THE RASCALLI WORKPACKAGES
WP 1 Cognitive Architecture and Modelling
focuses on building the core cognitive architecture of Rascalli. It will be a hybrid (symbolic/connectionist) general cognitive architecture providing the main structures and processes needed for reasoning, learning, communication, perception and action thus integrating them in a cognitively plausible way and allowing them to interact with each other in a dynamic way.
WP 2 Perception and Action
specifies the basic perception and action mechanisms of Rascalli, the links among them, and generic mechanisms for their extension during the lifetime of a Rascalla/o.
WP 3 Reasoning
The objective here is to endow Rascalli with reasoning capacities. In particular, Rascalli need be able to make inferences based on the knowledge they have and the perceived information and consequently plan their actions.
WP 4 Learning
While in WP1 we have addressed how memory is structured and how the pieces of knowledge interact, in WP4 we address topics regarding how the memory is filled, i.e. what the specific content of a Rascalla/o's memory is, and how it is represented technically. In particular, we pursue two strands of learning: a low level bottom up approach where the Rascallo's memory is filled and structured solely by user input, and a highly theory driven top down approach based on the cognitive model developed in WP1.
WP 5 Communication
In this workpackage we deal with the communication and interaction aspects of Rascalli. We distinguish two kinds of interaction 1) Rascalli interact with other Rascalli in the Rascalli society/community: 2) Rascalli interact with the user (communication between Rascalli and their users comprises uni-directional and dialogue aspects).
WP 6 Integration: Platform and Application Scenarios
WP 7 Evaluation
We distinguish two strands of evaluation that will accompany the development and implementation phases of the RASCALLI platform and demonstrators. These are: evaluation of learning and progress of Rascalli, and the assessment of usability of the Rascalli applications. While results from the former are expected to influence the development of the core of the RASCALLI system, results of the latter will inform about the attitudes of users towards the specific kind of digital presence featured in RASCALLI, and the overall satisfaction or dissatisfaction of using the applications.
WP 8 Dissemination and Exploitation
A central goal of the RASCALLI dissemination and exploitation will be to ensure that the research community will be able to share the results of the project fully. This is supposed to be achieved through scientific publication and via licensing of RASCALLI software components. We expect the project to produce results at least in the areas of cognitive artificial systems, natural language processing and human-computer interfaces.
RASCALLI is supported by the European Commission Cognitive Systems
RASCALLI develops a new type of personalized cognitive agents, the
Rascalli, that live and learn on the Internet.
combine Internet-based perception, action, reasoning, learning, and
Rascalli come into existence by creation
through the user. The users not only create their Rascalli but also
train them to fulfil specific tasks, such as be experts in a quiz
game or assist the user in a music portal.