Luca Vassena

The Context-Aware Browser

The typical scenario of a user seeking information on the Web requires significant effort to get the desired information. In a world where information is essential, it can be crucial for users to get the desired information quickly even when they are away from their desktop computers. The Context-Aware Browser for mobile devices senses the surrounding environment, infers the users current context, and proactively searches for and activates relevant Web documents and applications.

Retrieval of context-aware applications on mobile devices: how to evaluate?

This paper discusses the issue of evaluation of context-aware retrieval applications. We begin by describing MoBe, a specific architecture that allows automatic download and execution of context-aware applications on mobile devices. In MoBe, the most relevant applications are selected by matching context and application descriptors. Since several alternatives for descriptors implementation exist, it is important to compare their effectiveness. To this aim, we develop a TREC-like benchmark, in which the collection is made up by a set of application descriptors and the topics are made up by context descriptors. We then use the benchmark to evaluate the effectiveness of different descriptors components, and of structured and unstructured (i.e., free text) data, finding results that are somehow useful for future MoBe development. We also discuss the issue of the evaluation methodology for highly interactive and novel applications like context-aware retrieval systems, and MoBe in particular.

A social approach to context-aware retrieval

We present a general purpose solution to Web content and services perusal by means of mobile devices, named Social Context-Aware Browser. This is a novel approach for information access based on users’ context, that exploits social and collaborative models to overtake the limits of the existing solutions. Instead of relying on a pool of experts and on a rigid categorization, as it is usually done in the context-aware field, our solution allows the crowd of users to model, control, and manage the contextual knowledge through collaboration and participation. To have a dynamic and user-tailored context representation, and to enhance the process of retrieval based on users’ actual situation, the community of users is encouraged to define the contexts of interest, to share, use, and discuss them, and to associate context to content and resources (Web pages, services, applications, etc.). This paper provides an overall presentation of our solution, describing the idea, the implementation, and the evaluation through a benchmark based methodology.

AI Techniques in a Context-Aware Ubiquitous Environment

Nowadays, the mobile computing paradigm and the widespread diffusion of mobile devices are quickly changing and replacing many common assumptions about software architectures and interaction/communication models. The environment, in particular, or more generally, the so-called user context is claiming a central role in everyday’s use of cellular phones, PDAs, etc. This is due to the huge amount of data “suggested” by the surrounding environment that can be helpful in many common tasks. For instance, the current context can help a search engine to refine the set of results in a useful way, providing the user with a more suitable and exploitable information. Moreover, we can take full advantage of this new data source by “pushing” active contents towards mobile devices, empowering the latter with new features (e.g., applications) that can allow the user to fruitfully interact with the current context. Following this vision, mobile devices become dynamic self-adapting tools, according to the user needs and the possibilities offered by the environment. The present work proposes MoBe: an approach for providing a basic infrastructure for pervasive context-aware applications on mobile devices, in which AI techniques (namely a principled combination of rule-based systems, Bayesian networks and ontologies) are applied to context inference. The aim is to devise a general inferential framework to make easier the development of context-aware applications by integrating the information coming from physical and logical sensors (e.g., position, agenda) and reasoning about this information in order to infer new and more abstract contexts.

Collaborative annotation for context-aware retrieval

We discuss how collaborative annotations can be exploited to simplify and improve the management of context and resources in the context-aware retrieval field. We apply this approach to our Context Aware Browser, a general purpose solution to Web content perusal by means of mobile devices, based on the users context. Instead of relying on a pool of experts and on a rigid categorization, as it is usually done in the context-aware field, our solution allows the crowd of users to model, control and manage the contextual knowledge through collaboration and participation. We propose two models and we outline an example of application.

Context-Aware Browser

We propose the Context-Aware Browser (CAB), a new approach to context-aware Web content perusal by means of mobile devices. The CAB is a Web browser running on mobile devices, exploiting artificial intelligence techniques and mixing several ingredients. The novelty of the proposed approach is threefold. First, our proposal allows a different perspective in the development of Web contents, which should be conceived in a context-dependent way. Our framework makes it possible to develop information and web applications specifically for a given environment, which will become capable to transmit context information, and ad hoc, context-based applications to whom will enter that context. Second, we exploit several informations provided by the surrounding environment and not only the location in order to carry out a more refined search of web contents and applications. Third, the CAB approach involves an innovative two stage retrieval and a filtering process, that minimizes privacy issues.

Where do you Roll Today? Trajectory Prediction by SpaceRank and Physics Models

Pre-destination, the prediction of a user’s future destination, is recently gaining interest and importance in location-aware, ubiquitous, and mobile computing. An increasing amount of data related to position of people is becoming available because people usually take their mobile devices (phones, smart-phones, PDAs, etc.) with them. We propose to mine these data to derive the importance of the single locations in an area of interest, given by either a single user or a community. Then we use the importance of locations as basis for our approach to pre-destination, where well-known physics models (namely gravitation and electrical force) are exploited to estimate users trajectories and future destinations.

Evaluating Mobile Proactive Context-Aware Retrieval: An Incremental Benchmark

We present the evaluation of a novel application for Web content perusal by means of context-aware mobile devices that proactively query an external search engine. To this aim, we develop a TREC-like benchmark and we use it to evaluate different strategies for automatic query construction on the basis of users current context. We discuss both the methodology and the results.