In this article I’ll make the asked explanation on three of the topics we have spoken about recently:
This first topic I’m going to talk about is the “Humaine” or “Human-machine interaction Network on emotions” one of the current projets of the German Research Center for Artificial Intelligence.
Humaine aims to lay the foundations for European development of systems that can register, model and influence human emotional and emotion-related states and processes – ‘emotion-oriented systems’. Such systems may be central to future interfaces, but their conceptual underpinnings are not sufficiently advanced to be sure of their real potential or the best way to develop them.
One of the reasons is that relevant knowledge is dispersed across many disciplines. Humaine brings together leading experts from the key disciplines in a programme designed to achieve intellectual integration. It identifies six thematic areas that cut across traditional groupings and offer a framework for an appropriate division of labour – theory of emotion; signal/sign interfaces; the structure of emotionally coloured interactions; emotion in cognition and action; emotion in communication and persuasion; and usability of emotion-oriented systems. Teams linked to each area will run a workshop in it and carry out joint research to define an exemplar embodying guiding principles for future work in their area.
The second topic on which I am going to focus is the one called “Whiteboard“; a completed project of the same research centre. This project focused on the “Multilevel annotation for dimamic free text processing”.
The project aimed at designing, implementing, investigating and evaluating a new system architecture that facilitated the combination of different language technologies for a range of practical applications. Language technologies offered numerous means for a partial analysis of texts that could be employed for information retrieval, information extraction, language checking, and many other applications. Processing methods and tools differed along several dimensions, e.g., wrt. levels of linguistic description, depth of analysis, or the way knowledge of language is derived (linguistically or statistically).
Methods often overlaped in their functionality but differed in their strengths and weaknesses. Finding optimal combinations of heterogeneous techniques and processing components was one of the most difficult tasks in language processing – the challenge of the Whiteboard project. The novel architecture to be developed and explored in Whiteboard was based on the concept of an annotated text. The different LT components enriched an XML. Each component can exploit or disregard previously assigned annotations. Its architecture had a single shared data structure, which at the same time was the input, throughput, and output of the system. The envisaged architecture permited the pragmatic combination of different processing approaches, most notably novel ways of the combination of shallow and deep methods.
Finally, the last topic I had picked to focus on is the “Neca” or “The net environment for embodied emotional conversational agents”; one of the previous projects of the Austrian Research Institute for Artificial Intelligence.
The objective of the NECA project was to develop a new generation of mixed multi-user / multi-agent virtual spaces populated by affective conversational agents. The agents are be able to express themselves through synchronised emotional speech and non-verbal expression, generated from an abstract representation. This is the first time that such expressive capabilities are featured in Internet applications. The agents’ usefulness were evaluated in two concrete application scenarios. From a technical point of view, the NECA platform provides a confederation of dedicated components including an affective reasoner, co-ordinated generation of verbal and nonverbal aspects of communication, and emotional speech synthesis, thus providing a basis for the development of new Internet applications with emotional agents.
Sources:
http://www.dfki.de/pas/f2w.cgi?ltp/humaine-e
http://www.dfki.de/pas/f2w.cgi?ltc/whiteboard-e
http://www.ofai.at/research/nlu/projects/nlproject_neca.html
Etiquetas: Human LanguageTechnologies. (Q 2)
