verification methods.

ID-15 Security and Privacy

mailto:an.braeken@vub.bemailto:amirthan.puvaneswaran@student.oulu.fi

23

Towards Smart Blockchain-Based Framework for Privacy Preserving Data sharing

in Smart Cities

Driss EL MAJDOUBI, Hanan EL BAKKALI, Souad SADKI Smart Systems Laboratory- Rabat IT Center

Mohammed V University in Rabat {Driss.elmajdoubi, h.elbakkali, souad.sadki}@um5s.net.ma

Abstract— Nowadays, the adoption of smart cities worldwide is accelerating the digital transformation of urban environments. Besides, a huge amount of data is being exchanged through the smart devices, networks, cloud infrastructure and Internet of Things (IoT) applications both in private and public sector. This seamless integration of the cyber capabilities of the corresponding devices with the physical world generates new opportunities in many areas but it also raises new challenges in terms of security and privacy due to the diversity of sources and stakeholders, the centralized data management and the resulting lack of trustworthiness and governance. Hence, we introduce “SmartPrivChain” a Smart Blockchain Based Framework for privacy-preserving data sharing in a smart city environment. The proposed approach is different from the existing schemes on many aspects. The data privacy is preserved by combining data access control and data usage auditing measures based on smart contracts. In addition, the proposed solution, is compliant with the main privacy laws and regulation especially the requirements of European Union General Data Protection Regulation (EU GDPR). Lastly, we propose an enhanced Proof of Reputation (PoR) consensus scheme using a multidimensional Trust model. Keywords— Blockchain; Smart cities; Privacy Laws; Consensus Protocol; Proof-of-Trustworthiness; Smart Contract.

ID- 39 Security and Privacy

24

Big Data Storage and Analysis for Smart Farming

Fabrice Nolack Fote, Saïd Mahmoudi, Sidi Ahmed Mahmoudi and Amine Roukh Faculty of Engineering - ILIA / Infortech

University of Mons, Belgium

Abstract— Smart Farming has always been referred to as agriculture, but nowadays, that is no longer the case. Today, Smart farming is made up of Precision Agriculture (PA) and Precision Livestock Farming (PLF). Big Data technologies and algorithms can be relevant for managing and monitoring data related to any farm. Precision livestock farming concerns genetics, animal welfare, animal nutrition, reproduction, species protection and animal health. This paper presents a general overview of Big Data tools that can be applied in a smart farming application. New Technologies are offering many tools used to facilitate the management of data collection, risk minimization, climate change anticipation, secure storage and analysis, etc. The main purpose of Big Data tools is to increase productions in order to offer higher quantities while ensuring higher quality products. However, they remain some issues that need to be accomplished.

Keywords— Smart Farming, Big Data, Precision Agriculture, Precision Livestock Farming.

ID-58 Smart cities/Farming and Mobile Technologies

25

Performance Evaluation of Newly Implemented Resource Blocks (RB) Allocation Schemes

on NS-3 simulator for mMTC 5G NR (New Radio) Femtocells

Ismail Angri, Abdellah Najid Telecommunication Systems, Networks and Services (STRS) laboratory

National Institute of Posts and Telecommunications (INPT) Rabat, Morocco

ismail.angri@gmail.com, najid_abdellah@yahoo.fr

Mohammed Mahfoudi Transmission and Data Processing Laboratory

(LTTI) Superior School of Technology (EST-Fez) Fez, Morocco

mahfoudi.mohammed@gmail.com

Abstract— The new standard of mobile technologies called 5G allows enormous improvements,

comparing to the previous telecommunication network system LTE, in terms of user requirements by

offering different use cases (eMBB, URLLC and mMTC). With the use of the Internet of Things (IoT) by

5G networks, the number of radio devices by each user will drop from 2 to around 7 to 10 devices.

Despite this, the saturation of the system does not arise, thanks to the connected equipment’s high

density, offered by Massive machine type communications (mMTC). A Radio Resource Management

RRM procedure for efficient distribution of available radio resources between those devices is essential

for 5G systems. In this article, we have studied the behavior of scheduling algorithms in a 5G

environment, for a large number of connected objects and for different types of data flows, while

limiting to small cells (Femtocells) with a speed of 3 km/h of the User Equipment (UE). In this objective,

we program in C++ two new scheduling algorithms at the base station gNb, namely Exponential PF

(EXP/PF) and Exponential Rule (EXP-rule), in addition to those already existing (Maximum-Weight

(MW) and Proportional Fair (PF)), using the mmWave model of the famous NS-3 simulator. The

performance comparison of the different 5G scheduler schemes was inspected via two important

parameters, which are the user throughput and the Signal-to-Interference-plus-Noise Ratio (SINR).

Consequently, we have demonstrated that the scheduling algorithms used by LTE networks can be

implemented at the 5G gNB level. The results of our simulations have shown that the EXP-rule

algorithm provides the best SINR and DataRate values for voice, video and data streams.

Keywords— 5G; New Radio; NR numerology; Scheduling; Radio Resource Management; NS-3; eMBB; mMTC; URLLC.

ID-70

Smart cities/Farming and Mobile Technologies

mailto:najid_abdellah@yahoo.fr

26

Improvements of Centroid Localization Algorithm for Wireless Sensor Networks

Abdelali Hadir Hassan II University, National School of Business and Management, Casablanca, Morocco

a.hadir@encgcasa.ma

Khalid Zine-Dine Mohammed V University, Faculty of Sciences, Rabat, Morocco

khalid.zinedine@um5.ac.ma

Mohamed Bakhouya International University of Rabat, LERMA Lab., Sala Al Jadida, Morocco

mohamed.bakhouya@uir.ac.ma

Abstract— The accurate location of sensor nodes in Wireless Sensor Networks (WSNs) is considered as critical problem in many applications of IoT. In the last decades, many localization solutions have been proposed to provide location of sensor nodes, however a limited number of these solutions have been presented to accurately estimate the location of sensor nodes in the Internet of Things (IoT). In this paper we present three localization solutions, named Centroid + 4A, ICentroid, and ICentroid + 4A respectively, based on Centroid localization algorithm and a new weighted formula to calculate the position of unknown nodes. The performance of the proposed solutions was evaluated using the network simulator (OMNeT++) and compared to the basic Centroid localization algorithm. Results show that the proposed localization algorithms can significantly reduce the localization error of algorithms in Wireless Sensor Networks.

Keywords— Wireless Sensor Networks, Internet of Things Localization Algorithm, Centroid, ICentroid, Simula-tions and Performance Evaluation.

ID-96

Smart cities/Farming and Mobile Technologies

mailto:a.hadir@encgcasa.mamailto:khalid.zinedine@um5.ac.ma

27

An overview of recommender systems in the context of smart cities

Rabie MADANI, Abderrahmane EZ-ZAHOUT and Abdellah IDRISSI1 Intelligent Processing and Security of Systems Team,

Computer Sciences Department, Faculty of Sciences, Mohammed V University of Rabat, Morocco

rabie.madani@um5.ac.ma; abderrahmane.ezzahout@um5s.net.ma; idriab@gmail.com

Abstract—The concept of smart city appeared following technological, societal and organizational changes. A smart city uses The Internet of things (IoT) which refers to billions of physical devices connected to internet, to collect data and use it to effectively manage resources and to improve the quality of urban services. The use of recommender systems in smart cities plays an important role in the process of guiding citizens to find services that match with their preferences. Recommendations provided allow users to satisfy their needs in an efficient and easy way and make their daily lifes less complicated. In this paper, we present an overview of recommender systems in the context of smart cities and describe real-world application of recommender systems in IoT and in smart cities.

Keywords— Smart cities, Recommender systems, Internet of things. ID-64

Smart cities/Farming and Mobile Technologies

28

A re-engineering approach for extension of the Tourist Guide Knowledge Base

Asya Stoyanova-Doycheva, Todorka Glushkova, Emil Doychev Faculty of Mathematics and Informatics, University of Plovdiv “Paisii Hilendarski”

Plovdiv, Bulgaria astoyanova@uni-plovdiv.net, glushkova@uni-plovdiv.bg

Nevena Moraliyska

Department of Intelligent Systems Institute of Information and Communication Technologies (IICT), BAS

Sofia, Bulgaria nevena.uzunova@gmail.com

Abstract— The paper presents an extension of the knowledge base of the Tourist Guide with comprehensive information about Bulgarian cultural, historical, and natural sites, available in the databases created under the BECC project. To accomplish the task of enriching the knowledge base, the architecture of the Tourist Guide that was created as a reference architecture of the Virtual-Physical Space (ViPS) is presented and the restructuring process of the components in this architecture is described. In order to use the created databases in the BECC project, we had to re-engineer them on the basis of standards for the presentation of cultural and historical sites such as UNESCO and CCO (Cataloging Cultural Objects). Index Terms— Cataloging Cultural Objects (CCO), Virtual Physical Space (ViPS), Tourist Guide.

ID- 40

Smart cities/Farming and Mobile Technologies

29

NFaaS: A new cloud service for water distribution network monitoring

Rayane El Sibai Faculty of Engineering, Al Maaref University, Beirut, Lebanon

rayane.elsibai@mu.edu.lb

Jacques Bou Abdo Faculty of Natural and Applied Sciences, Notre Dame University Deir El Kamar, Lebanon

jbouabdo@ndu.edu.lb

Chady Abou Jaoude Ticket Lab., Faculty of Engineering Antonine University Baabda, Lebanon

chady.aboujaoude@ua.edu.lb

Jacques Demerjian LaRRIS, Faculty of Sciences Lebanese University Fanar, Lebanon

jacques.demerjian@ul.edu.lb

Yousra Chabchoub LISITE ISEP Issy-les-Moulineaux, 92130, France

yousra.chabchoub@isep.fr

Raja Chiky LISITE ISEP Issy-les-Moulineaux, 92130, France

raja.chiky@isep.fr Abstract— Water monitoring is one of the critical battles of sustainability for a better future of humanity. 44 countries are considered at high risk of the water crisis, 28 of which are developing countries and have limited capacity to deploy a national scale solution. As a response to the United Nation’s sustainability goals and initiatives, this paper proposes an intelligent water monitoring service which acts as a foundational infrastructure for all future water management systems. It also provides municipalities, Non-Governmental Organization and other private initiatives with the tools needed to establish local water monitoring in the scale of villages or rural areas with a very small initial investment.

Index Terms— —Cloud computing, data streams, data quality, sampling techniques, Software as a Service (SaaS). ID-11

Cloud services

mailto:rayane.elsibai@mu.edu.lbmailto:jbouabdo@ndu.edu.lbmailto:chady.aboujaoude@ua.edu.lbmailto:jacques.demerjian@ul.edu.lbmailto:yousra.chabchoub@isep.fr

30

An overall statistical analysis of AI tools deployed in Cloud computing and networking systems

Ikhlasse Hamzaoui

Research Foundation for Development and Innovation in Science and Engineering, Casablanca, Morocco

Engineering research laboratory, System Architecture Team, ENSEM, Hassan II University, 8118, Casablanca, Morocco

EIGSI, La Rochelle, France ikhlasse.h12@gmail.com

Benjamin Duthil, EIGSI, La Rochelle, France

L3i, La Rochelle University, La Rochelle, France duthil@eigsi.fr

Vincent Courboulay L3i, La Rochelle University, La Rochelle, France

vcourbou@univ-lr.fr

Hicham Medromi Research Foundation for Development and Innovation in Science and Engineering,

Casablanca, Morocco Engineering research laboratory, System Architecture Team, ENSEM, Hassan II University, 8118,

Casablanca, Morocco hmedromi@yahoo.fr

Abstract— As the vast amount of data destined to cloud systems never stop growing in seconds, minutes, hours and daily basis, the development of dynamic, autonomous and proactive architectures for cloud resources scheduling becomes a veritable prerequisite. This dominant trend is inciting to further seek for complete and accurate forecasting and predictive models to support decision making in several cloud-scheduling levels. In this perspective, this paper is a result of a meticulous statistical analysis of about five hundred relevant research articles dealing with proactive resources scheduling in cloud, fog, edge computing and networking systems, using a complete panoply of Artificial Intelligence (AI) predictive techniques. The first aim is to highlight for the first time current trends bridging the gap between cloud services management and AI tools.

Keywords— AI tools, Cloud Computing, Networking, Proactive scheduling ID-10

Cloud services

mailto:ikhlasse.h12@gmail.commailto:duthil@eigsi.frmailto:vcourbou@univ-lr.fr

31

Multi-task Offloading and Computational Resources Management in a Mobile Edge

Computing Environment

Mohamed EL GHMARY FSDM, LIIAN Laboratory, Sidi Mohamed Ben Abdellah University

Fez, Morocco mohamed.elghmary@usmba.ac.ma

Youssef HMIMZ FSDM, LIIAN Laboratory, Sidi Mohamed Ben Abdellah University

Fez, Morocco youssef.hmimz@usmba.ac.ma

Tarik CHANYOUR FSDM, LIIAN Laboratory, Sidi Mohamed Ben Abdellah University

Fez, Morocco tarik.chanyour@usmba.ac.ma

Mohammed Ouc¸amah CHERKAOUI MALKI FSDM, LIIAN Laboratory, Sidi Mohamed Ben Abdellah University

Fez, Morocco oucamah.cherkaoui@usmba.ac.ma

Abstract— In Mobile Cloud Computing, Smart Mobile Devices (SMDs) and Cloud Computing are combined to create a new infrastructure that allows data processing and storage outside the device. The Internet of Things refers to the billions of physical devices that are connected to the Internet. With the rapid development of these, it is clear that the requirements are largely based on the need for autonomous devices to facilitate the services required by applications that require rapid response time and flexible mobility. In this article, we study the management of computational resources and the trade-off between the energy consumption of an SMD and the processing time of its tasks. For this, we define a system model, a problem formulation and offer heuristic solutions for offloading tasks in order to jointly optimize the allocation of computing resources under limited energy and sensitive latency. In addition, we introduce the residual energy of the SMD battery and the sensitive latency of its tasks in defining the weighting factor of energy consumption and processing time. Keywords-context— mobile edge computing, computation offloading, energy, processing time, bi-objective optimization, heuristic, multi-task.

ID -102

Cloud services

mailto:mohamed.elghmary@usmba.ac.mamailto:youssef.hmimz@usmba.ac.mamailto:tarik.chanyour@usmba.ac.ma

32

Deployment of Containerized Deep Learning Applications in the cloud

Rim Doukha , Sidi Ahmed Mahmoudi, Mostapha Zbakh , Pierre Manneback Computer Science and Artificial Intelligence Department, Faculty of Engineering,

University of Mons Mons, Belgium

National School of Computer Science and Systems Analysis, Mohamed V University Rabat, Morocco

{rim.doukha, sidi.mahmoudi, pierre.manneback}@umons.ac.be m.zbakh@um5s.net.ma

Abstract— During the last years, the use of Cloud computing environment has increased as a result of the various services offered by Cloud providers (Amazon Web Services, Google Cloud, Microsoft Azure, etc.). Many companies are moving their data and applications to the Cloud in order to tackle the complex configuration effort, for having more flexibility, maintenance, and resource availability. However, it is important to mention the challenges that developers may face when using a Cloud solution such as the variation of applications requirements (in terms of computation, memory and energy consumption) over time, which makes the deployment and migration a hard process. In fact, the deployment will not depend only on the application, but it will also rely on the related services and hardware for the proper functioning of the application. In this paper, we propose a Cloud infrastructure for automatic deployment of applications using the services of Kubernetes, Docker, Ansible and Slurm. Our architecture includes a script to deploy the application depending of its requirement need. Experimentation are conducted with the analysis and the deployment of Deep Learning (DL) applications and more particularly images classification and object localization. Keywords— Cloud Computing, Application Deployment, Application Migration, Kubernetes, Docker, Ansible, Slurm, Deep Learning. ID- 51

Cloud services

mailto:pierre.manneback%7d@umons.ac.be

33

Autonomous Provisioning of Preemptive Instances in Google Cloud for Maximum Performance Per Dollar

Harek Haugerud, Jonas Kruger Svensson and Anis Yazidi

Department of Computer Science OsloMet –– Oslo Metropolitan University

Oslo, Norway

Abstract— Cloud computing and its popularity has boomed over the last decade, enabling anyone to rent computing power on demand. Cloud providers such as Amazon and Google rent out surplus computing power for a discounted price according to demand in their data centers, but with the trade off that it is revocable and can only be rented for a short amount of time. This paper investigates the use of surplus computing power in order to reduce the cost of batch computing. We rely on a simple economical principle, the most cost-efficient Virtual Machine (VM) in a public cloud is the one that offers the highest performance per dollar. Therefore by rescheduling the workloads to the most cost-efficient location in terms of performance per dollar our solution dynamically provisions preemptible VMs in Google Cloud while continuously monitoring the performance per dollar of all available resources in every region. The algorithm automatically relocates workloads to a less expensive location if any appears and handles revoked access of the resources. Our algorithm views the cost reduction problem as a linear optimization problem with constraints and solves it using a greedy procedure. In the experiment we spawn Docker containers to mine cryptocurrency. The experimental results show that 67% of the cost is saved compared to renting on-demand VMs. The system can readily be extended to containers processing similar types of workloads and more generally to applications where the performance per dollar is easy to measure.

Keywords— Autonomous VM, performance per dollar, containers, VM migration, global cloud.

ID -87

Cloud services

34

A Review of Credit Card Fraud Detection Using Machine Learning Techniques

Nadia Boutaher Laboratory of Computer Science, Systems

Modeling and Decision Support Faculty of Sciences AinChock, Hassan II University

Casablanca, Morocco nadia.boutaher1@gmail.com

Amina Elomri, Noreddine Abghour, Khalid Moussaid and Mohamed Rida Laboratory of Computer Science, Systems Modeling and Decision Support Faculty of Sciences

AinChock, Hassan II University Casablanca, Morocco {amina.elomri, noreddine.abghour, khalid.moussaid, mohamed.rida}@univh2c.ma

Abstract— Big Data technologies concern several critical areas such as Healthcare, Finance, Manufacturing, Transport, and ECommerce. Hence, they play an indispensable role in the financial sector, especially within the banking services which are impacted by the digitalization of services and the evolvement of ecommerce transactions. Therefore, there are various issues and challenges related to the banking sector and credit card transaction services generating by the emergence of the credit card use and the increasing number of fraudsters. Unfortunately, these issues obstruct the performance of Fraud Control Systems (Fraud Detection Systems & Fraud Prevention Systems) and abuse the transparency of online payments. Thus, financial institutions aim to secure credit card transactions and allow their customers to use e-banking services safely and efficiently. To reach this goal, they try to develop more relevant fraud detection techniques that can identify more fraudulent transactions and reduce the number of frauds and false alarms. This paper aims to present a comprehensive analysis of fundamental aspects of fraud detection, the current systems of fraud detection, the issues and challenges of frauds related to the banking sector, and the existing solutions based on machine learning techniques. Index Terms—big data; machine learning techniques; credit card fraud detection; legitimate transaction; financial institutions; digitalization; e-commerce. ID-53

Machine Learning

mailto:nadia.boutaher1@gmail.com

35

Machine Learning for Anomaly Detection. Performance Study considering Anomaly Distribution in an Unbalanced Dataset

Salma El Hajjami

IASSE Laboratory, ENSA, USMBA, Fez, Morocco salma.elhajjami@usmba.ac.ma

Jamal Malki L3i Laboratory, La Rochelle University, La Rochelle, France

jmalki@univ-lr.fr

Mohammed Berrada IASSE Laboratory, ENSA, USMBA, Fez, Morocco

mohammed.berrada@gmail.com

Bouziane Fourka aYaline R&D Laboratory, aYaline, Poitiers-Futuroscope, France

bfourka@ayaline.com

Abstract— The continuous dematerialization of real-world data greatly contributes to the increase in the volume of data exchanged. In this case, anomaly detection is increasingly becoming an important task of data analysis in order to detect abnormal data, which is of particular interest and may require action. Recent advances in artificial intelligence approaches, such as machine learning, are making an important breakthrough in this area. Typically, these techniques have been designed for balanced data sets or that have certain assumptions about the distribution of data. However, the real applications are rather confronted with an imbalanced data distribution, where normal data are present in large quantities and abnormal cases are generally very few. This makes anomaly detection similar to looking for the needle in a haystack. In this article, we develop an experimental setup for comparative analysis of two types of machine learning techniques in their application to anomaly detection systems. We study their performance taking into account anomaly distribution in an imbalanced dataset. Keywords— Anomaly Detection, Data Analysis, Artificial Intelligence, Machine Learning, Imbalanced Data. ID-37

Machine Learning

mailto:salma.elhajjami@usmba.ac.mamailto:jmalki@univ-lr.frmailto:mohammed.berrada@gmail.commailto:bfourka@ayaline.com

36

Novel Convex Polyhedron Classifier for Sentiment Analysis

EL MRABTI Soufiane, LAZAAR Mohamed ENSIAS, Mohammed V University

Rabat, Morocco elmrabtisouf@gmail.com; mohamed.lazaar@um5.ac.ma

AL ACHHAB Mohammed, OMARA Hicham ENSA, Abdelmalek Essaadi University

Tetuan, Morocco m.alachhab@uae.ac.ma; hichamomara@gmail.com

Abstract— In this paper, we propose a Novel Convex Polyhedron classifier (NCPC) based on the geometric concept convex hull. NCPC is essentially a piecewise linear classifier. It partitions linearly non-separable data into approximately linearly separable subset. Each subset of data is classified by a linear hyperplane. Experimental results on several real-world datasets indicate that the proposed classifier outperforms the other classifiers and achieved state-of-the-art performances. Keywords— Convex Polyhedron classifier; convex hull; Sentiment Analysis; Feature selection; Support vector machines. ID-73

Machine Learning

mailto:elmrabtisouf@gmail.commailto:mohamed.lazaar@um5.ac.mamailto:m.alachhab@uae.ac.mamailto:hichamomara@gmail.com

37

A Recommendation approach based on Correlation and Co-occurrence within social learning network

Sonia Souabi, Asmaâ Retbi, Mohammed Khalidi Idrissi, Samir Bennani

RIME TEAM-Networking, Modeling and e-Learning Team- MASI Laboratory- ENGINEERING.3S Research centerMohammadia School of Engineers (EMI)

MOHAMMED V UNIVERSITY IN RABAT, Morocco

soniasouabi@research.emi.ac.ma, retbi@emi.ac.ma, khalidi@emi.ac.ma, sbennani@emi.ac.ma

Abstract— In the context of e-learning, social learning is viewed as an evolving educational practice, namely in social networks. It is extensively associated with new educational technologies and fosters collaborative learning between learners. To handle the various pedagogical resources, several recommendation systems were proposed, with considerable emphasis on interactions and social relationships, except that they did not raise a critical aspect, namely the underlying nature of the relationship between the learners' actions and recommendations. To support the current recommendation systems, we propose a recommendation system that can measure the influence of learners' actions on the calculated recommendations. We therefore seek to evaluate the connection and link between these actions, and thus to combine the two parameters: correlation and co-occurrence by estimating the similarity of occurrences on the one hand and the probability of two actions taking place on the other hand. Keywords-component— Social learning; recommendation systems; correlation; co-occurrence; actions; social networks ID-7

Machine Learning

38

Graph-based Model for Negative e-WOM Influence in Social Media

Abderraouf DEMBRI and Mohamed GHARZOULI MISC Lab, University of Constantine 2 -Abdelhamid Mehri, Constantine, Algeria

abderraouf.dembri@univ-constantine2.dz, mohamed.gharzouli@univ-constantine2.dz

Abstract— Nowadays, several companies use social media marketing to increase profit and control the market. The customer’s feedback has a powerful influence on company reputation by conveying their experience in social media. Customers exchange their feedback about the services using electronic Word-of-Mouth (e-WOM). Negative feedback could help companies improve their service to increase profit. In this work, we propose an approach to determine the effect of negative e-WOM relating to a company’s products or services. Firstly, we apply a machine-learning algorithm called random forest to classify e-WOM on three classes based on polarity: Positive, negative, or neutral. Secondly, we group negative e-WOM into different clusters based on their topics using a similarity method named cosine similarity. Thirdly, we generate an influence graph of negative e-WOM based on time precedence and social ties. Finally, we analyze the resulted graph to identify risk patterns and convey useful information. The provided method is implemented using Python and is tested with collected data. Keywords— Social media marketing, Electronic word-of-mouth, Social commerce, Machine learning. ID-66

Machine Learning

39

Big Data Architectures Benchmark for Forecasting Electricity Consumption

Houda Daki, Asmaa El Hannani and Hassan Ouahmane Laboratory of Information Technologies, National School of Applied Sciences

University of Chouaib Doukkali, El Jadida, Morocco {daki.h, elhannani.a, ouahmane.h}@ucd.ac.ma

Abstract— It is in the sector of the forecast of the electrical consumption of educational buildings that we will be interested in this paper. In fact, buildings (residential and tertiary) are key elements in the electricity grid because they present one of the highest power consuming sectors. And for the network operators, the forecasts play a crucial role for the stability of the network and for maintaining the balance between electricity supply and demand. Fortunately, this is possible by combining Big Data analytics with statistical and machine learning techniques, which have been proven in recent research to be very effective in predicting renewable energy production and consumption at various time scales. It is in this perspective that the National School of Applied Sciences of El Jadida (NSASE), a Moroccan engineering school has decided to deploy a Big Data architecture to implement a custom system to predict electrical energy consumption by analyzing all factors that influence its electrical energy use. In this paper, we propose a benchmark of the main Big Data architectures in the field and that will cover all electrical energy data processing from data collection, data storage, data analytic and data visualization. The aim of this benchmark is to choose the optimal architecture in term of fault tolerance, resource management, data storage and data modelling to forecast electricity consumption in the NSASE. Keyword— Big Data architecture, Lambda architecture, SMACK architecture, Electrical forecasting, Smart grid. ID-52

Big data

40

A Big Data Placement Strategy in Geographically Distributed Datacenters

Laila Bouhouch, Mostapha Zbakh National School of Computer science and Systems Analysis,

Mohamed V University Rabat, Morocco

lailabouhouch94@gmail.com laila.bouhouch@mines-paristech.fr

m.zbakh@um5s.net.ma

Claude Tadonki Mines ParisTech-PSL, Centre de Recherche en Informatique (CRI)

Paris, France claude.tadonki@mines-paristech.fr

Abstract— With the pervasivness of the “Big Data” characteristic together with the expansion of geographically distributed datacenters in the Cloud computing context, processing largescale data applications has become a crucial issue. Indeed, the task of finding the most efficient way of storing massive data across distributed locations is increasingly complex. Furthermore, the execution time of a given task that requires several datasets might be dominated by the cost of data migrations/exchanges, which depends on the initial placement of the input datasets over the set of datacenters in the Cloud and also on the dynamic data management strategy. In this paper, we propose a data placement strategy to improve the workflow execution time through the reduction of the cost associated to data movements between geographically distributed datacenters, considering their characteristics such as storage capacity and read/write speeds. We formalize the overall problem and then propose a data placement algorithm structured into two phases. First, we compute the estimated transfer time to move all involved datasets from their respective locations to the one where the corresponding tasks are executed. Second, we apply a greedy algorithm in order to assign each dataset to the optimal datacenter w.r.t the overall cost of data migrations. The heterogeneity of the datacenters together with their caracteristics (storage and bandwith) are both taken into account. Our experiments are conducted using Cloudsim simulator. The obtained results show that our proposed strategy produces an efficient placement and actually reduces the overheads of the data movement compared to both a random assignment and a selected placement algorithm from the litterature. Keywords-component— Big Data, Cloud Computing, Data Placement, Greedy algorithm, Cloudsim. ID-59

Big data

mailto:laila.bouhouch@mines-paristech.frmailto:m.zbakh@um5s.net.ma

41

New Approach for implementing big Datamart using NoSQL Key-Value stores

KHALIL Abdelhak SIAD Laboratory, Hassan First University of Settat

Settat, Morocco a.khalil@uhp.ac.ma

BELAISSAOUI Mustapha SIAD Laboratory, Hassan First University of Settat

Settat, Morocco mustapha.belaissaoui@uhp.ac.ma

Abstract— Nowadays, NoSQL technologies are gaining significant ground and considered as the future of data storage, especially when it comes to huge amount of data, which is the case of data warehouse solutions. NoSQL databases provide high scalability and good performance among relational ones, which are really time consuming and can’t handle large data volume. The growing popularity of the term NoSQL these days and vaguely related phrases like big data make us think about using this technology in decision support systems. The purpose of this paper is to investigate the possibility to instantiate a big data mart under one of the most popular and least complicated types of NoSQL databases; namely key-value store, the main challenge is to make a good correlation between the old-school approach of data warehousing based on traditional databases that favor data integrity, and interesting opportunities offered by new generation of database management systems. The paper describes the transformation process from multidimensional conceptual schema to the logical model following three approaches, and outlines a list of strengths and weaknesses for each one based on practical experience under Oracle NoSQL Database. Keywords— DataMart, NoSQL, Decision support systems, Big Data. ID-41

Big data

mailto:a.khalil@uhp.ac.ma

42

Big data for sustainability: A qualitative analysis

Wail El Hilali, Abdellah El Manouar and Mohammed Abdou Janati Idrissi Mohammed V University

Rabat, Morocco wailelhilali@gmail.com; a.elmanouar@um5s.net.ma; a.janati@um5s.net.ma

Abstract— In this time and age, companies are facing difficult times to sustain their competitive advantages. The risk of disruption, the fierce and the bloody competition, the change in customer behaviors and the scarcity of resources, all of these are drivers that push companies to think out of the box and to change the way business is done. Digital capabilities, and especially big data, comes as a new tool to use in order to continue creating and capturing value in a sustainable manner. This paper is an attempt to enrich the literature about this subject. We conducted a qualitative analysis among three big size companies inside Morocco in order to assess the impact of using big data on these companies’ financial numbers, social footprints and environmental negative externalities. The finding showed that embracing this technology has helped these companies for their quest to reach sustainability.

Keywords- big data; sustainability; digital transformation; digital capabilities ID-6

Big data

mailto:wailelhilali@gmail.commailto:a.elmanouar@um5s.net.ma

43

Bloom filter and its variants for the optimization of MapReduce’s algorithms: A review

F.EZZAKI Laboratoire Informatique, Modélisation des Systèmes et Aide à la Décision (LIMSAD)

Faculty of Sciences, University of Hassan II Department of Mathematics and Informatics

Casablanca, Morocco ezzakifati@gmail.com

N.Abghour, A.Elomri, K.Moussaid, M.Rida Laboratoire Informatique, Modélisation des Systèmes et Aide à la Décision (LIMSAD)

Faculty of Sciences, University of Hassan II Department of Mathematics and Informatics

Casablanca, Morocco {noreddine.abghour, amina.elomri, khalid.moussaid, mohamed.rida}@univh2c.ma

Abstract— The bloom filter is a probabilistic model used to represent a set in order to test the existence of an element in this set, i.e., for any given item, the bloom filter could test if this element is a member or not of the set represented by the bloom filter. The bloom filter has many advantages due to its simplicity and efficiency in highly solving the issue of data representation in many fields and to support membership queries, it has been known as space and time-efficient randomized data structure, by filtering out redundant data and optimizing the memory consumption. However, bloom filters are limited to membership tests and don’t support the deletion of elements. They also generate the false positive probability as they are based on a probabilistic model, this error rate is generated when an element that doesn’t belong to a set is considered as a member of this set by the bloom filter. Our goal is to compare a number of wellexisted algorithms related to the boom filter for future work on the optimization of the join’s algorithms in MapReduce. This paper provides an overview of the different variants of the bloom filter and analyses the studies that have been interested in this area of research.

Keywords— big data storage; optimization; bloom filter; cuckoo filter; membership; data structure; join algorithms; mapreduce. ID-55

Big data

44

Deep Learning and Tensorflow for Tracking People’s Movements in a Video

Jemai Bornia Robotics, Computing and Complex Systems, National Engineering School of Tunis

University Tunis El Manar, Tunis, Tunisia bornia.jemai@gmail.com

Ali Frihida Robotics, Computing and Complex Systems, National Engineering School of Tunis

University Tunis El Manar, Tunis, Tunisia ali.frihida@enit.utm.tn

Olivier Debauche Faculty of Engineering – ILIA, Infortech Institut

University of Mons, Mons, Belgium

Sidi Ahmed Mahmoudi Faculty of Engineering – ILIA, Infortech Institut

University of Mons, Mons, Belgium

Pierre Manneback Faculty of Engineering – ILIA, Infortech Institut

University of Mons, Mons, Belgium

Abstract— With the advent of the digital age and more specifically videos, a huge amount of data is produced every day such as television archiving, video surveillance, etc. Faced with the need to keep control over this content, in terms of data analysis, classification, accurate AI (Artificial Intelligence) algorithms are required to perform this task efficiently and quickly. In this paper, we propose an approach for movement analysis from video sequences using deep learning technologies. The proposed approach splits video in set of images, detects objects/entities present in these images and stores their descriptions into a standard XML file. As result, we provide a Deep Learning algorithm using TensorFlow for tracking motion and animated entities in video sequences. Index Terms— Object tacking, motion tracking, TensorFlow, deep learning, video. ID-97

Deep Learning

mailto:bornia.jemai@gmail.commailto:ali.frihida@enit.utm.tn

45

SAHAR-LSTM: An enhanced Model for Sentiment Analysis of Hotels’Arabic Reviews based on LSTM

Manal Nejjari and Abdelouafi Meziane

Department of Computer Science Science College, Mohammed 1st University Oujda, Morocco

manal.nejjari10@gmail.com; Abdelouafi_meziane@yahoo.fr

Abstract— Over the last few years, many scientists have paid special attention to the field of Opinion Mining (OM) or Sentiment analysis (SA), thanks to its interesting and useful applications in different domains. The most popular studies have tackled the issue of SA in the English language, however those dealing with SA in the Arabic language are, up to