PhD defense of Nejat Arınık – 29 June 2021

29 June 2021

I will defend my thesis titled ‘Multiplicity in the Partitioning of Signed Graphs’ on Tuesday, June 29th at 2 p.m. The presentation will be in French and will take place in a hybrid mode: at the Blaise Pascal amphitheater of the CERI and via video conference. The video link is: You are all cordially invited. The jury will consist of: You are also invited to the thesis reception following the defense. Abstract: The partitioning of signed graphs is an important task from an application standpoint, as finding a balanced partition helps in understanding the system modeled by the signed graph. However, the standard approach in the literature aims to find a single partition, as if it adequately characterizes the system under study. Yet, multiple partitions may be needed to construct a fairer image of the studied system. Although this notion of multiplicity is crucial from the end-users’ perspective, it has been scarcely addressed in the literature. In this thesis, we aim to relax the assumption of a unique partition and search for multiple partitions, each within two distinct situations. The first situation concerns signed multiplex graphs. Such a graph consists of several separate graphs, referred to as layers, each Plus d'infos

HDR defense of Vincent Labatut – 16 June 2021

16 June 2021

I will defend my HDR entitled ‘Combining Heterogeneous Information: Contributions to the Extraction and Analysis of Feature-Rich Complex Networks’ on Wednesday, June 16, 2021, at 9 a.m. The presentation will be in English and will take place via video conference at the link You are all cordially invited to attend. Jury: Abstract: The concept of Complex Network is generally used in the literature to refer to a graph representing a real-world complex system. This confers such graphs so-called non-trivial topological properties that distinguish them from regular and random graphs. Among them, the most widely known are small-worldness and scale-freeness, whose study marked the beginning of a new research domain now called Network Science, and aiming at studying complex networks. It is a multidisciplinary field that relies largely on a number of pre-existing domains, in particular graph theory, quantitative sociology, computer science, operations research, statistical physics, and of course complex systems. Network Science is mainly a data science, as its starting point is the modeling of real-world systems. As such, its emergence is due not only to the convergence of interdisciplinary efforts, but also to the availability of the resources required to build and study large and/or numerous complex networks: Plus d'infos

PhD defense of Randa Abdelmonem – 26 January 2021

26 January 2021

Date: Tuesday 26 January 2021 at 9:00 AM. Title: Quality of Service and Privacy in Internet of Things Dedicated to Healthcare  Jury: Abstract: The Internet of Things (IoT) based healthcare systems usually composed of medical and environmental sensors, remote servers, and the network. These systems focus on providing remote monitoring, disease diagnosis, and treatment progress observation. The healthcare systems in IoT domain helps in realizing long-term economical, ubiquitous, and patient centered care systems, that result in improving treatment and patient outcomes. This research contributes to the domain by proposing a Cloud-Fog based architecture that can embrace multiple healthcare scenarios, and able to adapt dynamically with the context and status of the patients. It allows the mobility and physical activity of the patients in the environment through deployment and implementation of an appropriate Received Signal Strength (RSS) based handoff mechanism. It also proposes a mobility-aware task scheduling and allocation approach in cloud-fog computing paradigm, called MobMBAR, with the objective of minimizing the total schedule time (makespan). MobMBAR performs dynamic balanced healthcare tasks distribution between the cloud and fog devices. It is a data locality based approach that depends on changing the location where the data is computed to where it actually resides. Plus d'infos

PhD defense of Mayeul Matthias – 22 January 2021

2 January 2021

On January 22, 2021, Mayeul Matthias will defend his thesis entitled “Recommendation of Personalized Cultural Paths – An Interdisciplinary Study of Automated Visit Proposals.” This thesis is supervised by Juan-Manuel Torres and Didier Josselin, and co-supervised by Fen Zhou. The jury consists of: Abstract: This thesis focuses on recommending cultural visits through an interdisciplinary approach. These works combine techniques from Operations Research and natural language processing while drawing on concepts from audience sociology and geography. We propose new methods for evaluating cultural points of interest and automatically creating tourist routes that take into account the desires expressed by a visitor. These principles are applied on two different scales and contexts: museum visits and cultural paths within a city. In the first part, we concentrate on visits to art museums based on the preferences expressed by the visitor and the prestige of the artworks. This dual approach allows classifying the works both according to the cultural affinities of the visitor and their importance within the museum. The latter is calculated by applying automatic text summarization algorithms to the museum’s official descriptions of the works, providing a visit profile reflecting the discovery of a museum through its masterpieces. This profile can then Plus d'infos

PhD defense of Dina Tarek – 18 December 2020

18 December 2020

Dina Tarek will defend her thesis on Friday, December 18th at 2:30 PM. This thesis is jointly supervised with Egypt. Titre : Development of Spectrum Sharing Protocol for Cognitive Radio Internet of Things Jury: Abstract: Internet of Things (IoT) presents a new life style by developing smart homes, smart grids, smart city, smart transportation … etc., so IoT is developing rapidly. However recent researches focus on developing the IoT applications disregarding the IoT spectrum scarcity problem facing it. Integrating Internet of Things (IoT) technology and Cognitive Radio Networks (CRNs), forming Cognitive Radio Internet of Things (CRIoTs), is an economical solution for overcoming the IoT spectrum scarcity. The aim of this thesis is to solve the problem of spectrum sharing for CRIoT; the work in thesis is presented in three parts. Our first contribution is to propose two new protocols to solve the problem of channel status prediction for interweave CRNs. Both protocols uses Hidden Markov Model (HMM). In the training stage of both protocols, the available data are trained to produce two HMM models, an idle HMM model and a busy one. Both models are used together to produce the 2-model HMM. In the prediction stage the first protocol uses Bayes theorem and the 2-model HMM, while the Plus d'infos

HDR defense of Richard Dufour – 8 December 2020

8 December 2020

Defense of the HDR entitled ‘Natural Language Processing: Studies and Contributions at the Frontiers of Interdisciplinarity’, on Tuesday, December 8, 2020, at 2:00 PM in the Thesis Room of Avignon University (Hannah Arendt Campus – City Center). The defense committee will be composed of:

PhD defense of Cyril Sahuc – 1st December 2020

1 December 2020

On December 1st, 2020, at 2:00 PM in the meeting room of LIA. Title: Mathematical Approaches for Retail Area Planning: Linear Models, Algorithms, and Multi-Agent Systems. Thesis in collaboration with Cyrille Genre-Grandpierre (UMR Espace) as part of an industrial regional scholarship. Due to the situation, the defense will be held remotely (like all theses currently).

PhD defense of Carlos González – 18 December 2019

18 December 2019

Thesis defense of Carlos González entitled ‘Multimedia and Multilingual Automatic Summarization and Information Retrieval’ on Wednesday, December 18, 2019, at 2:00 PM in the Thesis Room (Saint Marthe – City Center). Jury: Abstract: As multimedia sources have become massively available online, helping users to understand the large amount of information they generate has become a major issue. One way to approach this is by summarizing multimedia content, thus generating abridged and informative versions of the original sources. This PhD thesis addresses the subject of text and audio-based multimedia summarization in a multilingual context. It has been conducted within the framework of the Access Multilingual Information opinionS (AMIS) CHISTERA-ANR project, whose main objective is to make information easy to understand for everybody. Text-based multimedia summarization uses transcripts to produce summaries that may be presented either as text or in their original format. The transcription of multimedia sources can be done manually or automatically by an Automatic Speech Recognition (ASR) system. The transcripts produced using either method differ from wellformed written language given their source is mostly spoken language. In addition, ASR transcripts lack syntactic information. For example, capital letters and punctuation marks are unavailable, which means sentences are nonexistent. To deal Plus d'infos

PhD defense of Titouan Parcollet – 3 December 2019

3 December 2019

Thesis defense of Titouan Parcollet, entitled “Artificial Neural Networks Based on Quaternion Algebra,” will take place on Tuesday, December 3, 2019, at 2:30 PM in the Blaise Pascal amphitheater (CERI). The thesis will be presented before a jury composed of: The defense will be conducted in French. You are also invited to the reception following the defense in Room 5. Abstract: In recent years, deep learning has become the preferred approach for developing modern artificial intelligence (AI). The significant increase in computing power, along with the ever-growing amount of available data, has made deep neural networks the most efficient solution for solving complex problems. However, accurately representing the multidimensionality of real-world data remains a major challenge for artificial neural architectures. To address this challenge, neural networks based on complex and hypercomplex number algebras have been developed. Thus, the multidimensionality of data is integrated into neurons, which are now complex and hypercomplex components of the model. In particular, quaternion neural networks (QNNs) have been proposed to process three-dimensional and four-dimensional data, based on quaternions representing rotations in our three-dimensional space. Unfortunately, unlike complex-valued neural networks, which are now accepted as an alternative to real-valued neural networks, QNNs suffer from several limitations, Plus d'infos

HDR defense of Mohamed Morchid – 26 November 2019

26 November 2019

On the next 26th of November at 4 PM in the thesis room (Hannah Arendt campus). This HDR entitled ‘Neural Networks for Natural Language Processing’ will be presented before a jury composed of: Reviewers: Mrs. Dilek Z. HAKKANI-TÜR Senior Principal Scientist, Alexa AI, USA Mr. Patrice BELLOT Professor, AMU Polytech’, LIS, Marseille Mr. Frédéric ALEXANDRE Research Director INRIA, Bordeaux Examiners: Mr. Yannick ESTÈVE Professor, AU, LIA, Avignon Mr. Frédéric BÉCHET Professor, AMU, LIS, Marseille

1 5 6 7 8