David Coquil

Extracting User Interests from Search Query Logs: A Clustering Approach

This paper proposes to enhance search query log analysis by taking into account the semantic properties of query terms. We first describe a method for extracting a global semantic representation of a search query log and then show how we can use it to semantically extract the user interests. The global representation is composed of a taxonomy that organizes query terms based on generalization/specialization (“is a”) semantic relations and of a function to measure the semantic distance between terms. We then define a query terms clustering algorithm that is applied to the log representation to extract user interests. The evaluation has been done on large real-life logs of a popular search engine.

Integrating similarity-based queries in image DBMSs

Until recently, issues in image retrieval have been handled in DBMSs and in computer vision as separate research works. Nowadays, the trend is towards integrating the two approaches (content- and metadata-based) for multi-criteria image retrieval. However, most existing works and proposals in this domain lack a formal framework to deal with a multi-criteria query. In this paper, we introduce a formal framework to address this subject of image retrieval under an ORDBMS model. We first propose an image data repository model interoperable with current popular standards. Then, we present an algebraic formalism for content-based operators on image database. We study the properties of these operators and discuss query optimization issues. To demonstrate the use of our algebra, we implemented an extension of our prototype called EMIMS. Experimental evaluations on our proposed query optimization techniques used in EMIMS are presented here.

Interactive TV Services on Mobile Devices

The recent digitalization of television creates new opportunities for enhancing the viewers experience with interactivity. Interactive TV (iTV) is often solely understood as the ability to change a programs storyline. Besides this interpretation, iTV in general means providing some kind of interactive add-ons or TV-related content and services. For example, the viewer might participate in a game show, gather additional information on news topics, or buy a product presented in a commercial. The combination of digital TV and modern set-top boxes facilitates the deployment of such innovative services. In this context, we developed a prototype platform that uses mobile devices to support multiuser and personalized access for iTV services. The mobile devices connect to the set-top box with ad hoc mechanisms over an existing home network, enabling inexperienced users to access and use the services without having to worry about configuration.

Efficient Execution of Service Composition for Content Adaptation in Pervasive Computing

Multimedia content adaptation has been proved to be an effective mechanism to mitigate the problem of devices and networks heterogeneity and constraints in pervasive computing environments. Moreover, it enables to deliver data taking into consideration the users preferences and the context of his/her environment. In this paper, we present an algorithm for service composition and protocols for executing service composition plan. Both the algorithm and the protocols are implemented in our distributed content adaptation framework (DCAF) which provides a service-based content adaptation architecture. Finally, a performance evaluation of the algorithm and the protocols is presented.

Improving SURF Image Matching Using Supervised Learning

Key points-based image matching algorithms have proven very successful in recent years. However, their execution time makes them unsuitable for online applications. Indeed, identifying similar key points requires comparing a large number of high dimensional descriptor vectors. Previous work has shown that matching could be still accurately performed when only considering a few highly significant key points. In this paper, we investigate reducing the number of generated SURF features to speed up image matching while maintaining the matching recall at a high level. We propose a machine learning approach that uses a binary classifier to identify key points that are useful for the matching process. Furthermore, we compare the proposed approach to another method for key point pruning based on saliency maps. The two approaches are evaluated using ground truth datasets. The evaluation shows that the proposed classification-based approach outperforms the adversary in terms of the trade-off between the matching recall and the percentage of reduced key points. Additionally, the evaluation demonstrates the ability of the proposed approach of effectively reducing the matching runtime.

Evaluating the Robustness of Publish/Subscribe Systems

In recent years, publish/subscribe (pub/sub) systems have evolved as serious candidates for the implementation of multicast communication. However, to date they have not been properly analysed with respect to the critical dimension of robustness. Indeed, the robustness evaluations found in the literature only cover a subset of the types of failure situations and generally lack a comprehensive architectural classification. In this context, this paper introduces the necessary foundations for systematic evaluation of the robustness of pub/sub system: a comprehensive classification scheme and a complete taxonomy of failures. We argue that both robustness evaluation of pub/sub systems and development methods of robust pub/sub systems should take rational behavior into account. We demonstrate this point through an experimental study.

CReaM: User-Centric Replication Model for Mobile Environments

Data replication can improve data availability in mobile networks; but due to typical resource limitations in such environments, specific replication mechanisms are needed. In this paper, we propose CReaM, user-Centric REplicAtion Model for mobile environment. This model puts users at the centre by letting them determine the amount of resources they are willing to share. CReaM is suitable for dynamic environments where each node needs a certain level of decision autonomy. In this paper, we present CReaMs general principles and focus on one of its core function, which is its autonomic behavior that generates replication requests based on resources monitoring and user settings.

SimAttack: private web search under fire

Web Search engines have become an indispensable online service to retrieve content on the Internet. However, using search engines raises serious privacy issues as the latter gather large amounts of data about individuals through their search queries. Two main techniques have been proposed to privately query search engines. A first category of approaches, called unlinkability, aims at disassociating the query and the identity of its requester. A second category of approaches, called indistinguishability, aims at hiding user’s queries or user’s interests by either obfuscating user’s queries, or forging new fake queries. This paper presents a study of the level of protection offered by three popular solutions: Tor-based, TrackMeNot, and GooPIR. For this purpose, we present an efficient and scalable attack – SimAttack – leveraging a similarity metric to capture the distance between preliminary information about the users (i.e., history of query) and a new query. SimAttack de-anonymizes up to 36.7 % of queries protected by an unlinkability solution (i.e., Tor-based), and identifies up to 45.3 and 51.6 % of queries protected by indistinguishability solutions (i.e., TrackMeNot and GooPIR, respectively). In addition, SimAttack de-anonymizes 6.7 % more queries than state-of-the-art attacks and dramatically improves the performance of the attack on TrackMeNot by 23.6 %, while retaining an execution time faster by two orders of magnitude.

Semantic video content annotation at the object level

A vital prerequisite for fine-grained video content processing (indexing, querying, retrieval, adaptation, etc.) is the production of accurate metadata describing its structure and semantics. Several annotation tools were presented in the literature generating metadata at different granularities (i.e. scenes, shots, frames, objects). These tools have a number of limitations with respect to the annotation of objects. Though they provide functionalities to localize and annotate an object in a frame, the propagation of this information in the next frames still requires human intervention. Furthermore, they are based on video models that lack expressiveness along the spatial and semantic dimensions. To address these shortcomings, we propose the Semantic Video Content Annotation Tool (SVCAT) for structural and high-level semantic annotation. SVCAT is a semi-automatic annotation tool compliant with the MPEG-7 standard, which produces metadata according to an object-based video content model described in this paper. In particular, the novelty of SVCAT lies in its automatic propagation of the object localization and description metadata realized by tracking their contour through the video, thus drastically alleviating the task of the annotator. Experimental results show that SVCAT provides accurate metadata to object-based applications, particularly exact contours of multiple deformable objects.

On reliability in publish/subscribe systems: a survey

In recent years, publish/subscribe (pub/sub) systems have evolved as serious candidate for the implementation of group communication for distributed systems. However, they have not yet been properly evaluated with respect to the critical dimension of reliability. In particular, existing surveys do not consider the case of selfish behaviour of the nodes. This paper aims to bridge this gap by reviewing 25 pub/sub systems of the literature in order to evaluate their Byzantine, altruistic, rational tolerance capabilities, i.e. how well they are able to prevent or tolerate Byzantine failures as well as rational (selfish) behaviour. To this end, we also provide the foundations required for systematically evaluating such capabilities: we define a comprehensive classification scheme of pub/sub systems and a taxonomy of elementary failures. The latter covers not only failures that may be caused by arbitrary faults but also those arising from selfish behaviour of the nodes.