Friday, September 9. 2011
I am happy to announce my PhD defense September 28th, at 15:00 at Orsay. If you are interested, please let me know and I will provide full information about the place.
Title: Distributed Data Management with Access Control
Social Networks and Data of the Web
Abstract: The amount of information on the Web is spreading very rapidly. Users as well as companies bring data to the network and are willing to share with others. They quickly reach a situation where their information is hosted on many machines they own and on a large number of autonomous systems where they have accounts. Management of all this information is rapidly becoming beyond human expertise. We introduce WebdamExchange, a novel distributed knowledge-base model that includes logical statements for specifying information, access control, secrets, distribution, and knowledge about other peers. These statements can be communicated, replicated, queried, and updated, while keeping track of time and provenance. The resulting knowledge guides distributed data management.
WebdamExchange model is based on WebdamLog, a new rule-based language for distributed data management that combines in a formal setting deductive rules as in Datalog with negation, (to specify intensional data) and active rules as in Datalog¬¬ (for updates and communications). The model provides a novel setting with a strong emphasis on dynamicity and interactions (in a Web 2.0 style). Because the model is powerful, it provides a clean basis for the speciﬁcation of complex distributed applications. Because it is simple, it provides a formal framework for studying many facets of the problem such as distribution, concurrency, and expressivity in the context of distributed autonomous peers. We also discuss an implementation of a proof-of-concept system that handles all the components of the knowledge base and experiments with a lighter system designed for smartphones. We believe that these contributions are a good foundation to overcome the problems of Web data management, in particular with respect to access control.
Sunday, June 12. 2011
Presentation at WebDB2011 Posted by Alban Galland in Presentations at 08:57
I gave a presentation about my work on corroboration, June 12th, 2011, at the 14th International Workshop on the Web and Databases
Title: Web information management with access control
Authors: Serge Abiteboul, Alban Galland, Neoklis Polyzotis
Abstract: We investigate the problem of sharing private information on the Web, where the information is hosted on diﬀerent machines that may use diﬀerent access control and distribution schemes. We introduce a distributed knowledge-base model, termed WebdamExchange, that comprises logical statements for specifying data, access control, distribution and knowledge about other peers. The statements can be communicated, replicated, queried, and updated, while keeping track of time and provenance. This uniﬁed base allows applications to reason declaratively about what data is accessible, where it resides, and how to retrieve it securely.
Slides of the presentation: webdb-06-12-2011
Wednesday, April 13. 2011
Demonstration at ICDE2011 Posted by Alban Galland in Presentations at 14:18
I gave a a presentation about WebdamExchange, April 12-13th 2011, at the demonstration session of the ICDE2011 conference.
Title: Social networking on top of the WebdamExchange System
Authors: Emilien Antoine, Alban Galland, Kristian Lyngbaek, Amelie Marian, Neoklis Polyzotis
Live demonstration: SocialRock
Friday, December 10. 2010
Presentation at LIG/Hadas Posted by Alban Galland in Presentations at 14:14
I gave a presentation about my work on web data management, December 10th, 2010, at the Hadas working group.
Title: WebdamExchange and WebdamLog: some models for web data management
Abstract: The emergence of Web 2.0 and social network applications has enabled more and more users to share sensitive information over the Web. The information we manipulate has many facets: data, annotations, localization (e.g., bookmarks), login and keys, access rights, ontologies, beliefs, time and provenance information, etc. To find data, one typically has to perform a number of complex tasks such as search/query, authentication, data extraction. More and more, we also want to control how our personal data is used.
We will argue that all this should be viewed in the holistic context of a distributed knowledge base. More precisely, we use extensions of distributed datalog. Logical statements are used to capture these different facets of information that are typically considered in isolation. Knowledge can be communicated, replicated, queried, updated, and monitored. The fact that we use a formal model allows complex reasoning for searching information, within a rich mix of very different scenarios ranging from information in centralized servers to massively distributed, from fully trusted to untrusted, and providing encrypted or clear information, which is the reality of today's Web.
Slides of the presentation: grenoble-10-12-2010
Tuesday, September 14. 2010
Demonstration at VLDB2010 Posted by Alban Galland in Presentations at 08:17
I gave a presentation about Axart and data-centric workflows, September 14-15th, 2010, at the demonstration session of the VLDB2010 conference.
Title: Axart, Enabling Collaborative works with Active XML Artifacts
Authors: Serge Abiteboul, Pierre Bourhis, Alban Galland, Bogdan Marinoiu
Tuesday, April 20. 2010
Friday, April 2. 2010
I gave a presentation about my work on access control on social networks, April 1st, 2010, at the DAHU working group.
Title: Pastis, A Model for Data Access on the Web
Abstract: The emergence of Web 2.0 and social network applications has enabled more and more users to share sensitive information over the Web. This strongly stresses the need for a comprehensive logical model to specify, manipulate and control information in a distributed environment. We introduce such a model here, namely Pastis. The main originality of our approach is that we articulate the problem as a general distributed knowledge-base management problem. Logical statements are used to specify data, access rights, secret keys, localization, time, provenance and knowledge about other peers, aspects that are typically considered separately. Knowledge can be communicated, replicated, queried, updated, and monitored.
Relying on this knowledge base, we show how to control that information is not used improperly. The fact that the model is formally defined allows formally proving or disproving desirable properties such as soundness (data is only acquired legally) and completeness (one can acquire all data that one can legally claim).
We illustrate the use of the model with very different scenarios based on servers ranging from centralized to massively distributed, from fully trusted to untrusted, and providing encrypted or clear information. We briefly discuss a system we implemented that supports these different scenarios. The fact that they are all described in the same model opens the way for exploring situations that combine these different scenarios in arbitrarily rich ways, which is the reality of today's Web.
Slides of the presentation: dahu-1-04-2010
Tuesday, March 16. 2010
Presentation at Datalog2.0 workshop Posted by Alban Galland in Presentations at 14:20
Thursday, February 4. 2010
Presentation at WSDM2010 Posted by Alban Galland in Presentations at 16:20
I gave a presentation about my work on corroboration, February 5th, 2010, at the international conference WSDM2010.
Title: Corroborating Information from Disagreeing Views
Authors: Alban Galland, Serge Abiteboul, Pierre Senellart, Amelie Marian
Abstract: We consider a set of views stating possibly conflicting facts. Negative facts in the views may come, e.g., from functional dependencies in the underlying database schema. We want to predict the truth values of the facts. Beyond simple methods such as voting (typically rather accurate), we explore techniques based on ``corroboration'', i.e., taking into account trust in the views. We introduce three fixpoint algorithms corresponding to different levels of complexity of an underlying probabilistic model. They all estimate both truth values of facts and trust in the views. We present experimental studies on synthetic and real-world data. This analysis illustrates how and in which context these methods improve corroboration results over baseline methods. We believe that corroboration can serve in a wide range of applications such as source selection in the semantic Web, data quality assessment or semantic annotation cleaning in social networks. This work sets the bases for a wide range of techniques for solving these more complex problems.
Slides of the presentation: Corrob-WSDM-02-2010
Video of the presentation: VideoLecturers
Program and datasets: page of the project
Wednesday, February 3. 2010
I gave a presentation about my work on social navigation, February 3rd, 2010, at the poster session of the SSM2010 workshop.
Title: Birds of a tag flock together
Authors: Serge Abiteboul, Sihem Amer-Yahia, Alban Galland, Pierre Senellart, Amelie Marian
Tuesday, December 1. 2009
I gave a presentation about my work on access control on social networks, December 1st, 2009, at the SMIS working group.
Title: Access control on Social Network
Abstract: Social network systems allow users to share private data with privileged other users of the systems, called friends. The development of these systems is leading to issues in access control. In particular, the management of data in centralized providers such as Facebook raises privacy concerns and suggests distributed solutions. Even if there have been some works in that direction, we believe that a general model is still missing. We propose such a logical model based on the distribution of data, access control and keys. We show how different systems can be modeled using it. We discuss these models with respect to some general desirable properties. Finally, we discuss how to express access control with a richer semantics and mention consequences on tractability.
Slides of the presentation:: SN-PRISMSeminar-12-2009
Saturday, October 24. 2009
I gave a presentation about my work on corroboration, October 24th, 2009, at the french conference BDA2009.
Title: Corroboration de vues discordantes fondée sur la confiance,
Authors: Alban Galland, Serge Abiteboul, Pierre Senellart, Amelie Marian
Abstract: Cet article traite de la corroboration d'informations, dans le contexte de vues exprimant des opinions sur des faits de façon éventuellement contradictoire. Il s'agit de prédire si un fait est vrai ou faux. Des méthodes d'agrégation simples comme le vote donnent déjà de bons résultats, mais nous présentons dans cet article des algorithmes qui tiennent compte de la confiance dans les vues pour améliorer les prédictions. Les trois algorithmes proposés sont des algorithmes de point fixe correspondant à différents niveaux de complexité du modèle probabiliste sous-jacent. Ils estiment à la fois la valeur de vérité des faits et la confiance dans les vues. Cet article présente une étude expérimentale sur des données synthétiques et réelles. Ces expériences montrent dans quelle mesure et dans quel contexte nos algorithmes peuvent améliorer les résultats par rapport au vote. La corroboration apparaît naturellement dans un grand nombre d'applications, comme la sélection de sources dans le Web sémantique, les tests de qualité de données ou le nettoyage d'annotations sémantiques dans les réseaux sociaux. Ce travail pose donc les bases de techniques plus complexes pour traiter les problèmes précédents.
Slides of the presentation: Corrob-BDA-10-2009
Friday, October 9. 2009
Presentation at GEMO annual seminar 2009 Posted by Alban Galland in Presentations at 09:00
Friday, February 6. 2009
Presentation at GEMO seminar Posted by Alban Galland in Presentations at 14:00
I gave a presentation about my work on corroboration, February 6th, 2009, at the GEMO seminar
Title: On Corroboration and Trust
Abtract: On a decentralized network such as the web or a P2P network, the full information is the combination of the information given by all the nodes, which could be conflicting. Considering a set of views stating possibly conflicting facts and potential functional dependencies between facts, one could try to predict which facts are true and which ones are false. A natural idea, corroboration, is to take into account trust in the views. Without further information, trust is then recursively defined as how accurate a source is on the data. This talk will present our works on this problem, introducing a general probabilistic model and different corroboration algorithms and discussing experiments on synthetic and real-world data. Fayçal Hamdi: "Alignment-based Partitioning of Large-scale Ontologies" Ontology alignment is an important task for information integration systems that can make different resources, described by various and heterogeneous ontologies, interoperate. However very large ontologies have been built in some domains such as medicine or agronomy and the challenge now lays in scaling up alignment techniques that often perform complex tasks. In this talk, we will briefly present the state-of-the-art of partitioning methods and their limits when used to align ontologies. Then we will present two partitioning methods that take the alignment objective into account during the partitioning process. We will show how these methods transform the two ontologies to be aligned into two sets of blocks of a limited size and discuss the location of the elements of the two ontologies that have to be aligned according to each of them. Finally we will present some experimental results performed by the two methods on various pairs of ontologies.
Slides of the presentation:Corrob-GEMOSeminar-02-2009
« previous page (Page 1 of 1, totaling 14 entries) next page »
Who is the author ?
My name is Alban Galland. You will find more personal information on my web site.
Last entry: 2011-09-09 11:26
22 entries written
0 comments have been made