Projects

Generelles Ziel des Projekts „Automated Coding and Categorizing of Innovation Areas“ (ACCIA) war es, ein intelligentes automatisiertes System zu schaffen, das Belegstellen für problembezogene, innovationsrelevante Äußerungen, welche aus unterschiedlichen Online-Quellen extrahiert wurden, identifizieren, analysieren und kategorisieren kann. Durch die Einbindung von Verfahren aus den Bereichen Textverarbeitung, Document Clustering und Document Classification konnte ein Prozessmodell erarbeitet werden, dass eine bisher rein manuell durchgeführte Innovationsfeldanalyse in ein automatisiertes Modell überführt, in dem die manuelle, expertinnengetriebene Analyse mit automatischen, computerlinguistisch gestützten Verfahren verschränkt wird.

The central aim of the study is the realistic presentation of the potential of AAL Robotics, based on the analysis of parameters drawn from user needs, technical readiness, and existing business models. Instead of compiling yet another collection of individual solutions and projects, the PotenziAAL Study will aggregate existing knowledge gained from analysis of secondary sources with knowledge generated from primary sources such as expert interviews, a workshop and user focus groups in order to achieve a comprehensive picture of the state of the art in AAL Robotics and its future potential. The study will develop categories and criteria in order to foster exact characterization, comparability and quality assurance in AAL Robotics.

The aim of the project is the development of a virtual agent that interactively supports older persons in the narration of autobiographic stories, from listener feedback to raw transcription.

TAPPING FRIEND is an interactive science game for everyone between 6 and 99 years and is aimed to provide a playful experience of synchronization and cooperation between one or two humans and a virtual partner – the maestro. The players tap in time with the maestro on little drums and the system provides immediate feedback on their synchronization success by showing their taps relative to the maestro’s taps and by counting the taps that were on time and the taps that were too early or too late. The aim of the game is to achieve as many on-time taps as possible.

The FemSMA project aims at the development of automatic methods to determine the gender of the author of a given user-created social media posting (e.g. Tweets, blog or internet forum entries). The approach is motivated by recent developments in the social sciences, as well as market and opinion research, which increasingly consider social media platforms as a valuable source of data, especially for trend and sentiment analyses. Current instruments for social media analysis are incapable of providing gender-aware information regarding the topics, products or (political) programmes under investigation. This deficit, as well as the associated lack of orientation towards active demand, will be countered effectively in FemSMA.

he so-called “hubness” phenomenon is a general problem of machine learning in high dimensional data spaces. Hubs are data points which keep appearing unwontedly often in nearest neighbor lists of many other data points. This effect is particularly problematic in algorithms for similarity search, as the same “similar” objects are found over and over again. But it has also adverse effects for the many machine learning algorithms that make use of distance information. The effect has been shown to be a natural consequence of high dimensionality and as such is yet another aspect of the curse of dimensionality. The main goal of this project was to conduct an in-depth study of the hubness problem in the context of MIR.

CYBEREMOTIONS is a European IP (Large Scale Integrating Project) which focuses on the role of collective emotions in creating, forming and breaking-up e-communities. Understanding these phenomena is important in view of the growing role of ICT-mediated social interactions and some specific features of e-communities. The challenge of this interdisciplinary project is to combine psychological models of emotional interactions and algorithmic methods for detection and classification of human emotions in the Internet with probabilistic models of complex systems and data driven simulations based on heterogeneous emotionally-reacting agents.

MIReS was an EU FP7-funded 18-month project focusing on the future impact of music technology research on academia and industry. The project created a research roadmap of MIR field, by expanding its context and addressing challenges such as multimodal information, multiculturalism and multidisciplinarity.

Language varieties are a primary means of expressing a person's social affiliation and identity. Hence, computer systems that can adapt to the user by displaying a familiar socio-cultural identity are expected to raise the acceptance within certain contexts and target groups dramatically. However, current systems are far from achieving the fidelity required for realization of these benefits. For example, promising early results have been obtained in the context of speech synthesis through localized pronunciation, but it is clear that language variety is a multi-faceted concept that involves deviations from standard language on various linguistic levels. Our goal is to develop algorithmic methods that are capable of capturing and reproducing all major idiosyncrasies displayed by a language variety, be they syntactic, lexical or phonetic in nature. Conceptually, part of this task can be understood as a machine translation problem, which is, however, characterized by unique properties.