Ben De Meester

Declarative Data Transformations for Linked Data Generation: The Case of DBpedia

Mapping languages allow us to define how Linked Data is generated from raw data, but only if the raw data values can be used as is to form the desired Linked Data. Since complex data transformations remain out of scope for mapping languages, these steps are often implemented as custom solutions, or with systems separate from the mapping process. The former data transformations remain case-specific, often coupled with the mapping, whereas the latter are not reusable across systems. In this paper, we propose an approach where data transformations (i) are defined declaratively and (ii) are aligned with the mapping languages. We employ an alignment of data transformations described using the Function Ontology ( Open image in new window ) and mapping of data to Linked Data described using the rdf Mapping Language (rml). We validate that our approach can map and transform dbpedia in a declaratively defined and aligned way. Our approach is not case-specific: data transformations are independent of their implementation and thus interoperable, while the functions are decoupled and reusable. This allows developers to improve the generation framework, whilst contributors can focus on the actual Linked Data, as there are no more dependencies, neither between the transformations and the generation framework nor their implementations.

An Ontology to Semantically Declare and Describe Functions

Applications built on top of the Semantic Web are emerging as a novel solution in different areas, such as decision making and route planning. However, to connect results of these solutions – i.e., the semantically annotated data – with real-world applications, this semantic data needs to be connected to actionable events. A lot of work has been done (both semantically as non-semantically) to describe and define Web services, but there is still a gap on a more abstract level, i.e., describing interfaces independent of the technology used. In this paper, we present a data model, specification, and ontology to semantically declare and describe functions independently of the used technology. This way, we can declare and use actionable events in semantic applications, without restricting ourselves to programming language-dependent implementations. The ontology allows for extensions, and is proposed as a possible solution for semantic applications in various domains.

A digital-first authoring environment for enriched e-books using EPUB 3

The overall majority of books are currently being made with primarily a printed outcome in mind. To make a digital version of these books, most manuscripts need to be re-processed, which usually results in customary built e-books. This need for a customized authoring workflow for every electronic version of a book makes it impossible to build e-books in a cost-effective way. In this paper, we propose a novel workflow that incorporates both print and digital book authoring. By charting the currently most widespread workflow Flemish publishers use to author print books and e-books, we are able to identify the most pressing problems. These are the print-first approach, the vendor lock-in situation of the e-reader market, and the high cost of updating and/or maintaining the content of an (e-)book. To overcome the aforementioned problems, we devise a new workflow that follows a digital-first approach using Open Web standards, separating content, structure, and layout. We evaluate the proposed workflow by building a proof-of-concept authoring environment. Using this new workflow, both digital and print books can be built without significant additional costs. The proof of concept is evaluated using an experts group of Flemish publishers, and received general positive reception, with concerns on how to incorporate the proposed workflow into production environments. By not limiting the proof of concept to a fixed data model, it could handle content from more content providers, facilitating further research into the possibilities and future requirements of the EPUB 3 specification.

Ontology Reasoning using Rules in an eHealth Context

Traditionally, nurse call systems in hospitals are rather simple: patients have a button next to their bed to call a nurse. Which specific nurse is called cannot be controlled, as there is no extra information available. This is different for solutions based on semantic knowledge: if the state of care givers (busy or free), their current position, and for example their skills are known, a system can always choose the best suitable nurse for a call. In this paper we describe such a semantic nurse call system implemented using the EYE reasoner and Notation3 rules. The system is able to perform OWL-RL reasoning. Additionally, we use rules to implement complex decision trees. We compare our solution to an implementation using OWL-DL, the Pellet reasoner, and SPARQL queries. We show that our purely rule-based approach gives promising results. Further improvements will lead to a mature product which will significantly change the organization of modern hospitals.

Specification and Implementation of Mapping Rule Visualization and Editing: MapVOWL and the RMLEditor

Visual tools are implemented to help users in defining how to generate Linked Data from raw data. This is possible thanks to mapping languages which enable detaching mapping rules from the implementation that executes them. However, no thorough research has been conducted so far on how to visualize such mapping rules, especially if they become large and require considering multiple heterogeneous raw data sources and transformed data values. In the past, we proposed the RMLEditor, a visual graph-based user interface, which allows users to easily create mapping rules for generating Linked Data from raw data. In this paper, we build on top of our existing work: we (i) specify a visual notation for graph visualizations used to represent mapping rules, (ii) introduce an approach for manipulating rules when large visualizations emerge, and (iii) propose an approach to uniformly visualize data fraction of raw data sources combined with an interactive interface for uniform data fraction transformations. We perform two additional comparative user studies. The first one compares the use of the visual notation to present mapping rules to the use of a mapping language directly, which reveals that the visual notation is preferred. The second one compares the use of the graph-based RMLEditor for creating mapping rules to the form-based RMLx Visual Editor, which reveals that graph-based visualizations are preferred to create mapping rules through the use of our proposed visual notation and uniform representation of heterogeneous data sources and data values.

Improving OWL RL Reasoning in N3 by Using Specialized Rules

Semantic Web reasoning can be a complex task: depending on the amount of data and the ontologies involved, traditional OWL DL reasoners can be too slow to face problems in real time. An alternative is to use a rule-based reasoner together with the OWL RL/RDF rules as stated in the specification of the OWL 2 language profiles. In most cases this approach actually improves reasoning times, but due to the complexity of the rules, not as much as it could. In this paper we present an improved strategy: based on the TBoxes of the ontologies involved in a reasoning task, we create more specific rules which then can be used for further reasoning. We make use of the EYE reasoner and its logic Notation3. In this logic, rules can be employed to derive new rules which makes the rule creation a reasoning step on its own. We evaluate our implementation on a semantic nurse call system. Our results show that adding a pre-reasoning step to produce specialized rules improves reasoning times by around 75i¾?%.

Using rule-based reasoning for RDF validation

The success of the Semantic Web highly depends on its ingredients. If we want to fully realize the vision of a machine-readable Web, it is crucial that Linked Data are actually useful for machines consuming them. On this background it is not surprising that (Linked) Data validation is an ongoing research topic in the community. However, most approaches so far either do not consider reasoning, and thereby miss the chance of detecting implicit constraint violations, or they base themselves on a combination of different formalisms, e.g. Description Logics combined with SPARQL. In this paper, we propose using Rule-Based Web Logics for RDF validation focusing on the concepts needed to support the most common validation constraints, such as Scoped Negation As Failure (SNAF), and the predicates defined in the Rule Interchange Format (RIF). We prove the feasibility of the approach by providing an implementation in Notation3 Logic. As such, we show that rule logic can cover both validation and reasoning if it is expressive enough.

Sustainable Linked Data Generation: The Case of DBpedia

dbpedia ef, the generation framework behind one of the Linked Open Data cloud’s central interlinking hubs, has limitations with regard to quality, coverage and sustainability of the generated dataset. dbpedia can be further improved both on schema and data level. Errors and inconsistencies can be addressed by amending (i) the dbpedia ef; (ii) the dbpedia mapping rules; or (iii) Wikipedia itself from which it extracts information. However, even though the dbpedia ef and mapping rules are continuously evolving and several changes were applied to both of them, there are no significant improvements on the dbpedia dataset since its limitations were identified. To address these shortcomings, we propose adapting a different semantic-driven approach that decouples, in a declarative manner, the extraction, transformation and mapping rules execution. In this paper, we provide details regarding the new dbpedia ef, its architecture, technical implementation and extraction results. This way, we achieve an enhanced data generation process, which can be broadly adopted, and that improves its quality, coverage and sustainability.

Towards robust and reliable multimedia analysis through semantic integration of services

Thanks to ubiquitous Web connectivity and portable multimedia devices, it has never been so easy to produce and distribute new multimedia resources such as videos, photos, and audio. This ever-increasing production leads to an information overload for consumers, which calls for efficient multimedia retrieval techniques. Multimedia resources can be efficiently retrieved using their metadata, but the multimedia analysis methods that can automatically generate this metadata are currently not reliable enough for highly diverse multimedia content. A reliable and automatic method for analyzing general multimedia content is needed. We introduce a domain-agnostic framework that annotates multimedia resources using currently available multimedia analysis methods. By using a three-step reasoning cycle, this framework can assess and improve the quality of multimedia analysis results, by consecutively (1) combining analysis results effectively, (2) predicting which results might need improvement, and (3) invoking compatible analysis methods to retrieve new results. By using semantic descriptions for the Web services that wrap the multimedia analysis methods, compatible services can be automatically selected. By using additional semantic reasoning on these semantic descriptions, the different services can be repurposed across different use cases. We evaluated this problem-agnostic framework in the context of video face detection, and showed that it is capable of providing the best analysis results regardless of the input video. The proposed methodology can serve as a basis to build a generic multimedia annotation platform, which returns reliable results for diverse multimedia analysis problems. This allows for better metadata generation, and improves the efficient retrieval of multimedia resources.

Reconnecting Digital Publications to the Web using their Spatial Information

Digital publications can be packaged and viewed via the Open Web Platform using the EPUB 3 format. Meanwhile, the increased amount of mobile clients and the advent of HTML5s Geolocation have opened a whole range of possibilities for digital publications to interact with their readers. However, EPUB 3 files often remain closed silos of information, no longer linked with the rest of the Web. In this paper, we propose a solution to reconnect digital publication with the (Semantic) Web. We will also show how we can use that connection to improve contextualization for a user, specifically via spatial information. We enrich digital publications by connecting the detected concepts to their URIs on, e.g., DBpedia, and by devising an algorithm to approximate the location of any detected concept, we can provide a user with the spatial center of gravity of his reading position. The evaluation of the location approximation algorithm showed a high recall, and the high correlation between estimation error and standard deviation can provide the user with a sense of correctness (or spread) of an approximation. This means relevant locations (and their possible radius) can be shown for a user, based on the content he or she is reading, and based on his or her location. This methodology can be used to reconnect digital publications with the online world, to entice readers, and ultimately, as a novel location-based recommendation technique.