School of Electrical & Computer Engineering 
Intelligent Systems Laboratory   Technical University of Crete 
border graphic Main Menu  border graphic
border graphic border graphic border graphic border graphic border graphic
- Home
- News & Events
- People
- Courses
- Research
- Publications
- System Prototypes
- Funding
- Contact
border graphic   border graphic
border graphic Members Area  border graphic
border graphic border graphic border graphic border graphic border graphic
- Web Mail
- Reservations
border graphic   border graphic
border graphic About  border graphic
border graphic border graphic border graphic border graphic border graphic
- School of ECE
border graphic   border graphic
Intelligence Cloud Infrastructure


System Prototypes

CHΟROS, a system for reasoning over spatial information in OWL ontologies. CHOROS reasoner applies over spatial (RCC Topological and Cone-shaped Directional) relations, infers implied relations, detects inconsistencies and retains soundness, completeness and tractability over the supported relations using path consistency. The updated version of the reasoner (spatialReasoner2), relies on the idea of decomposing directional relations into two smaller sets of basic relations yielding fewer compositions.[See related article] [View System]
CHRONOS, a system for reasoning over temporal information in OWL ontologies. It applies over temporal relations (Allen's interval algebra), infers implied relations, detects inconsistencies and retains soundness, completeness and tractability over the supported relations using path consistency. The updated version of the reasoner (chronos_x28), computes a minimal set of relation compositions at run-time (based on a tractable subset of Allen relations) while being sound and complete.[See related article] [View System]
SOWL QL is a high-level query language for querying spatio-temporal ontologies in OWL 2.0. SOWL QL extends SPARQL with a powerful set of temporal and spatial operators, including Allen operators, and is capable of querying both quantitative and qualitative temporal and spatial information (e.g., information expressed using dates, times, temporal intervals or information expressed using natural language terms such as “before” or “after”). The language is model independent and supports reasoning during the querying process. SOWL QL is implemented in full along with a Graphical User Interface (GUI) facilitating ontology loading and parsing, query formulation and execution and, results viewing. [See related article:"SOWL QL: Querying Spatio - Temporal Ontologies in OWL"]. Use the following link to download the SOWL QL Prototype: [file]   
SWRL rules for qualitative spatial and temporal reasoning for Point Algebra, Interval Algebra, RCC8 Topological Relations and Cone Shaped Directional Relations. (If you need access to the implementation, please contact) [See related articles:"SOWL: A Framework for Handling Spatio-Temporal Information in OWL" and "Temporal Reasoning for Supporting Temporal Queries in OWL 2.0"].
The acquisition of accurate estimates regarding the power output of inherently intermittent and potentially distributed renewable energy sources is a major concern for scientists and the wider public alike. RENES (Renewable ENergy EStimator) is an interactive web-based tool that enables its users to acquire short-to-mid term predictions on the output of photovoltaic (PV) systems and wind turbine generators (given online weather forecasts provided by meteorological websites). RENES is not a commercial product, and its usage is available to all, free of charge. We envisage RENES to be employed by Artificial Intelligence and Multiagent Systems researchers, for activities such as accumulating data, designing experiments or evaluating new algorithms. To this end, we also provide a web-based application program interface (API), thus enabling the user to fully exploit the tool's requests. [View System]
Handling temporal ontologies in Protege can be complex and requires that the user be familiar with the peculiarities of the underlying temporal representation (N-ary relations). CHRONOS Ed is a Tab plug-in for Protege OWL editor that facilitates creation and editing of temporal ontologies in OWL. CHRONOS Ed enables handling of temporal relations in Protege the same way static ones are handled [See related articles:article 1, article 2, article 3]. Use the following link to download the plugin: [file]   
DOGi (Dog OntoloGy Image annotator) is a complete and fully automated semantic annotation system for images of dog breeds. Annotation relies on feature extraction and on associating low-level features with image concepts in an ontology. Because general purpose ontologies for all image types are not yet available, we choose the problem of annotating images of dog breeds as a case study and for the evaluation of our methodology. Nevertheless, DOGi can be adapted to more image types provided that an ontology for a new image domain becomes available. Therefore, DOGi offers an ideal test-bed for experimentation and sets the grounds for the annotation and evaluation of virtually any image type. Evaluation results are realized using images collected from the Web. Almost 95% of the test images is correctly annotated (i.e., DOGi identified their class correctly). [View System]
We design and implement a Plug-In for the Protégé editor that facilitates the crafting (i.e., creating, editing) of temporal ontologies. Particular emphasis is given to making the Plug-In portable and easy to use by ordinary users of the semantic Web. The Plug-In is realized as a Protégé tab that provides a front-end interface which is easy to use, handles temporal ontologies similarly to static ontologies and does not require that users are familiar with peculiarities of the underlying representation of temporal information (the 4D- fluents approach in our case) [See related article]. Use the following link to download the 4D-fluents plugin: [file]   
PoTrain is a Web-based application that was developed on behalf of project PoTrain: Positive Thinking Training' (EU Life-Long Learning, Comenius, Project No. 509925). PoTrain aims at encouraging the collection and dissemination of positive (good) news by students across the EU using the Web. The media, especially TV are transmitting mostly negative news, so establishing negative psychology and pessimistic mode to young people. Collecting positive news from different Web sources, will counterbalance the flow of negatives news. PoTrain aims to offer non-formal training and to assist children in developing useful skills and positive way of thinking. [View System]
OntoGain is a method for ontology learning from multi-word concept terms extracted from plain text. OntoGain follows an ontology learning process defined by distinct processing layers. Building upon plain term extraction a concept hierarchy is formed by clustering the extracted concepts. The derived term taxonomy is then enriched with non-taxonomic relations. [View System]
The Larvabot is a prototype modular robotic mechanism which combines three modes of locomotion: snake-like, caterpillar-like, and loop-like. The loop-like locomotion is modeled after the motion of the Chlorochlamys Chloroleucaria larva. The Larvabot could be used in various applications, such as search-and-rescue missions. [View System] [file]   
Web Summarization by Image Content. It is the process of extracting the most important or characteristic images from a Web site. This process is complementary to text summarization which works by extracting brief text summaries (e.g. important phrases or sentences) from Web sites. To the best of our knowledge, Web image summarization methods are not known to exist. This is exactly the problem this system is dealing with. [View System]
Information Retrieval System for MedLine and Medical Information. MedSearch is a complete retrieval system for Medline, the premier bibliographic database of the U.S. National Library of Medicine (NLM). MedSearch implements the Semantic Similarity based Retrieval Model (SSRM), a novel information retrieval method for discovering similarities between documents containing semantically similar but not necessarily lexically similar terms. [View System]
Semantic Similarity relates to computing the similarity between concepts (terms) which are not necessarily lexically similar. We investigate approaches to computing semantic similarity by mapping terms to an ontology and by examining their relationships in that ontology. More specifically, we investigate approaches to computing the semantic similarity between natural language terms (using WordNet as the underlying reference ontology) and between medical terms (using the MeSH ontology of medical and biomedical terms). The most popular semantic similarity methods are implemented and evaluated using WordNet and MeSH. We also focus on cross ontology methods which are capable of computing the semantic similarity between terms stemming from different ontologies (WordNet and MeSH). This is a far more difficult problem (than in the single ontology) which has not been investigated adequately in the literature. X-Similarity, a novel cross-ontology similarity method is also a contribution of this work. [View System]
IntelliSearch is a complete and fully automated information retrieval system for the Web. It supports fast and accurate responses to queries addressing text and images in Web pages by incorporating state-of-the-art text and Web link information indexing and rertieval methods in conjunction with efficient ranking of Web pages and images by importance (authority). Searching by semantic similarity for discovering information related to user's requests (but not explicitly specified in the queries) is a distinguishing feature of the system. IntelliSearch stores a crawl of the Web with more than 1,5 million Web pages with images and is accessible on the Internet. It offers an ideal test-bed for experimentation and training and serves as a framework for a realistic evaluation of many Web image retrieval methods. [View System]


© 2018 Intelligent Systems Laboratory, TUC