Search academic texts

Information × Registration Number 0218U003044, 0117U005086 , R & D reports Title Innovative System for tactile adaptation of documents for the peoples with special needs (veterans of war, students, aging...) popup.stage_title Head Stirenko Sergiy Grygorovych, Registration Date 12-01-2018 Organization The informatics and computer science faculty the National Technical University of Ukraine "KPI" popup.description2 The analysis of available concepts and approaches, the study of requirements, system components and the principles of integrating components to care for people with special needs (including the elderly and people with disabilities) showed that they do not meet modern requirements and new challenges. There is a need to seek truly innovative innovations that will solve new problems and problems. The proposed integrated ecosystem provides the basis for effective individual care for people with special needs through the introduction of multimodal personalized communication channels. This allows end users to derive a cumulative effect from a mixture of information and communication technologies, such as Internet of Things (IoT) and Internet of Everything (IoE), multimodal approaches based on Augmented Reality (AR) and predictable approaches based on machine learning and eliminate the obsessive technological interactions between human-to-machine (H2M) and machine-to-human (M2H), shifting them to machine- to-machine (M2M), which are encapsulated at the growing level computations, and enhancing pleasant multimedia and interactive non-verbal interaction H2M / M2H. In the context of the development of new approaches for multimodal nonverbal and situational communications, a formalized approach is proposed for the generation and evaluation of symbols, which can be transmitted by a wide range of non-verbal means (alphabets), and the approach to their creation. Several variants of experimental setups and some methods of machine learning for evaluating the utility and effectiveness of components for non-verbal communication (alphabets and systems based on them) are considered. The use of machine learning methods for learning and recognizing the intensity of the aforementioned non-verbal communication allowed us to analyze the mental and physical stress on individuals under the influence of various stimuli, including mathematical operations, verbal and non-verbal communication, using multimodal channels (acceleration, heart activity and brain activity). The results obtained by machine learning methods, in fact, neural networks of in-depth training for data obtained by multimodal channels (acceleration, heart activity and brain activity), open new perspectives for assessing the intensity and effectiveness of the aforementioned nonverbal communication by attracting new channels of nonverbal communication based on a brain-computer interface and electromyographic sensors. According to the results of the work, 8 presentations were prepared, submitted and presented jointly with our French partners at international scientific conferences and 4 papers for international scientific journals (which are taken into account by the international scientometric databases Web of Science, Scopus, Copernicus). Product Description popup.authors Гордієнко Юрій Григорович popup.nrat_date 2020-04-02 Close