The shape of new technologies is not set in advance. But how do different social circumstances and different groups of actors influence the forms and outcomes of scientific and technical innovation? What are the roles of governance institutions and policies, industry structures and strategies, discursive practices and imaginaries? How can processes of high-tech development be modulated to guide the progress of new technologies in responsive and responsible ways − and how might these approaches be adapted to different conditions across the world? The multidisciplinary contributions of this volume from experts in the social sciences and humanities, as well as practitioners in industry and consultancy, try to answer these questions by addressing nanotechnology and other emerging technoscientific fields, such as geoengineering and synthetic biology. Shaping Emerging Technologies constitutes the fourth volume showcasing the research activities of the vibrant, multidisciplinary community of S.NET, the Society for the Study of Nanoscience and Emerging Technologies.

Natural Language Generation (NLG) systems in interactive settings often face a multitude of choices, given that the communicative effect of each utterance they generate depends crucially on the interplay between its situational circumstances, addressee and interaction history. This is particularly true in interactive and situated settings. Traditionally, the generation process has been divided into distinct stages of decision making, e.g. content selection, utterance planning and surface realisation. However, this sequential model does not account for the interdependencies that exist among these stages, which in practice can become manifest in inefficient, ineffective communication and an increased cognitive load for the user.This book presents a joint optimisation framework for NLG in dialogue that is based on Hierarchical Reinforcement Learning and learns the best utterance for a context by optimisation through trial and error. The joint model considers decisions at different NLG stages interdependently and produces more context-sensitive utterances than a model that considers decisions in isolation. To enhance the human-likeness of the presented framework, we integrate graphical models, trained from human data, as generation space models for natural surface realisation. The proposed technique is evaluated in a study with human participants and confirms that the hierarchical learner is able to learn an adaptive policy that leads to smooth and successful interactions. Results also suggest that a joint optimisation leads to substantially higher user satisfaction and task success and is better perceived by human users than its isolated counterpart.

This work is about human beings and Geographic Information Systems (GIS). More specifically, it introduces methods and techniques to foster natural interaction among the two; an interaction that becomes continuously more important as technological improvements bring GIS technology within general public‘s reach. A glaring example concerns, for instance, Volunteered Geographic Information projects that relies on the intervention of volunteers − i.e., not necessarily GIS experts − to build and share freely accessible spatial datasets.According to the current state of the art, there is a fundamental representational gap that, by hindering the coveted natural interaction, prevents non-experts from fully exploiting GIS power: if on the one hand human beings prefer to represent and reason about space in a qualitative manner, on the other hand, GIS mainly resort to quantitative representations. To bridge such a gap, a new typology of queries is presented in this work: Qualitative Spatial Relation Queries (QSRQ) involve spatial predicates defined in mathematical structures known as qualitative calculi. Such predicates provide both a computer-readable definition and a cognitively-adequate semantics to represent and reason about given spatial aspects − e.g., topology or direction. An extensible theoretical and practical framework is presented that allows for enabling QSRQ in GIS. Moreover, a classification of such queries is provided based on their solving complexity and the focus is shifted on the discussion of a new type of spatial access methods − Qualitative Spatial Access Methods (QSAM) − designed for resolving the hardest QSRQ category, termed Qualitative Spatial Configuration Queries. QSAM draw upon a qualitative spatial relation reduction/reconstruction paradigm in order to face the exponential explosion in the amount of information bits introduced by the qualitative formalism. Four different types of QSAM are presented, evaluated, and compared with respect to the space-time performance that they allow for reaching with both synthetic and real spatial datasets.

Extreme events cause drastic transformations in human settlements; as a consequence, descriptions of the struck environment, based on „quantitative“ spatial data collected before the event, becomes suddenly unreliable. On the other hand, time critical actions taken for responding to extreme events require up-to-date spatial information. Traditional methods for spatial data collection are not able to provide updated information rapidly enough, calling for the development of new data collection methods. Reports provided by actors involved in the response operations can be considered as a source of „qualitative“ spatial information. There has been considerable research on spatial representations and analysis based on metric measurements that lead to the development of Geographic Information Systems (GIS). A different field of research focuses on developing representation and reasoning techniques for dealing with qualitative spatial information that describes qualities and relations holding among objects in space abstracting from quantitative representations. Relatively little work exists on finding techniques to combine the different representation and reasoning methodologies. A hybrid reasoning system that combines techniques developed separately for qualitative spatial reasoning and quantitative data analysis is developed in this work. The integrated approach performs different reasoning operations to retrieve new information pieces from a given set. Derived spatial information can be either qualitative or quantitative and complements the given one in supporting tasks related to visualization (e.g., maps and graphics) or querying (e.g., support for decision making). Three main system functionalities are investigated: The first, called quantification, allows for translating qualitative information into quantitative descriptions. The second, called qualification, aims at translating quantitative information into qualitative relations. Finally, the third allows for performing inference operations with information given partly qualitatively and partly quantitatively for boosting the spatial knowledge the system is able to produce.

Nachdem in der medizinischen Forschung über viele Jahre die Auswahl oder Programmierung einzelner IT-Anwendungen im Vordergrund stand, bedarf es heute eines übergreifenden Konzeptes für die Integration verschiedener Infrastrukturkomponenten in die IT-Gesamtlösung von Forschungseinrichtungen und -verbünden. Solche Komponenten, die heute vorliegen, reichen von Datenbanken und Software-Anwendungen über die Ansteuerung von Robotik bis zur Kommunikation mit mobilen Geräten. Notwendige Voraussetzung für die Integration und Weiterentwicklung ist eine kontinuierliche Bestands- und Bedarfsanalyse, die mit dem ersten Bericht des IT-Reviewing Boards der TMF beginnt. Die künftig jährlich erscheinende Publikation wird auch die Grundlage sein für konsentierte Handlungsempfehlungen als Hilfestellung im politischen Diskurs. Die Bundesregierung hat in ihrem Koalitionsvertrag von November 2013 darauf hingewiesen, dass die Gesundheitsforschung auch weiterhin eine herausgehobene Stellung einnehmen wird. Mit den Deutschen Zentren der Gesundheitsforschung und vergleichbaren Forschungsverbünden wird der Ausgestaltung der IT-Infrastrukturen in der patienten–orientierten Forschung auch weiterhin eine große Bedeutung zukommen. Schwerpunkt des ersten Berichtes sind die Primärdatenquellen. Dabei werden klinische Studien, Register und Kohorten, Bildverarbeitungssysteme, Biobanken und mobile IT-Werkzeuge sowie das verbindende Thema Identitätsmanagement in den Blick genommen. Die Bestandsaufnahme zeigt, dass für diese Bereiche bereits sehr gute Lösungen existieren.

Recent years have witnessed a remarkable proliferation of activities related to nanoscience and emerging technologies. Exploratory research has expanded to high-tech innovation, national investment has expanded to global collaboration, and interest in the social, ethical, and philosophical dimensions of the molecular sciences has expanded to issues of policy, education, and governance. Little by Little: Expansions of Nanoscience and Emerging Technologies traces these developments by attending to the diversification of cultural meanings, the engagement of various publics, the transformation of innovation regimes, and the broadening of responsibilities around new technologies. Bringing together contributions and perspectives from experts in the social sciences, the humanities, and the natural sciences, along with incisive reports from experts in public outreach, science education, and technology policy, Little by Little showcases ongoing research activities of the vibrant, multidisciplinary community of S.NET, the Society for the Study of Nanoscience and Emerging Technologies.