Alle Publikationen
- <<
- <
- 1
2018
-
(2018) : A Concept for Productivity Tracking based on Collaborative Interactive Learning Techniques In: ARCS: ARCS Workshop 2018; 31th International Conference on Architecture of Computing Systems: Berlin, Germany: Axel Springer SE, S. 1-8. Online verfügbar unter https://ieeexplore.ieee.org/document/8385431/
Abstract: The academic success of individual students differs widely and it depends on various factors, ranging from financial to social and to health aspects. In this article, we propose a concept for a novel productivity tracking system that provides the basis for a self-assessment of academic behaviour and that can be used by students to support their academic success. The development of such a system requires interdisciplinary efforts, most of them located in the field of collaborative interactive learning (CIL) that is grounded on a socio-technical system perspective. The system is interactive since it is based on bidirectional communication, collaborative in the sense that it uses students, other students, and external sources such as the Internet for generation of knowledge, and learning in the sense that it continuously and autonomously acquires knowledge. It is further self-organised as it decides about interaction partners and self-adaptive in terms of modifying its behaviour according to changing conditions.
-
(2018): The knowledge level in cognitive architectures. Current limitations and possible developments. In: COGNITIVE SYSTEMS RESEARCH 48, S. 39-55. DOI: 10.1016/j.cogsys.2017.05.001
Abstract: In this paper we identify and characterize an analysis of two problematic aspects affecting the representational level of cognitive architectures (CAs), namely: the limited size and the homogeneous typology of the encoded and processed knowledge. We argue that such aspects may constitute not only a technological problem that, in our opinion, should be addressed in order to build artificial agents able to exhibit intelligent behaviors in general scenarios, but also an epistemological one, since they limit the plausibility of the comparison of the CAs' knowledge representation and processing mechanisms with those executed by humans in their everyday activities. In the final part of the paper further directions of research will be explored, trying to address current limitations and future challenges. (C) 2017 Elsevier B.V. All rights reserved.
Keywords: Computerwissenschaft, Disziplin, Favoriten, formale Sprache, Formalisierung, Intellektualtechnik, Kogn. Architektur, Kognitionswissenschaft/Social Sciences/Humanities, kognitive Architekturen, Künstliche Intelligenz, Mensch-Technik-Relationen (MTR), natürliche Sprache, Ontologie, Philosophie, Realtechnik, Sprachverarbeitung, Technik, Technikphilosophie, Überblick, Weltwissen -
(2018): Argumentation mining. How can a machine acquire common sense and world knowledge?. In: Argument & Computation 9 (1), S. 1-14. DOI: 10.3233/AAC-170025Keywords: Allgemeiner Sinn, Ambiguität, Argumentation, argumentation mining, Argumentationssuche, Argumentationstheorie, argumentative text processing, Common Sense, Deep Learning, fortgeschrittene interaktive Geräte, frame semantics, Kognitionswissenschaft/Social Sciences/Humanities, Lebenswissen, Linguistik, Mensch-Technik-Relationen (MTR), Natural language understanding, Neural Networks, Realtechnik, representation learning, Sprachverarbeitung, Sprachverstehen, Technik, unsupervised, Voraussetzungen für sozial angemessenes Verhalten, Weltwissen, world knowledge
2017
-
(2017): Online Recognition of Daily Activities by Color-Depth Sensing and Knowledge Models. In: SENSORS 17 (7)
Abstract: Visual activity recognition plays a fundamental role in several research fields as a way to extract semantic meaning of images and videos. Prior work has mostly focused on classification tasks, where a label is given for a video clip. However, real life scenarios require a method to browse a continuous video flow, automatically identify relevant temporal segments and classify them accordingly to target activities. This paper proposes a knowledge-driven event recognition framework to address this problem. The novelty of the method lies in the combination of a constraint-based ontology language for event modeling with robust algorithms to detect, track and re-identify people using color-depth sensing (Kinect® sensor). This combination enables to model and recognize longer and more complex events and to incorporate domain knowledge and 3D information into the same models. Moreover, the ontology-driven approach enables human understanding of system decisions and facilitates knowledge transfer across different scenes. The proposed framework is evaluated with real-world recordings of seniors carrying out unscripted, daily activities at hospital observation rooms and nursing homes. Results demonstrated that the proposed framework outperforms state-of-the-art methods in a variety of activities and datasets, and it is robust to variable and low-frame rate recordings. Further work will investigate how to extend the proposed framework with uncertainty management techniques to handle strong occlusion and ambiguous semantics, and how to exploit it to further support medicine on the timely diagnosis of cognitive disorders, such as Alzheimer’s disease.
Keywords: activities of daily living, activity recognition, assisted living, Bildsemantik, color-depth sensing, complex events, Ereigniserkennung, Ereignisverständnis, frame semantics, Informations- & Kommunikationstechnik, Knowledge representation, Kognitive Skills/Social Cognition, Mensch-Technik-Relationen (MTR), people detection and tracking, Realtechnik, senior monitoring, Sprachverstehen, Technik, Voraussetzungen für sozial angemessenes Verhalten, Weltwissen 2016
-
(2016) : Context-aware Planning by Refinement for Personal Robots in Smart Homes In: VDE: Proceedings of ISR 2016: 47st International Symposium on Robotics: Berlin: VDE Verlag GmbH. Online verfügbar unter https://ieeexplore.ieee.org/document/7559161/authors
Abstract: The idea of integrating robots and smart environments is becoming more popular. An important challenge for robotics and large advanced applications, such as Ambient Assisted Living, is to enable robots to seamlessly operate as part of smart spaces.They evolve in an unpredictable and highly dynamic environment where context information change quickly and where smart devices can join or leave it at anytime. In this situation, context-aware task planning is a key enabler. Actually, as the environment evolves, plans can become outdated, putting robots in blocking situations. Currently, few planners are able to consider smart spaces constraints during execution phase. In this paper, we propose a novel planning approach called DHTN (Dynamic HTN) based on HTN (Hierarchical Task Networks) planner. DHTN is able to generate and adapt the plan at execution and has the capability to smartly probe the environment by exclusively querying devices that provide useful data. Our approach was implemented and evaluated through simulation and a real life scenario using Nao robot in a smart office.
2007
-
(2007): Mixed-Initiative Human–Robot Interaction Using Hierarchical Bayesian Networks. In: IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans 37 (6), S. 1158-1164. DOI: 10.1109/TSMCA.2007.906570
Abstract: As the usage of service robots becomes more sophisticated, direct communication by means of human language is required to increase the efficiency of their performance. In natural speech interaction, however, people often omit some words and rely on background knowledge or the context, resulting in ambiguity. In order to develop smarter service robots, therefore, managing the context of interaction is essential. In this correspondence, we have investigated the mixed-initiative interaction that prompts for missing information and clarifies ambiguous statements based on hierarchically designed Bayesian networks. Simulation with the Kephera II robot and a usability test have demonstrated the usefulness of the proposed method.
2006
-
(2006): A PERCEPTUALLY-BASED THEORY OF MIND FOR AGENT INTERACTION INITIATION. In: International Journal of Humanoid Robotics 03 (03), S. 321-339. DOI: 10.1142/S0219843606000783
Abstract: We endow agents with the capability to open interactions based on their perception of the gaze and direction of attention of others in a virtual environment. The capability is geared towards the earliest part of interaction initiation, where agents may be at some distance from each other and may not initially have knowledge of each other's presence. An important idea in our work is that the start of interaction be initiated in a graceful manner involving exchanges of subtle cues before overt interaction commitments are made. Synthetic vision, attention and memory are used to implement the perceptually-based agent theory of mind. Theory of Mind is used to infer information about the intention of the other to interact based on their eye, head and body directions, locomotion and greeting gestures. An agent's interaction behavior is therefore driven not only by its interaction goal, but also by its theory of the goal of the other based on perception. We have implemented this system and used it to automate and evaluate social interaction behaviors between humans and agents in a virtual environment.
Keywords: Begrüßungsgesten, Blick / Gaze, Blickrichtung, Favoriten, Fortbewegung/locomotion, Gesten, Gestik/Mimik, Körperhaltung, Modell zur Interpretation von Handlungsvorhaben für Roboter, Richtung der Kopfhaltung, Richtung der Körperhaltung, Subtilität der Kommunikationsgestik, theory of mind, Weltwissen
- <<
- <
- 1