Bounded-memory computability continues to be in the focus of those areas of AI and databases that deal with feasible computations over streams be it feasible arithmetical calculations on low-level streams or feasible query answering for declaratively specified queries on relational data streams or even feasible query answering for high-level queries on streams w.r.t. a set of constraints in an ontology such as in the paradigm of Ontology-Based Data Access (OBDA). In classical OBDA, a high-level query is answered by transforming it into a query on data source level. The transformation requires a rewriting step, where knowledge from an ontology is incorporated into the query, followed by an unfolding step with respect to a set of mappings. Given an OBDA setting it is very difficult to decide, whether and how a query can be answered efficiently. In particular it is difficult to decide whether a query can be answered in bounded memory, i.e., in constant space w.r.t. an infinitely growing prefix of a data stream.

We focus on handling conflicting and uncertain information in lightweight ontologies, where uncertainty is represented in a possibilistic logic setting. We use DL-Lite, a tractable fragment of Description Logic, to specify terminological knowledge (i.e., TBox). We assume the TBox to be stable and coherent, while its combination with a set of assertional facts (i.e., ABox) may be inconsistent. We address the problem of dealing with conflicts when the reliability relation between sources is only partially ordered. We propose to represent the uncertain ABox as a symbolic weighted base, where a strict partial preorder is applied on the weights.

There is a growing need for scalable semantic web repositories which support inference and provide efficient queries. There is also a growing interest in representing uncertain knowledge in semantic web datasets and ontologies. In this paper, I present a bit vector schema specifically designed for RDF (Resource Description Framework) datasets. I propose a system for materializing and storing inferred knowledge using this schema. I show experimental results that demonstrate that this solution simplifies inference queries and drastically improves results. I also propose and describe a solution for materializing and persisting uncertain information and probabilities. Thresholds and bit vectors are used to provide efficient query access to this uncertain knowledge. My goal is to provide a semantic web repository that supports knowledge inference, uncertainty reasoning, and Bayesian networks, without sacrificing performance or scalability.

We describe an approach to reducing the computational cost of identifying coreferent instances in heterogeneous semantic graphs where the underlying ontologies may not be informative or even known. The problem is similar to coreference resolution in unstructured text, where a variety of linguistic clues and contextual information is used to infer entity types and predict coreference. Semantic graphs, whether in RDF or another formalism, are semi-structured data with very different contextual clues and need different approaches to identify potentially coreferent entities. When their ontologies are unknown, inaccessible or semantically trivial, coreference resolution is difficult. For such cases, we can use supervised machine learning to map entity attributes via dictionaries based on properties from an appropriate background knowledge base to predict instance entity types, aiding coreference resolution. We evaluated the approach in experiments on data from Wikipedia, Freebase and Arnetminer and DBpedia as the background knowledge base.

The success of the Web of Data (WOD) is based on the thorough understanding of, and agreement upon, the se- mantics of data and ontologies. But the Web of Data as a whole is complex, and inherently messy, contex- tualised, opinionated, in short: it is a market-place of ideas, rather than a database. Existing paradigms are in- appropriate for dealing with this new type of knowledge structures. The urgency of dealing with the non-standard charac- teristics of the Web of Data has been recognised, and separate initiatives try to tackle its individual manifes- tations, e.g.

When the amount of RDF data is very large, it becomes more likely that the triples describing entities will contain errors and may not include the specification of a class from a known ontology. The work presented here explores the utilization of methods from machine learning to develop classifiers for identifying the semantic categorization of entities based upon the property names used to describe the entity. The goal is to develop classifiers that are accurate, but robust to errors and noise. The training data comes from DBpedia, where entities are categorized by type and densely described with RDF properties. The initial experimentation reported here indicates that the approach is promising.

Much of the focus on big data has been on the problem of processing very large sources. There is an equally hard problem of how to normalize, integrate, and transform the data from many sources into the format required to run large-scale analysis and visualization tools. We have previously developed an approach to semi-automatically mapping diverse sources into a shared domain ontology so that they can be quickly combined. In this paper we describe our approach to building and executing integration and restructuring plans to support analysis and visualization tools on very large and diverse datasets.

We present an implemented approach to transform natural language sentences into SPARQL, using background knowledge from ontologies and lexicons. Therefore, eligible technologies and data storage possibilities are analyzed and evaluated. The contributions of this paper are twofold. Firstly, we describe the motivation and current needs for a natural language access to industry data. We describe several scenarios where the proposed solution is required.

Privacy policies are used to communicate company data practices to consumers and must be accurate and comprehensive. Each policy author is free to use their own nomenclature when describing data practices, which leads to different ways in which similar information types are described across policies. A formal ontology can help policy authors, users and regulators consistently check how data practice descriptions relate to other interpretations of information types. In this paper, we describe an empirical method for manually constructing an information type ontology from privacy policies. The method consists of seven heuristics that explain how to infer hypernym, meronym and synonym relationships from information type phrases, which we discovered using grounded analysis of five privacy policies. The method was evaluated on 50 mobile privacy policies which produced an ontology consisting of 355 unique information type names. Based on the manual results, we describe an automated technique consisting of 14 reusable semantic rules to extract hypernymy, meronymy, and synonymy relations from information type phrases. The technique was evaluated on the manually constructed ontology to yield .95 precision and .51

Knowledge graph completion (a.k.a.~link prediction), i.e.,~the task of inferring missing information from knowledge graphs, is a widely used task in many applications, such as product recommendation and question answering. The state-of-the-art approaches of knowledge graph embeddings and/or rule mining and reasoning are data-driven and, thus, solely based on the information the input knowledge graph contains. This leads to unsatisfactory prediction results which make such solutions inapplicable to crucial domains such as healthcare. To further enhance the accuracy of knowledge graph completion we propose to loosely-couple the data-driven power of knowledge graph embeddings with domain-specific reasoning stemming from experts or entailment regimes (e.g., OWL2). In this way, we not only enhance the prediction accuracy with domain knowledge that may not be included in the input knowledge graph but also allow users to plugin their own knowledge graph embedding and reasoning method. Our initial results show that we enhance the MRR accuracy of vanilla knowledge graph embeddings by up to 3x and outperform hybrid solutions that combine knowledge graph embeddings with rule mining and reasoning up to 3.5x MRR.