CURRENT RESEARCH PROJECTS
hackAIR- H2020-ICT, (2016-present)
Collective awareness platform for outdoor air pollution
The overall objective of hackAIR is to develop and pilot test an open platform that will enable communities of citizens to easily set up air quality monitoring networks and engage their members in measuring and publishing outdoor air pollution levels, leveraging the power of online social networks, mobile and open hardware technologies, and engagement strategies.
KRISTINA – H2020-ICT, (2015-present)
A Knowledge-Based Information Agent with Social Competence and Human Interaction Capabilities
KRISTINA’s overall objective is to research and develop technologies for a human-like socially competent and communicative agent that is run on mobile communication devices and that serves for migrants with language and cultural barriers in the host country as a trusted information provision party and mediator in questions related to basic care and healthcare.
Mining and Understanding of multilinguaL contenT for Intelligent Sentiment Enriched coNtext and Social Oriented inteRpretation
The consumption of large amounts of multilingual and multimedia content regardless of its reliability and cross-validation can have important consequences on the society. To break this spiral, we need multilingual technologies with sentiment, social and spatiotemporal competence that are able to interpret, relate and summarise economic information and news created from various local subjective and biased views and disseminated via TV, radio, mass media websites and social media. To achieve multidimensional integration of heterogeneous resources, MULTISENSOR proposes a content integration framework that builds upon multimedia mining, knowledge extraction, analysis of computer-mediated interaction, topic detection, semantic and multimodal representation as well as hybrid reasoning.
The developed technologies will be validated with the aid of 2 main use cases: a) International mass media news monitoring and b) SME internationalisation.
Home Made Explosives and Recipes
The goal of HOMER is to expand the knowledge of European bodies about Homemade Explosives (HME) and to improve the capacity of security and law enforcement agencies to cope with current and anticipated threats in order to reduce the probability that HMEs will be used by terrorists. The concept of the HOMER project is based on four pillars: a) data collection and analysis based on HME literature study, information mining from internet resources, industry best practices, as well as chemical and compound filtering analysis; b) knowledge management to support the sustainability of HME knowledge content; c) case study experiments and validation; d) standardization of HME vocabulary and definition among the security and law enforcement agencies and also the precursor’s manufacturers, for the purpose of preventing, detecting and mitigating the threat resulting from the manufacture of HMEs by terrorist and criminal elements. ITI-CERTH will lead the discovery and analysis of HME recipes in the Web and will be involved in the knowledge management of HME resources.
The European Network on Integrating Vision and Language (iV&L Net): Combining Computer Vision and Language Processing For Advanced Search, Retrieval, Annotation and Description of Visual Data
The central goal of iV&L Net is to build a European CV/NLP research community, targeting 4 focus themes: (i) Integrated Modelling of Vision and Language for CV and NLP Tasks; (ii) Applications of Integrated Models; (iii) Automatic Generation of Image & Video Descriptions; and (iv) Semantic Image & Video Search. iV&L Net will organise annual conferences, technical meetings, partner visits, data/task benchmarking, and industry/end-user liaison.
PAST RESEARCH PROJECTS
Multilingual and Multifaceted Interactive Information Access, Cost Action IC1002.
Multilingual and Multifaceted Interactive Information Access (MUMIA) research and development will be a key part of the next generation of search technology. Machine Translation (MT), Information Retrieval (IR) and Multifaceted Interactive Information Access (MIIA) are three disciplines which address the main components of MUMIA. This Action will launch a much needed initiative to coordinate the collaboration between these disciplines, fostering research and technology transfer in these areas and play an important role in the definition of the future of search. To form a common basis for collaboration, the domain of patent retrieval has been selected as it provides highly sophisticated and information intensive search tasks that have significant economic ramifications; as well as providing a large scale unifying test bed of multilingual and dynamic data. This Action will explore innovative frameworks to empower the synergies from the disparate research fields of MT/IR/MIIA within the specific context of patent search and other next generation Web applications.
Personalized Environmental Service Configuration and Delivery Orchestration
The overall goal of PESCaDO is to research and develop strategies for user-centred environmental service orchestration and multilingual information delivery in the context of active decision support for environment conditioned tasks. These strategies, realized as an operational workbench, will support the user during the entire process of environmental information-based decision-making from the formulation of the problem, via the search for adequate environmental service nodes and their connection (including data, knowledge and language synchronization), up to multilingual information, advice, hints, etc. delivery. PESCaDO will adapt its strategies and language preferences to the profile and the needs of the user. Furthermore, it will allow for intervention and feedback at any stage of the process, incorporate the received feedback and facilitate several loops of the process until the full satisfaction of the user. The backbone of the involved strategies will be semantics. In addition, PESCaDO will develop visual analytics strategies that will be specific to incremental environmental conditions oriented decision support and problem solving.
Coordinated approacH to the EurOpean effoRt on aUdio-visual Search engines
CHORUS is a Coordination Action which aims at creating the conditions of mutual information and cross fertilisation between the projects that will run under Strategic objective 2.6.3 (Advanced search technologies for digital audio-visual content) and beyond the IST initiative.
CHORUS workplan includes the set up of a coordination and information exchange platform (cluster meetings, newsletter, website) to enhance the interactions and to invoke discussions between the key players of the IST ‘Cluster’ and thus to stimulate the creation of the European Research Area (ERA).
An important task of the project consists of supporting the preparation of an analysis and a roadmap for the realisation of Audio-visual search engines in EU. That activity will need a multi-disciplinary (societal, economic, regulatory and technological) and strategic approach and will imply the participation of major Stakeholders of the domain through the creation of a Think-Tank. The Think Tank, to which several representative organizations have already agreed to participate, will contribute to give an industrial relevance to the coordination action and its visions. The areas a priori investigated by the project are described in WP2 activities.
Interaction and dissemination will be further ensured through the set-up of a conference and dedicated workshops addressing a large community which complement the coordination instruments proposed by CHORUS.
Patents belong to the few types of public information that have a big impact on the European economy, and whose proper monitoring, retrieval, representation, interpretation, and assessment so clearly depend on the access to its content, and, thus, on advances in semantics-based techniques. However, research and development in the area of patent processing still focuses on selected traditional tasks such as text retrieval, classification, and shallow linguistic analysis. PATExpert’s overall scientific objective is to change the paradigm currently followed for patent processing from textual (viewing patents as text blocks enriched by picture material that are interpreted as ‘canned’ material, sequences of morpho-syntactic tokens, or collections of syntactic structures) to semantic (viewing patents as multimedia knowledge objects) processing.
MKLab is closely collaborating with the Department of Electrical and Computer Engineering of the Faculty of Engineering of the Aristotle University of Thessaloniki – AUTH and in particular with the Information Processing Laboratory (IPL) of AUTH, in the research activities that are carried out in the PATexpert project.
Knowledge Space of Semantic Inference for Automatic Annotation and Retrieval of Multimedia Content
K-Space will create a sustainable network of world-leading research teams from academia and industry to conduct integrative research and dissemination activities in semantic inference for automatic and semi-automatic annotation and retrieval of multimedia content, aiming at closing the gap between the low-level content descriptions that can be computed automatically by a machine and the richness and subjectivity of semantics in high-level human interpretations of audiovisual media. Specifically, K-Space integrative research will focus on three key areas: Content-based multimedia analysis; Knowledge extraction; and Semantic multimedia. To ensure that the output of the targeted research becomes the primary feeding ground for industrial innovation, an Industrial Advisory Board with a core of four key industrial players has been set up and has been consulted from the very outset of proposal building. Its mission is to assist the network in defining research directions and to advise on potential exploitation of research results as they emerge.
Enhanced Learning Unlimited.
The objective of Enhanced Learning Unlimited (ELU) is to look on how interactive digital TV can be used to increase learning opportunities in the home, office and school. The television is a familiar and reliable consumer device with around 95-99% penetration in European households. It is also perceived to be a source of learning. However, in its more traditional role it has tended to be used in a passive viewing mode.
COST292: Semantic Multimodal Analysis of Digital Media (2006-2008)
Semantic Multimodal Analysis of Digital Media, European Cooperation in the Field of Scientific and Technical Research, 2004 – 2008. Objective of this action is to push forward the frontiers of current research on semantic analysis, inference and conceptualisation for high-level annotation and retrieval of digital audiovisual content.
New Forms of Distributed Organization and Access to Cultural Heritage
The overall aim of REACH is to provide a web portal enabling access to different databases of cultural content with emphasis on multimedia material. The objective is to use ontologies to provide enhanced unified access to heterogeneous distributed cultural heritage digital databases. In order to achieve this, the project employs ontology-based representation of cultural content, multimedia content-based search algorithms, hybrid ontology and content-based access and user-friendly search interfaces. This fusion approach provides users with resources for building queries of multimedia repositories sequentially using multiple individual search tools, and performs higher quality searching. The system will be tested using real content such as inscriptions and coins from the Greco-Roman time period.