RECHERCHE  |  
Thèses de doctorat

Personalized Information Retrieval based on Time-Sensitive User Profile
Directeur(s):

Professeur Rim FAIZ

, Professeur Mohand BOUGHANEM

Année soutenance: 2017
- Abstract: Recently, search engines have become the main source of information for many users and have been widely used in different fields. However, Information Retrieval Systems (IRS) face new challenges due to the growth and diversity of available data. An IRS analyses the query submitted by the user and explores collections of data with unstructured or semi-structured nature (e.g. text, image, video, Web page etc.) in order to deliver items that best match his/her intent and interests. In order to achieve this goal, we have moved from considering the query-document matching to consider the user context. In fact, the user profile has been considered, in the literature, as the most important contextual element which can improve the accuracy of the search. It is integrated in the process of information retrieval in order to improve the user experience while searching for specific information. As time factor has gained increasing importance in recent years, the temporal dynamics are introduced to study the user profile evolution that consists mainly in capturing the changes of the user behavior, interests and preferences, and updating the profile accordingly. Prior work used to discern short-term and long-term profiles. The first profile type is limited to interests related to the user's current activities while the second one represents user's persisting interests extracted from his prior activities excluding the current ones. However, for users who are not very active, the short-term profile can eliminate relevant results which are more related to their personal interests. This is because their activities are few and separated over time. For users who are very active, the aggregation of recent activities without ignoring the old interests would be very interesting because this kind of profile is usually changing over time. Unlike those approaches, we propose, in this thesis, a generic time-sensitive user profile that is implicitly constructed as a vector of weighted terms in order to find a trade-off by unifying both current and recurrent interests. User profile information can be extracted from multiple sources. Among the most promising ones, we propose to use, on the one hand, searching history. Data from searching history can be extracted implicitly without any effort from the user and includes issued queries, their corresponding results, reformulated queries and click-through data that has relevance feedback potential. On the other hand, the popularity of Social Media makes it as an invaluable source of data used by users to express, share and mark as favorite the content that interests them. First, we modeled a user profile not only according to the content of his activities but also to their freshness under the assumption that terms used recently in the user's activities contain new interests, preferences and thoughts and should be considered more than old interests. In fact, many prior works have proved that the user interest is decreasing as time goes by. In order to evaluate the time-sensitive user profile, we used a set of data collected from Twitter, i.e a social networking and microblogging service. Then, we apply our re-ranking process to a Web search system in order to adapt the user's online interests to the original retrieved results. Second, we studied the temporal dynamics within session search where recent submitted queries contain additional information explaining better the user intent and prove that the user hasn't found the information sought from previous submitted ones. We integrated current and recurrent interactions within a unique session model giving more importance to terms appeared in recently submitted queries and clicked results. We conducted experiments using the 2013 TREC Session track and the ClueWeb12 collection that showed the effectiveness of our approach compared to state-of-the-art ones. Overall, in those different contributions and experiments, we prove that our time-sensitive user profile insures better performance of personalization and helps to analyze user behavior in both session search and social media contexts.

Chercheur: Ameni Kacem

Contribution à l'étude de l'exploitation des données temporelles en présence d'imperfections.
Directeur(s):

Prof. Boutheina Ben Yaghlane

, Prof. Allel HadjAli

Année soutenance: 2017
ISG Tunis

Chercheur: Aymen GAMMOUDI

Influencers characterization in a social network for viral marketing perspectives
Directeur(s):

Boutheina Ben Yaghlane

, Arnaud Martin

Année soutenance: 2016
France

Chercheur: Siwar Jendoubi

Towards a decision support systemfor the vehicle routing problems
Directeur(s):

Saoussen Krichen

Année soutenance: 2016
Institut Supérieur de Gestion de Tunis

Chercheur: Takwa Tlili

Decion making under possibilistic uncertainty with multi-objective influence diagrams: Complexity results and algorithms
Directeur(s):

Nahla BEN AMOR

, Hélène FARGIER

Année soutenance: 2016

Chercheur: Essghaier Fatma

Learning possibilistic graphical models from data
Directeur(s):

Pr. Nahla Ben Amor

, Pr. Philippe Leray

Année soutenance: 2016
Institut Supérieur de Gestion de Tunis Université de Nantes
[BibTeX...]
@phdthesis{DBLP:phd/hal/Haddad16, author = {Maroua Haddad}, title = {Learning possibilistic graphical models from data. (Apprentissage de mod{\`{e}}les graphiques possibilistes {\`{a}} partir de donn{\'{e}}es)}, school = {University of Nantes, France}, year = {2016}, url = {https://tel.archives-ouvertes.fr/tel-01442642}, timestamp = {Tue, 07 Mar 2017 18:00:58 +0100}, biburl = {http://dblp.uni-trier.de/rec/bib/phd/hal/Haddad16}, bibsource = {dblp computer science bibliography, http://dblp.org} }

Chercheur: Haddad Maroua

Improved Methods for Dimensionality Reduction of Data Streams
Directeur(s):

Mohamed Limam

Année soutenance: 2016
Institut Supérieur de Gestion de Tunis

Chercheur: Haïfa Nakouri

On Feature Selection for Credit Scoring
Directeur(s):

mohamed limam

Année soutenance: 2015
Institut supérieur de gestion

Chercheur: Bouaguel Waad

Decision Making for Ontology Matching under the Theory of Belief Functions
Directeur(s):

Pr. Boutheina Ben Yaghlane

, Pr. Arnaud Martin, Grégory Smits

Année soutenance: 2015
Thèse en cotutelle Université de Tunis et Université de Rennes 1

Chercheur: Amira ESSAID

New AHP Methods For Handling Uncertainty Within The Belief Function Theory
Directeur(s):

Pr. Zied ELOUEDI (ISG, Tunis)

, Pr. Eric Lefevre (Université d’Artois, France)

Année soutenance: 2015

Chercheur: Ennaceur Amel

Stable and Efficient Feature Selection Methods for High Dimensional Data
Directeur(s):

Prof. Mohamed Limam

Année soutenance: 2015
Institut Supérieur de Gestion de Tunis

Chercheur: Ben Brahim Afef

New Methods for Maintaining Case Based Reasoning based on Clustering Techniques and Competence Models
Directeur(s):

Pr. Zied ELOUEDI

Année soutenance: 2014
Institut Supérieur de Gestion Tunis

Chercheur: SMITI Abir

New Danger Theory Classification Methods in an Imprecise Framework
Directeur(s):

Dr. Zied ELOUEDI

Année soutenance: 2014
Institut Supérieur de Gestion Tunis

Chercheur: Zaineb Chelly

Efficient k-means based methods for overlapping clustering
Directeur(s):

Dr. Nadia Essoussi

Année soutenance: 2014
<p>ISG Tunis, Universit&eacute; de Tunis</p>

Chercheur: Ben N'Cir Chiheb-Eddine

A Framework for Adaptive Risk-driven Intrusion Detection and Response Systems
Directeur(s):

Pr. Mhamed Ali Elaroui (FSEG Nabeul)

, Pr. Belgacem Raggad (Pace Univ.)

Année soutenance: 2014
Institut Supérieur de Gestion de Tunis(ISG)

Chercheur: Katar Chaker

Evolutionary Resolution Approaches for the Multi-Objective and Multi-Mode Assignment and Scheduling problem
Directeur(s):

Dr. Saoussen Krichen

, Dr. Adel Guitouni

Année soutenance: 2014
Institut supérieur de gestion de Tunis

Chercheur: Dridi Olfa

Source Independence in the Theory of Belief Functions
Directeur(s):

Pr. Boutheina Ben Yaghlane

, Pr. Arnaud Martin

Année soutenance: 2014
Université de Rennes 1

Chercheur: Mouna Chebbah

Causal Modeling Under a Belief Function Framework
Directeur(s):

Zied ELOUEDI

, Salem BENFERHAT

Année soutenance: 2013
Institut Supérieur de Gestion de Tunis & Université d'Artois

Chercheur: Imen BOUKHRIS

Dynamic Clustering within the Belief Function Framework
Directeur(s):

Pr. Zied ELOUEDI

Année soutenance: 2013
ISG, Tunis

Chercheur: Sarra Ben Hariz

Compiling Possibilistic Graphical Models: From Inference to Decision
Directeur(s):

Nahla Ben Amor

, Salem Benferhat

Année soutenance: 2013
Institut Supérieur de Gestion de Tunis et Université d'Artois

Chercheur: Raouia Ayachi

Un modèle de recherche d'information agrégée basé sur les réseaux bayésiens
Directeur(s):

Pr. Rim FAIZ

, Pr. Mohand BOUGHANEM

Année soutenance: 2013
IRIT/Université Toulouse 3 - Paul Sabatier

Chercheur: Naffakhi Najeh

Frequent Itemset Mining and Maintenance in Imperfect Databases
Directeur(s):

Boutheina Ben Yaghlane

Année soutenance: 2012
ISG de Tunis
[BibTeX...]
@PhdThesis{Bac12, title = "{Frequent Itemset Mining and Maintenance in Imperfect Databases}", author = "Mohamed Anis Bach Tobji", school = "University of Tunis, ISG", month = "November", year = "2012", }

Chercheur: Bach Tobji Mohamed Anis

A New Process-Based Approach for Implementing an Integrated Management System (QSE): Algorithms and Tools.
Directeur(s):

Pr. Nahla Ben Amor

Année soutenance: 2012
Institut Supérieur de Gestion de Tunis (ISG, Tunis).

Chercheur: Badreddine Ahmed

Possibilistic Decision Theory : From Theoretical Foundations to Influence Diagrams Methodology
Directeur(s):

Pr. Nahla Ben Amor

, Dr. Hélène Fargier

Année soutenance: 2012
IRIT Toulouse et ISG Tunis
[BibTeX...]
@PhDThesis{ Gu2012.2, author = {Guezguez, Wided}, title = "{Théorie de la décision possibiliste: des fondations théoriques aux diagrammes d'influence}", year = {2012}, month = {mai}, type = {Thèse de doctorat}, school = {Université Paul Sabatier}, address = {Toulouse, France}, language = {français}, URL = {ftp://ftp.irit.fr/IRIT/ADRIA/PapersFargier/TheseGuezguez.pdf} }

Chercheur: Guezguez Wided

Belief Rough Set Classifier
Directeur(s):

Zied Elouedi

, Pawan Lingras

Année soutenance: 2011
Institut supérieur de gestion de Tunis

Chercheur: Trabelsi Salsabil

Qualitative Possibilistic Graphical Models : From Independence to propagation Algorithms
Directeur(s):

Khaled Mellouli (IHEC, Carthage)

, Salem Benferhat (IRIT, Toulouse)

Année soutenance: 2002
Institut Supérieur de Gestion Tunis
[BibTeX...]
{larodec.com}

Chercheur: Nahla BEN AMOR

Belief Decision Tree: A New Decision Tree Approach under Uncertainty
Directeur(s):

Khaled Mellouli (ISG, Tunis)

, Philippe Smets (ULB, Bruxelles)

Année soutenance: 2002
ISG Tunis

Chercheur: Zied ELOUEDI