Jan Noessner

Leveraging terminological structure for object reconciliation

It has been argued that linked open data is the major benefit of semantic technologies for the web as it provides a huge amount of structured data that can be accessed in a more effective way than web pages. While linked open data avoids many problems connected with the use of expressive ontologies such as the knowledge acquisition bottleneck, data heterogeneity remains a challenging problem. In particular, identical objects may be referred to by different URIs in different data sets. Identifying such representations of the same object is called object reconciliation. In this paper, we propose a novel approach to object reconciliation that is based on an existing semantic similarity measure for linked data. We adapt the measure to the object reconciliation problem, present exact and approximate algorithms that efficiently implement the methods, and provide a systematic experimental evaluation based on a benchmark dataset. As our main result, we show that the use of light-weight ontologies and schema information significantly improves object reconciliation in the context of linked open data.

Log-linear description logics

Log-linear description logics are a family of probabilistic logics integrating various concepts and methods from the areas of knowledge representation and reasoning and statistical relational AI. We define the syntax and semantics of log-linear description logics, describe a convenient representation as sets of first-order formulas, and discuss computational and algorithmic aspects of probabilistic queries in the language. The paper concludes with an experimental evaluation of an implementation of a log-linear DL reasoner.

Benchmarking matching applications on the semantic Web

The evaluation of matching applications is becoming a major issue in the semantic web and it requires a suitable methodological approach as well as appropriate benchmarks. In particular, in order to evaluate a matching application under different experimental conditions, it is crucial to provide a test dataset characterized by a controlled variety of different heterogeneities among data that rarely occurs in real data repositories. In this paper, we propose SWING (Semantic Web INstance Generation), a disciplined approach to the semi-automatic generation of benchmarks to be used for the evaluation of matching applications.

ELOG: a probabilistic reasoner for OWL EL

Log-linear description logics are probabilistic logics combining several concepts and methods from the areas of knowledge representation and reasoning and statistical relational AI. We describe some of the implementation details of the log-linear reasoner ELOG. The reasoner employs database technology to dynamically transform inference problems to integer linear programs (ILP). In order to lower the size of the ILPs and reduce the complexity we employ a form of cutting plane inference during reasoning.

Probabilistic-logical web data integration

The integration of both distributed schemas and data repositories is a major challenge in data and knowledge management applications. Instances of this problem range from mapping database schemas to object reconciliation in the linked open data cloud. We present a novel approach to several important data integration problems that combines logical and probabilistic reasoning. We first provide a brief overview of some of the basic formalisms such as description logics and Markov logic that are used in the framework. We then describe the representation of the different integration problems in the probabilistic-logical framework and discuss efficient inference algorithms. For each of the applications, we conducted extensive experiments on standard data integration and matching benchmarks to evaluate the efficiency and performance of the approach. The positive results of the evaluation are quite promising and the flexibility of the framework makes it easily adaptable to other realworld data integration problems.

An infrastructure for probabilistic reasoning with web ontologies

We present an infrastructure for probabilistic reasoning with ontologies based on our Markov logic engine RockIt. Markov logic is a template language that combines first-order logic with log-linear graphical models. We show how to translate OWL-EL as well as RDF schema to Markov logic and how to use RockIt for applying MAP inference on the given set of formulas. The resulting system is an infrastructure for log linear logics that can be used for probabilistic reasoning with both extended OWL-EL and RDF schema. We describe our system and illustrate its benefits by presenting experimental results for two types of applications. These are ontology matching and knowledge base verification, with a special focus on temporal reasoning. Moreover, we illustrate two further use cases which are Activity Recognition and Root Cause Analysis. Our infrastructure has been applied to these use cases in the context of a cooperation with industry partners. The experiments indicate that our system, which is based on a well-founded probabilistic semantics, is capable of solving relevant problems as good as or better than state of the art systems that have specifically been designed for the respective problem. The heterogeneity of the presented uses cases illustrates the wide applicability of our infrastructure.

LODE: linking digital humanities content to the web of data

Numerous digital libraries projects maintain their data collections in the form of text, images, and metadata. While data may be stored in many formats, from plain text to XML to relational databases, the use of the resource description framework (RDF) as a standardized representation has gained considerable traction during the last five years. Almost every digital humanities meeting has at least one session concerned with the topic of digital humanities, RDF, and linked data, including JCDL. While most existing work in linked data has focused on improving algorithms for entity matching, the aim of our Linked Open Data Enhancer Lode is to work “out of the box”, enabling their use by humanities scholars, computer scientists, librarians, and information scientists alike. With Lode we enable non-technical users to enrich a local RDF repository with high-quality data from the Linked Open Data cloud. Lode links and enhances the local RDF repository without reducing the quality of the data. In particular, we support the user in the enhancement and linking process by providing intuitive user-interfaces and by suggesting high quality linking candidates using state of the art matching algorithms. We hope that the Lode framework will be useful to digital humanities scholars complementing other digital humanities tools.

Coherent top-k ontology alignment for OWL EL

The integration of distributed information sources is a key challenge in data and knowledge management applications. Instances of this problem range from mapping schemas of heterogeneous databases to object reconciliation in linked open data repositories. In this paper, we approach the problem of aligning description logic ontologies. We focus particularly on the problem of computing coherent alignments, that is, alignments that do not lead to unsatisfiable classes in the resulting merged ontologies. We believe that considering coherence during the alignment process is important as it is this logical concept that distinguishes ontology alignment from other data integration problems. Depending on the heterogeneity of the ontologies it is often more reasonable to generate alignments with at most k correspondences because not every entity has a matchable counterpart. We describe both greedy and optimal algorithms for computing coherent top-k alignments between OWL EL ontologies and assess their performance relative to state-of-the-art matching systems.

User-Centric Data Integration with the MappingAssistant

Data integration is the problem of transferring complex data from one into another representation in order to support exchange between different systems. From a technical point of view data integration has intensively been studied. However, less attention has been paid to user-centric aspects of data integration. In this work, we present the MappingAssistant which supports the user in finding and validating mapping rules between heterogeneous data sources. Compared to existing approaches we focus on an user-centric approach where the user inspects the consequences of the data integration rules on the instance level rather than being confronted with complex data integration rules. We performed a study which shows that the user-centric approach leads to better integration results, especially for users with little or no technical knowledge and is perceived as being more intuitive.