Search Results (1 - 11 of 11 Results)

Sort By  
Sort Dir
 
Results per page  

Sengupta, KunalA Language for Inconsistency-Tolerant Ontology Mapping
Doctor of Philosophy (PhD), Wright State University, 2015, Computer Science and Engineering PhD
Ontology alignment plays a key role in enabling interoperability among various data sources present in the web. The nature of the world is such, that the same concepts differ in meaning, often so slightly, which makes it difficult to relate these concepts. It is the omni-present heterogeneity that is at the core of the web. The research work presented in this dissertation, is driven by the goal of providing a robust ontology alignment language for the semantic web, as we show that description logics based alignment languages are not suitable for aligning ontologies. The adoption of the semantic web technologies has been consistently on the rise over the past decade, and it continues to show promise. The core component of the semantic web is the set of knowledge representation languages -- mainly the W3C (World Wide Web Consortium) standards Web Ontology Language (OWL), Resource Description Framework (RDF), and Rule Interchange Format (RIF). While these languages have been designed in order to be suitable for the openness and extensibility of the web, they lack certain features which we try to address in this dissertation. One such missing component is the lack of non-monotonic features, in the knowledge representation languages, that enable us to perform common sense reasoning. For example, OWL supports the open world assumption (OWA), which means that knowledge about everything is assumed to be possibly incomplete at any point of time. However, experience has shown that there are situations that require us to assume that certain parts of the knowledge base are complete. Employing the Closed World Assumption (CWA) helps us achieve this. Circumscription is a very well-known approach towards CWA, which provides closed world semantics by employing the idea of minimal models with respect to certain predicates which are closed. We provide the formal semantics of the notion of Grounded Circumscription, which is an extension of circumscription with desirable properties like decidability. We also provide a tableaux calculus to reason over knowledge bases under the notion of grounded circumscription. Another form of common sense logic, is default logic. Default logic provides a way to specify rules that, by default, hold in most cases but not necessarily in all cases. The classic example of such a rule is: If something is a bird then it flies. The power of defaults comes from the ability of the logic to handle exceptions to the default rules. For example, a bird will be assumed to fly by default unless it is an exception, i.e. it belongs to a class of birds that do not fly, like penguins. Interestingly, this property of defaults can be utilized to create mappings between concepts of different ontologies (knowledge bases). We provide a new semantics for the integration of defaults in description logics and show that it improves upon previously known results in literature. In this study, we give various examples to show the utility and advantages of using a default logic based ontology alignment language. We provide the semantics and decidability results of a default based mapping language for tractable fragments of description logics (or OWL). Furthermore, we provide a proof of concept system and qualitative analysis of the results obtained from the system when compared to that of traditional mapping repair techniques.

Committee:

Pascal Hitzler, Ph.D. (Advisor); Krzysztof Janowicz, Ph.D. (Committee Member); Krishnaprasad Thirunarayan, Ph.D. (Committee Member); Prabhaker Mateti, Ph.D. (Committee Member)

Subjects:

Computer Science

Keywords:

description logic; ontology mapping language; ontology alignment lanuguage; ontology alignment; ontology Mapping; non monotonic reasoning; default logic; circumscription; semantic web; web ontology language

Cheatham, Michelle AndreenThe Properties of Property Alignment on the Semantic Web
Doctor of Philosophy (PhD), Wright State University, 2014, Computer Science and Engineering PhD
Ontology alignment is an important step in enabling computers to query and reason across the many linked datasets on the semantic web. This is a difficult challenge because the ontologies underlying different linked datasets can vary in terms of subject area coverage, level of abstraction, ontology modeling philosophy, and even language. The alignment approach presented here centers on string similarity metrics. Nearly all ontology alignment systems use a string similarity metric in one form or another, but it seems that the choice of a particular metric is often arbitrary. We begin this dissertation with the most comprehensive survey to date on the performance of string similarity metrics and string preprocessing strategies for ontology alignment. Based on this work we present practical guidelines for choosing string metrics in the face of different types of ontologies and different alignment goals. Additionally, we show that string similarity metrics alone can perform competitively with state-of-the-art alignment systems on the most popular benchmarks in the field. One of the contributions of our string similarity metric survey is quantification of the difference in performance between aligning classes and aligning properties (relations between classes). Put simply: aligning properties is hard, and existing string similarity metrics are not of great help. We therefore take on the task of developing a new string-based alignment approach that performs better on properties. Unfortunately, evaluating that approach is difficult because the only existing alignment benchmark that includes properties is, in our view, unrealistic since all relations in the reference alignment are presented as completely certain. Human experts do not have this degree of confidence when asked to align an ontology. We therefore present a more nuanced version of this benchmark that we have created through a combination of expert survey and crowdsourcing. We then present our new string-based property alignment system and evaluate its performance on both the current benchmark and our proposed revision. Our property-centric string metric can be configured for either high precision or high recall. The results show a five-fold increase in precision and a doubling of recall over an approach based on the best current string metric. Finally, we apply our system to a real-world test case and analyze the results.

Committee:

Pascal Hitzler, Ph.D. (Advisor); Isabel Cruz, Ph.D. (Committee Member); Krishnaprasad Thirunarayan, Ph.D. (Committee Member); Mateen Rizki, Ph.D. (Committee Member)

Subjects:

Computer Science

Keywords:

ontology alignment; string similarity metrics; semantic data integration

Amini, ReihanehTowards Best Practices for Crowdsourcing Ontology Alignment Benchmarks
Master of Science (MS), Wright State University, 2016, Computer Science
Ontology alignment systems establish the semantic links between ontologies that enable knowledge from various sources and domains to be used by automated applications in many different ways. Unfortunately, these systems are not perfect. Currently, the results of even the best-performing automated alignment systems need to be manually verified in order to be fully trusted. Ontology alignment researchers have turned to crowdsourcing platforms such as Amazon's Mechanical Turk to accomplish this. However, there has been little systematic analysis of the accuracy of crowdsourcing for alignment verification and the establishment of best practices. In this work, we analyze the impact of the presentation of the context of potential matches and the way in which the question is presented to workers on the accuracy of crowdsourcing for alignment verification. Our overall recommendations are that users interested in high precision are likely to achieve the best results by presenting the definitions of the entity labels and allowing workers to respond with true/false to the question of whether or not an equivalence relationship exists. Conversely, if the alignment researcher is interested in high recall, they are better o presenting workers with a graphical depiction of the entity relationships and a set of options about the type of relation that exists, if any.

Committee:

Michelle Cheatham, Ph.D. (Advisor); Mateen M. Rizki, Ph.D. (Committee Member); Derek Doran, Ph.D. (Committee Member)

Subjects:

Computer Science

Keywords:

Ontology Alignment, Crowdsourcing, Mechanical Turk

Xia, WeiguoAUTOMATIC SELECTION OF MEDIATING ONTOLOGY FOR ALIGNING BIOMEDICAL ONTOLOGIES
Master of Computer Science, Miami University, 2015, Computer Science & Software Engineering
Ontologies are increasingly important in the Semantic Web and biomedical information system fields. Ontology alignment (OA) is the process of finding semantic mappings between the concepts of two given ontologies. OA systems have begun using mediating ontologies pre-selected to improve OA performance. This research investigates the automatic selection of a set of mediating ontologies from a large set of ontologies in the biomedical domain. BioPortal, an online library offering biomedical ontologies via web API and web browsing, is used as the background knowledge source. The anatomy and the large biomedical ontologies tracks of the Ontology Alignment Evaluation Initiative are used to evaluate this approach which is implemented as a software component accessed from a leading OA system LogMap. The experimental results show automatically selected mediating ontologies improve the recall and f-measure for the anatomy track. For the large biomedical ontologies track, three of the six tasks show some overall improvement.

Committee:

Valerie Cross (Advisor); Dhananjai Rao (Committee Member); Ernesto Jimenez-Ruiz (Committee Member)

Subjects:

Computer Science

Keywords:

ontology; mediating ontology; Bio-portal; ontology alignment;

McCurdy, Helena BrookeWikiMatcher: Leveraging Wikipedia for Ontology Alignment
Master of Science in Computer Engineering (MSCE), Wright State University, 2016, Computer Engineering
As the Semantic Web grows, so does the number of ontologies used to structure the data within it. Aligning these ontologies is critical to fully realizing the potential of the web. Previous work in ontology alignment has shown that even alignment systems utilizing basic string similarity metrics can produce useful matches. Researchers speculate that including semantic as well as syntactic information inherent in entity labels can further improve alignment results. This paper examines that hypothesis by exploring the utility of using Wikipedia as a source of semantic information. Various elements of Wikipedia are considered, including article content, page terms, and search snippets. The utility of each information source is analyzed and a composite system, WikiMatcher, is created based on this analysis. The performance of WikiMatcher is compared to that of a basic string-based alignment system on two established alignment benchmarks and two other real-world datasets. The extensive evaluation shows that although WikiMatcher performs similarly to that of the string metric overall, it is able to find many matches with no syntactic similarity between labels. This performance seems to be driven by Wikipedia's query resolution and page redirection system, rather than by the particular information from Wikipedia that is used to compare entities.

Committee:

Michelle Cheatham, Ph.D. (Advisor); Mateen Rizki, Ph.D. (Committee Member); Krishnaprasad Thirunarayan, Ph.D. (Committee Member)

Subjects:

Computer Engineering; Computer Science

Keywords:

ontology alignment; wikipedia

Gu, ChenOntology Alignment Techniques for Linked Open Data Ontologies
Master of Computer Science, Miami University, 2013, Computer Science & Software Engineering
Ontology alignment (OA) addresses the Semantic Web challenge to enable information interoperability between related but heterogeneous ontologies. Traditional OA systems have focused on aligning well defined and structured ontologies from the same or closely related domains and producing equivalence mappings between concepts in the source and target ontologies. Linked Open Data (LOD) ontologies, however, present some different characteristics from standard ontologies. For example, equivalence relations are limited among LOD concepts; thus OA systems for LOD ontology alignment should be able to produce subclass and superclass mappings between the source and target. This thesis overviews the current research on aligning LOD ontologies. An important research aspect is the use of background knowledge in the alignment process. Two current OA systems are modified to perform alignment of LOD ontologies. For each modified OA system, experiments have been performed using a set of LOD reference alignments to evaluate their alignment results using standard OA performance measures.

Committee:

Valerie Cross, PhD (Advisor); Eric Bachmann, PhD (Committee Member); Michael Zmuda, PhD (Committee Member)

Subjects:

Computer Science

Keywords:

linked open data, ontology alignment, background knowledge

Silwal, PramitONTOLOGY ALIGNMENT USING SEMANTIC SIMILARITY WITH REFERENCE ONTOLOGIES
Master of Computer Science, Miami University, 2012, Computer Science & Software Engineering
I would like to give my sincere thanks and appreciation to Prof. Valerie Cross for her advice, support and constant encouragement. Without her guidance, I could not have finished this work. I also want to thank my thesis committee members, Prof. Michael Zmuda and Prof. James Kiper for their time in reviewing this work. I would also like to thank Cosmin Stroe from UIC for his support with the Agreementmaker software. Lastly I would like to thank my family and friends for their constant encouragement and support.

Committee:

Valerie Cross (Advisor); James Kiper (Committee Member); Michael Zmuda (Committee Member)

Subjects:

Computer Science

Keywords:

Ontology Alignment; semantic similarity, Mediating Ontology; OAEI

Chen, XiExploiting BioPortal as Background Knowledge in Ontology Alignment
Master of Science, Miami University, 2014, Computer Science & Software Engineering
Ontology alignment (OA) is the process of taking as input two ontologies and producing mappings between the source concepts and the target concepts. Over the last few years, OA systems have made only minor improvements. To improve performance, some OA systems have included a semi-automatic matching approach which incorporates user interaction to assess low confidence mappings. This research investigates replacing the human expert with an automated expert or “oracle” that relies on specialized knowledge sources in the biomedical domain, BioPortal. BioPortal provides access to different resources including a wide variety of ontologies, classes within ontologies and mappings between the classes of different ontologies. A leading OA system LogMap has been used to evaluate the automated expert on the anatomy and Large Biomed Track of the Ontology Alignment Evaluation Initiative (OAEI). The experimental results are reported and show that the automated expert has a positive impact in the Large Biomed Track with four out of six of the track’s matching tasks having better OA standard performance measure for F-measure. In the Anatomy Track, using the automated expert improves the OA standard performance measure for precision. However, to the detriment of the recall measure, the result is a slight improvement in the F-measure.

Committee:

Valerie Cross (Advisor); Ernesto Jimenez-Ruiz (Committee Member); Dhananjai Rao (Committee Member)

Subjects:

Computer Science

Keywords:

ontology alignment, ontology matching, human expert, automated expert, semi-automatic matching, user interaction, knowledge source, biomedical, BioPortal, LogMap, background knowledge, perfect oracle, error rate

Hu, XuehengSEMANTIC SIMILARITY IN THE EVALUATION OF ONTOLOGY ALIGNMENT
Master of Computer Science, Miami University, 2011, Computer Science & Software Engineering
A challenge for the Semantic Web is enabling information interoperability between related but heterogeneous ontologies. Ontology alignment (OA) addresses this challenge by identifying correspondences between entities in different ontologies. Traditional OA evaluation strategies use a gold standard reference alignment determined by a domain expert and considered correct and complete. Such strategies suffer a drawback. In many cases a reference alignment is not available. This thesis overviews OA evaluation methods and proposes the semantic alignment quality (SAQ) measure that utilizes semantic similarity measures in OA evaluation. The performance of SAQ is compared to standard OA quality measures including precision, recall, and the f-measure. This investigation relies on existing OA results from the OA evaluation initiative (OAEI). The results indicate much variation in SAQ performance depending on the selected semantic similarity measure. Its high correlation with the traditional OA precision measure does not necessarily support its use as an OA performance measure.

Committee:

Valerie Cross (Advisor); Michael Zmuda (Committee Member); Keith Frikken (Committee Member)

Subjects:

Computer Science

Keywords:

Semantic similarity; ontological similarity; ontology alignment; information content; semantic alignment quality

Joshi, Amit KrishnaExploiting Alignments in Linked Data for Compression and Query Answering
Doctor of Philosophy (PhD), Wright State University, 2017, Computer Science and Engineering PhD
Linked data has experienced accelerated growth in recent years due to its interlinking ability across disparate sources, made possible via machine-processable RDF data. Today, a large number of organizations, including governments and news providers, publish data in RDF format, inviting developers to build useful applications through reuse and integration of structured data. This has led to tremendous increase in the amount of RDF data on the web. Although the growth of RDF data can be viewed as a positive sign for semantic web initiatives, it causes performance bottlenecks for RDF data management systems that store and provide access to data. In addition, a growing number of ontologies and vocabularies make retrieving data a challenging task. The aim of this research is to show how alignments in the Linked Data can be exploited to compress and query the linked datasets. First, we introduce two compression techniques that compress RDF datasets through identification and removal of semantic and contextual redundancies in linked data. Logical Linked Data Compression is a lossless compression technique which compresses a dataset by generating a set of new logical rules from the dataset and removing triples that can be inferred from these rules. Contextual Linked Data Compression is a lossy compression technique which compresses datasets by performing schema alignment and instance matching followed by pruning of alignments based on confidence value and subsequent grouping of equivalent terms. Depending on the structure of the dataset, the first technique was able to prune more than 50% of the triples. Second, we propose an Alignment based Linked Open Data Querying System (ALOQUS) that allows users to write query statements using concepts and properties not present in linked datasets and show that querying does not require a thorough understanding of the individual datasets and interconnecting relationships. Finally, we present LinkGen, a multipurpose synthetic Linked Data generator that generates a large amount of repeatable and reproducible RDF data using statistical distribution, and interlinks with real world entities using alignments.

Committee:

Pascal Hitzler , Ph.D. (Advisor); Guozhu Dong, Ph.D. (Committee Member); Krishnaprasad Thirunaraya, Ph.D. (Committee Member); Michelle Cheatham, Ph.D. (Committee Member); Subhashini Ganapathy, Ph.D. (Committee Member)

Subjects:

Computer Science

Keywords:

Linked Data; RDF Compression; Ontology Alignment; Linked Data Querying; Synthetic RDF Generator; SPARQL

Pramit, SilwalOntology Alignment using Semantic Similarity with Reference Ontologies
Master of Computer Science, Miami University, 2012, Computer Science & Software Engineering
Measuring the similarity between concepts in two different ontologies can be done in many different ways as evidenced by the numerous kinds of matchers found in ontology alignment (OA) systems. Some produce a mapping between two concepts in different ontologies by finding an identical bridge concept in a reference ontology to which both concepts are mapped. A new matcher incorporating semantic similarity measurement within one or more reference ontologies has been integrated into the AgreementMaker OA system. Experiments using the Ontology Alignment Evaluation Initiative (OAEI) anatomy track are performed with both the Uberon and the FMA ontologies as reference ontologies. The results of these experiments are compared to the OAEI 2011 and the 2012 results for the anatomy track and show this approach performs better than most other OA systems in the OAEI competition. This thesis demonstrates that using semantic similarity measures within a reference ontology can improve the ontology alignment process.

Committee:

Valerie Cross (Advisor); Michael Zmuda (Committee Member); James Kiper (Committee Member)

Subjects:

Computer Science

Keywords:

ontology alignment; reference ontologies; semantic similarity