Mariana Damova

<p>The exploitation of rivers and hydropower reservoirs involves daily monitoring of the water resources, the meteorological conditions, the status of the coast, the flood areas, etc. Providing with timely and easy to consume information, analytics and early warnings for current and upcoming statuses or events helps water resources managers and high level officials to adequately observe and plan operations for sustainable development of river areas. We present an intelligent web-based workflow that combines different methods of AI, e.g. linked data, deep learning and resoning, to provide an integrated information system that ensures interoperability between spatial information of GIS systems, remote sensing information, symbolic and numerical data like meteorological data and proprietary measurements and creates an actionable knowledge value chain for the needs of rivers and hydropower reservoirs exploitation with embedded early warning capability. We show how hydrodynamic modelling using Telemac with forecasted water economic data, produced from earth observation and in-situ measurements applied to a series of neural network architectures, derive predictive river models, that are integrated into the work-flow and made available for querying, reviewing, projecting the changes in the navigational conditions of navigable rivers, geo-spatial visualization on GIS. The intelligent work-flow further provides with functional features like forecasts generation for river discharge, turbidity, water level, alerting and querying of a variety of correlations and synchronized visualizations in tables, graphs and GIS maps. It helps improve the<br />operational efficiency by providing ability to interact with and view all water resources management information at ones, ensures accuracy and decision making ability by correlating historic and forecast data with satellite imagery and data, gives automated forecasting of water economic data using satellite meteorological data, reduce risk through automated alerts. We demonstrate on the example of Danube the advantages of the presented intelligent web-based work-flow for the monitoring of rivers and their environment for sustainable development and planning.</p> <p>Acknowledgement<br />This work has been carried out within ESA Contract No 4000133836/21/NL/SC</p> <p> </p>

The exploitation of rivers and hydropower reservoirs involves daily monitoring of  the water resources, the meteorological conditions, the status of  the river banks, the flood areas, etc. As maximum river discharge often results in flooding, it is of importance to provide with timely and reliable forecasts of discharge and water levels. Predicting river discharge and water levels has been a subject of hydrological modelling and a topic of serious research. However, only in recent years scholars and practitioners have turned to consider earth observation data for their studies, mainly to compare evidence of flood mapping. We present an approach of using earth observation data to feed AI architectures – EO4AI – and produce forecasts for discharge and water level with significant degrees of accuracy.  Our starting point is that river discharge and water levels depend on a variety of meteorological and environmental factors like precipitations, snow cover, soil moisture, vegetation index and satellite data offer rich variety of datasets, supplying this information. We adopt a pipeline of deep learning architectures consisting of GAN, CNN, LSTM and EA to actually generate forecasts for river discharge and water level by using historic satellite data of the meteorological features listed above, and in-situ measurements for water level and discharge. The satellite data are provided by ADAM  via the NoR service of ESA. ADAM provides data access to satellite datasets from different satellites with semantic relevance for the construction of sediment transport and deposition forecast model as discussed above.  We explain the purpose of the pipeline components. Our forecast models are calibrated for 3, 5, 7, 30 days ahead, and our experiments provide predictions for one year ahead with each of the calibrated models. We discuss experiment results carried out with data from the Danube and Arda rivers, including three dams from cascade Arda and compare them with predictions derived with other methods. We demonstrate the viability of the approach and the reliability of the forecasting results. We further show how the forecasts can be used in hydrodynamic modelling context, for early warning applications and for routine water resources management and monitoring tasks.

D4.3A is an annex to the D4.3 deliverable of WP4 of the MOLTO project. It aims to address the reviewers’ remarks and recommendations for D4.3, as well as to present a final overview of the prototypes built in the scope of MOLTO with respect to grammar-ontology interoperabilty. D4.3A also describes the work after M24 and gives a general overview of the achievements in MOLTO with focus on WP4 - Knowledge Engineering, WP7 - Patents use case, and WP8 - Cultural Heritage use case.

Large Lexical Data Bases are one of the earliest applications of NLP. The initial stage of their rise, with the admiration for the automation of lexicographic work itself, came to an end long ago. In the following stages LexicalData Bases (LDB) began to extend considerably the range of their application and the scope of CL problems put forward by them [see Calzolari 1991, Calzolari and Zampolli 1988 and Boguraev et a1.1988]. It is worth discussing a new version of LDB (for a concrete new language) only in the present-day context of these problems. This does not, however, relieve the creators of LDB for a new language of the solution of the trivial problems standing at the lower foot of the ladder used to "storm" the lexical wealth of language. After overcoming these obstacles, there is prototype version available or a core of LDB, which cannot be called large especially when its volume is concerned. Speaking of volume, quite naturally, the following question arises: in wha...

THIS DOCuMENT DESCRIBES THE DATA AND THE DATA MODELS TO BE INCLuDED IN THE KNOWLEDGE REPRESENTATION INFRASTRuCTuRE (KRI) DESCRIBED IN DELIVERABLE 4.1. THE ADOPTED APPROACH OF LINKED DATA IS ExTENDED TO MuLTILINGuAL COVERAGE, AND APPLIED TO DATA FROM THE CuLTuRAL HERITAGE SuBJECT DOMAIN. THIS DOCuMENT PRESENTS THE METHODS OF INTERLINKING THE DIFFERENT RESOuRCES AT SCHEMA AND AT INSTANCE ...

This document describes the interoperability between the Grammatical Framework (GF) grammar and a semantic repository, which aims to bridge the gap between natural language (NL) and formal knowledge. Two different approaches are presented: semi-automatic with manual intervention and template-based GF grammar generation.

EXECUTIVE SUMMARY This deliverable describes the envisaged requirements for the Semantic TV Broker component. It is based on the use cases obtained by the WP7 tasks and the architectural requirements posed by WP6 task 6.1. In addition, it presents an overview of current SWS technology applied to the SWS Broker in the light of the technologies presented in D6. 1.

RENDER FP7-ICT-2009-5 Contract no.: 257790 www.render-project.eu ... Author(s): Maurice Grinberg, Ontotext; Mariana Damova, Ontotext; Atanas Kiryakov, Ontotext Deliverable Nature: Prototype (P) ... Suggested Readers: Research staff working on the data ...

The European Cultural Heritage Strategy for the 21st century has led to an increased demand for fast, efficient and faithful 3D digitization technologies for cultural heritage artefacts. Yet, unlike the digital acquisition of cultural goods in 2D which is widely used and automated today, 3D digitization often still requires significant manual intervention, time and money. To overcome this, the authors have developed CultLab3D, the world's first fully automatic 3D mass digitization technology for collections of three-dimensional objects. 3D scanning robots such as the CultArm3D-P are specifically designed to automate the entire 3D digitization process thus allowing to capture and archive objects on a large-scale and produce highly accurate photo-realistic representations.

Abstract. The Linked Open Data (LOD) cloud is rapidly becoming the largest interconnected source of structured data on diverse domains. The potential of the LOD cloud is enormous, ranging from solving chal-lenging AI issues such as open domain question answering to automated knowledge discovery. However, due to an inherent distributed nature of LOD and a growing number of ontologies and vocabularies used in LOD datasets, querying over multiple datasets and retrieving LOD data re-mains a challenging task. In this paper, we propose a novel approach to querying linked data by using alignments for processing queries whose constituent data come from heterogeneous sources. We also report on our Alignment based Linked Open Data Querying System (ALOQUS) and present the architecture and associated methods. Using the state of the art alignment system BLOOMS, ALOQUS automatically maps concepts in users ’ SPARQL queries, written in terms of a conceptual upper ontol-ogy or domain specific ontolo...

Abstract. Linking Open Data (LOD) facilitates the emergence of a web of linked data by publishing and interlinking open data on the web in RDF. One can explore linked data across servers by following the links in the graph. The LOD cloud has 203 datasets and more than 14 billion RDF triples

As the amount of cultural data avail-able on the Semantic Web is expand-ing, the demand of accessing this data in multiple languages is increas-ing. Previous work on multilingual access to cultural heritage informa-tion has shown that at least two dif-ferent problems must be dealt with when mapping from ontologies to nat-ural language: (1) mapping multilin-gual metadata to interoperable knowl-edge sources; (2) assigning multilin-gual knowledge to cultural data. This paper presents our effort to deal with these problems. We describe our expe-riences with processing museum data extracted from two distinct sources, harmonizing this data and making its content accessible in natural language. We extend prior work in two ways. First, we present a grammar-based sys-tem that is designed to generate co-herent texts from Semantic Web on-tologies in 15 languages. Second, we describe how this multilingual system is exploited to form queries using the standard query language SPARQL. The generati...

This paper deals with the notion of aspect as it is understood in the eventuality structure based formal approaches to aspect. These approaches typically link aspect to the interpretation of the philosophical and ontological notion of event, seen as a conceptual entity with rigid edges: beginning, protraction and end, and analyse and study extensively the end part of events ((Vendler, 1967),(Moens & Steedman, 1988), (Smith, 1991), (Pustejovsky, 1991), (Krifka, 1989), (Partee, 1984), (Hinrichs, 1986), etc.). The beginning, a semantic counterpart of the culmination on the other hand, has not been discussed so much at large. We analyse various language means that convey beginning and argue for the need of a mechanism to provide a uniform interpretation for them. We define the aspectual type BEGIN, and develop its semantic representation along the general lines of accounts of temporal reference of Discourse Representation Theory ((Kamp, 1979); (Kamp & Reyle, 1993)). We extend the DRT an...

This paper deals with the notion of aspect as it is understood in the eventuality structure based formal approaches to aspect. These approaches typically link aspect to the interpretation of the philosophical and ontological notion of event, seen as a conceptual entity with rigid edges: beginning, protraction and end, and analyse and study extensively the end part of events ((Vendler, 1967),(Moens & Steedman, 1988), (Smith, 1991), (Pustejovsky, 1991), (Krifka, 1989), (Partee, 1984), (Hinrichs, 1986), etc.). The beginning, a semantic counterpart of the culmination on the other hand, has not been discussed so much at large. We analyse various language means that convey beginning and argue for the need of a mechanism to provide a uniform interpretation for them. We define the aspectual type BEGIN, and develop its semantic representation along the general lines of accounts of temporal reference of Discourse Representation Theory ((Kamp, 1979); (Kamp & Reyle, 1993)). We extend the DRT an...

Digital museum databases have extremely heterogeneous data structures which require advanced mapping and vocabulary integration for them to benefit from the interoperability enabled by semantic technologies. In addition to establishing ways of extracting and manipulating digitally encoded cultural material, there exists a need to make this material available and accessible to human users in different forms and languages that are available to them. In this paper we describe a method to manage and access museum data by integrating it within a series of interlinked ontological models. The method allows querying and generation of query results in natural language. We report on the results of applying this method from experiments we have been pursuing.

In this position paper, we illustrate how Linked Data can be eectively used in a Technology-enhanced Learning scenario. Specically, we aim at using structured data to semiautomatically generate artifacts to support learning delivery and assessment: natural language facts, Q&A systems and quizzes, also used with a gaming avour, can be creatively generated to help teachers and learners to support and improve the learning path. Moreover, those artifacts can in turn be published on the Web as Linked Data, thus directly contributing to make the Web a global data space also for learning purposes.

Cultural heritage appears to be a very useful use case for Semantic Web technologies. The domain provides with plenty of circumstances where linkages between different knowledge sources are required to ensure access to rich information and respond to the needs of professionals dealing with cultural heritage content. Semantic Web technologies offer the technological backbone to meet the requirement of integrating heterogeneous data easily, but they are still more adapted to be consumed by computers than by humans, especially non-engineers or developers. This chapter is about a technique which allows interaction in natural language with semantic knowledge bases. The proposed technique offers a method that allows querying a semantic repository in natural language and obtaining results from it as a coherent text. This unique solution includes several steps of transition from natural language to SPARQL and from RDF to coherent multilingual descriptions, using the Grammatical Framework, G...

In this paper we propose an ontological framework for tools facilitating creative writing and story reading. It is based on an ontology implemented as a topic map and employs linguistic analysis methods for discovering conceptual entities in the text.

In this paper we present a multilingual SPARQL-based [1] retrieval interface for querying cultural heritage data in natural language (NL). The presented system offers an elegant grammar-based approach which is based on Grammatical Framework (GF) [2], a grammar formalism supporting multilingual applications. Using GF, we are able to present a cross-language SPARQL grammar covering 15 languages and a cross-language retrieval interface that uses this grammar for interacting with the Semantic Web. To our knowledge, this is the first implementation of SPARQL generation and parsing via GF that is published as a knowledge representation infrastructure-based prototype. Querying the Semantic Web in natural language, more specifically, using English to formulate SPARQL queries with the help of controlled natural language (CNL) syntax has been developed before [3,4]. Such approaches, based on verbalization methods are adequate for English, but in a multilingual setting where major challenges s...

The collection of the specialized vocabulary of a particular domain (terminology) is an important initial step of creating formalized domain knowledge representations (ontologies). Terminology Extraction (TE) aims at automating this process by collecting the relevant domain vocabulary from existing lexical resources or collections of domain texts.

Contents 1 Introduction 4 2 Implementation of Transfer 6 2.1 Preliminary Remarks : : : : : : : : : : : : : : : : : : : : : : : : : : : 6 2.2 The Predicate tau/2 : : : : : : : : : : : : : : : : : : : : : : : : : : : 7 2.3 The Predicates tau lex/f2,5g and sem lex/2 : : : : : : : : : : : : 10 2.4 The Predicates tau prep/4 and tau pred/3 : : : : : : : : : : : : : : 10 3 The Transfer of Nouns 12 3.1 Rigid Nouns : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 12 3.2 Common Nouns : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 12 3.3 Relational Nouns : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 13 3.4 Proper Names : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 14 3.5 Pronouns : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 15 3.6 Determiners : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 17 4 The Transfer of Verbs 18 4.2 Main Verbs : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 19 4.2.

Die vorliegende Arbeit wurde im Rahmen des Verbundvorhabens Verbmobil vom Bundesministerium f ur Bildung, Wissenschaft, Forschung und Technologie (BMBF) unter dem F orderkennzeichen 01 IV 101 U gef ordert. Die Verantwortung f ur den Inhalt dieser Arbeit liegt bei den ...

This abstract is part of a project proposal submitted for : H2020-REFLECTIVE-7-2014, Coordinated by University of Salerno. Abstract author: G. Laquidara, Et al. The project aims to build a HPW (High Power Web) framework, enforced with semantic and computational processors, able to harvest, analyze, modelize, semantically annotate 3D complex objects, and assorting them into a networked catalogue of built environments, fully explorable and cooperatively hackable through a suite of professional and consumer applications. The framework will be built within 3 different coexistent layers, accessible concurrently through 3 distinguished profiles, in order to be co-used by: #1. Researchers and technical experts in the field of archeological / architectural 3D modeling, predictive modeling, models building, large or short scale regional analysis; #2. Professionals creatives, CCIs and creative-labs; #3. users, visitors and a wide target of customers. Given the (triple) multi-tiered structure,...