[The original web site of the ConCreTe project is down. This stub has been created to provide basic information about it.]
The ConCreTe project was a 3 year Collaborative project funded by the European Comission FETOpen FP7 program. It took place from October 2013 to September 2016.
Some basic information about the project has been compiled below.
Original Project Abstract
ConCrete aims to study conceptual creativity in humans and machines. Hierarchical memory representations and
techniques for conceptual blending are implemented in context of a cognitive architecture of creativity based on
information theoretic measures. ConCreTe serves a long-term vision of computer systems that can behave in
ways comparable with human creativity, autonomously and interactively, with better interaction between human
and machine, better autonomous systems in general, and possibly creativity of new kinds, not yet exhibited by
humans. We anticipate on-line creative learning environments, to teach or support creative pursuits and promote
creativity in humans. We anticipate immersive gaming and edutainment systems that respond creatively to users’
actions. We anticipate reasoning systems that can propose new technology not intended by their designers.
This becomes possible with computationally-creative reasoning when necessary domain knowledge is made
available. We use Semantic Web technology to avoid the bottleneck of domain modeling, so creative reasoning
can be ready in advance. We focus on mechanisms for generating examples in the creative domain from a
learned model, and mechanisms for evaluating generated examples according to novelty and value. We develop
AI methodology for creative systems, to exploit the potential of creative computational resources for society. We
develop computer systems to simulate conceptualisation by study of deliberately guided methods. We develop
computer systems that can conceptualise new meaning in terms of, but not restricted by, its existing knowledge.
We develop and implement a cognitive architecture that simulates human creativity, study it as a creative entity
in its own right, and behaviourally and neuroscientifically as a model of human creativity. We develop new
evaluation methods for computational creativity founded in behavioural study and user responses of software
distributed by our exploitation partner.
List of partners
- Queen Mary and Westfield College (London, United Kingdom) [Coordinator]
- University of Helsinki (Finland)
- University of Twente (Netherlands)
- Universidade de Coimbra (Portugal)
- Institut Jozef Stefan (Lubljana, Slovenia)
- Universidad Complutense de Madrid (Spain)
- Chatterbox Analytics Ltd. (London, United Kingdom)
Work packages
- WP 1 Cognitive Architecture
- WP 2 Conceptual Representations
- WP 3 Concept Creation
- WP 4 Neuroscience of Concept Formation
- WP 5 Applications
- WP 6 Integration of technology and data
- WP 7 Exploitation
- WP 8 Evaluation
- WP 9 Dissemination
- WP 10 Project Management
Description of UCM Contribution
Within the ConCreTe project, UCM was leader of Work Package 5 on Applications.
This resulted in the development of a number of concept to text generation tools, the Grafeno library for constructing graphs of concepts from text, applications for text summarization, and refinements to pre-existing applications for the automatic generation of poetry. Details on these research efforts may be found under the publications section of this web site.