Search

A large amount of information is stored in data tables. Users can search for data tables using a keyword-based query. A table is composed primarily of data values that are organized in rows and columns providing implicit structural information. A table is usually accompanied by secondary information such as the caption, page title, etc., that form the textual information. Understanding the connection between the textual and structural information is an important yet neglected aspect in table retrieval as previous methods treat each source of information independently. In addition, users can search for data tables that are similar to an existing table, and this setting can be seen as a content-based table retrieval. In this paper, we propose StruBERT, a structure-aware BERT model that fuses the textual and structural information of a data table to produce context-aware representations for both textual and tabular content of a data table. StruBERT features are integrated in a new end-to-end neural ranking model to solve three table-related downstream tasks: keyword- and content-based table retrieval, and table similarity. We evaluate our approach using three datasets, and we demonstrate substantial improvements in terms of retrieval and classification metrics over state-of-the-art methods., Comment: The Proceddings of The ACM Web Conference 2022

Table search aims to retrieve a list of tables given a user's query. Previous methods only consider the textual information of tables and the structural information is rarely used. In this paper, we propose to model the complex relations in the table corpus as one or more graphs and then utilize graph neural networks to learn representations of queries and tables. We show that the text-based table retrieval methods can be further improved by graph-based predictions which fuse multiple field-level information.

Hierarchical Task Network (HTN) planning uses task-subtask relationships to break complex problems into more manageable subtasks, similar to how human problem-solvers plan. However, one limitation of HTN planning is that it requires domain knowledge in the form of planning methods to perform this task decomposition. Recent work has partially alleviated this knowledge engineering requirement by learning HTN methods from traces of observed behavior. Although this greatly reduces the amount of knowledge that must be encoded by a domain expert, it requires a large collection of traces in order to infer important landmark states that are used during trace segmentation and method learning. In this paper we present a novel method for landmark inference that transfers knowledge of landmarks from previously encountered environments to new environments without requiring any traces from the new environment. We evaluate our work in a logistics planning domain and show that our approach performs comparably to the existing landmark inference method but requires far fewer traces.

Ranking models are the main components of information retrieval systems. Several approaches to ranking are based on traditional machine learning algorithms using a set of hand-crafted features. Recently, researchers have leveraged deep learning models in information retrieval. These models are trained end-to-end to extract features from the raw data for ranking tasks, so that they overcome the limitations of hand-crafted features. A variety of deep learning models have been proposed, and each model presents a set of neural network components to extract features that are used for ranking. In this paper, we compare the proposed models in the literature along different dimensions in order to understand the major contributions and limitations of each model. In our discussion of the literature, we analyze the promising neural components, and propose future research directions. We also show the analogy between document retrieval and other retrieval tasks where the items to be ranked are structured documents, answers, images and videos., Published in the Information Retrieval Journal (2021)

Pretrained contextualized language models such as BERT have achieved impressive results on various natural language processing benchmarks. Benefiting from multiple pretraining tasks and large scale training corpora, pretrained models can capture complex syntactic word relations. In this paper, we use the deep contextualized language model BERT for the task of ad hoc table retrieval. We investigate how to encode table content considering the table structure and input length limit of BERT. We also propose an approach that incorporates features from prior literature on table retrieval and jointly trains them with BERT. In experiments on public datasets, we show that our best approach can outperform the previous state-of-the-art method and BERT baselines with a large margin under different evaluation metrics., Accepted at SIGIR 2020 (Long)

A search engine’s ability to retrieve desirable datasets is important for data sharing and reuse. Existing dataset search engines typically rely on matching queries to dataset descriptions. However, a user may not have enough prior knowledge to write a query using terms that match with description text. We propose a novel schema label generation model which generates possible schema labels based on dataset table content. We incorporate the generated schema labels into a mixed ranking model which not only considers the relevance between the query and dataset metadata but also the similarity between the query and generated schema labels. To evaluate our method on real-world datasets, we create a new benchmark specifically for the dataset retrieval task. Experiments show that our approach can effectively improve the precision and NDCG scores of the dataset retrieval task compared with baseline methods. We also test on a collection of Wikipedia tables to show that the features generated from schema labels can improve the unsupervised and supervised web table retrieval task as well.

Due to the decentralized nature of the Semantic Web, the same real-world entity may be described in various data sources with different ontologies and assigned syntactically distinct identifiers. In order to facilitate data utilization and consumption in the Semantic Web, without compromising the freedom of people to publish their data, one critical problem is to appropriately interlink such heterogeneous data. This interlinking process is sometimes referred to as Entity Matching , i.e., finding which identifiers refer to the same real-world entity. In this paper, we propose two candidate selection algorithms to improve the scalability of entity matching systems. First of all, we propose HistSim that utilizes the matching histories of the instances to prune instance pairs that are not sufficiently similar to the same pool of other instances. A sigmoid function based thresholding method is proposed to automatically adjust the threshold for such commonality on-the-fly. Furthermore, we propose DisNGram that selects candidate instance pairs by computing a character-level similarity metric on discriminating literal values that are chosen using domain-independent unsupervised learning. Instances are indexed on the chosen predicates’ literal values to enable efficient look-up for similar instances. Finally, in order to be able to handle heterogeneous datasets with a large number of predicates, a mechanism for automatically determining predicate comparability is proposed. We evaluate our two candidate selection algorithms against six state-of-the-art systems on three Semantic Web datasets, and demonstrate that our proposed algorithms frequently outperform state-of-the-art systems on F1-score and runtime.

Impoverished descriptions and convoluted schema labels are common challenges in data-centric tasks such as schema matching and data linking, especially when datasets can span domains. To address these issues, we consider the task of schema label generation. Typically, schema labels are created by dataset providers and are useful for users to understand a dataset. The motivation behind the task is that a lot of data linking systems require overlapping information between two datasets and rely on unique identifiers of schema labels. Moreover, it is common for schema labels in different datasets to have different identifiers even when they refer to the same concept. With no naming standard for schema labels, unintelligible labels are widely found in real-world datasets. For example, many schema labels contain abbreviations and compound nouns that hinder automated matching of attributes in corresponding datasets. Through schema label generation, more common (and thus understandable) schema labels can be provided to allow for broader schema matches in contexts such as dataset search and data linking. We develop a variety of features based on analysis of dataset content to enable machine learning methods to recommend useful labels. We test our approach on two real-world data collections and demonstrate that our method is able to outperform the alternative approach.

A key challenge for the semantic Web is to acquire the capability to effectively query large knowledge bases. As there will be several competing systems, we need benchmarks that will objectively evaluate these systems. Development of effective benchmarks in an emerging domain is a challenging endeavor. In this paper, we propose a requirements driven framework for developing benchmarks for semantic Web knowledge base systems (SW KBSs). In this paper, we make two major contributions. First, we provide a list of requirements for SW KBS benchmarks. This can serve as an unbiased guide to both the benchmark developers and personnel responsible for systems acquisition and benchmarking. Second, we provide an organized collection of techniques and tools needed to develop such benchmarks. In particular, the collection contains a detailed guide for generating benchmark workload, defining performance metrics, and interpreting experimental results

We describe our method for benchmarking Semantic Web knowledge base systems with respect to use in large OWL applications. We present the Lehigh University Benchmark (LUBM) as an example of how to design such benchmarks. The LUBM features an ontology for the university domain, synthetic OWL data scalable to an arbitrary size, fourteen extensional queries representing a variety of properties, and several performance metrics. The LUBM can be used to evaluate systems with different reasoning capabilities and storage mechanisms. We demonstrate this with an evaluation of two memory-based systems and two systems with persistent storage.

Cervical cancer is the second most common type of cancer for women. Existing screening programs for cervical cancer, such as Pap Smear, suffer from low sensitivity. Thus, many patients who are ill are not detected in the screening process. Using images of the cervix as an aid in cervical cancer screening has the potential to greatly improve sensitivity, and can be especially useful in resource-poor regions of the world. In this paper, we develop a data-driven computer algorithm for interpreting cervical images based on color and texture. We are able to obtain 74% sensitivity and 90% specificity when differentiating high-grade cervical lesions from low-grade lesions and normal tissue. On the same dataset, using Pap tests alone yields a sensitivity of 37% and specificity of 96%, and using HPV test alone gives a 57% sensitivity and 93% specificity. Furthermore, we develop a comprehensive algorithmic framework based on Multimodal Entity Coreference for combining various tests to perform disease classification and diagnosis. When integrating multiple tests, we adopt information gain and gradient-based approaches for learning the relative weights of different tests. In our evaluation, we present a novel algorithm that integrates cervical images, Pap, HPV, and patient age, which yields 83.21% sensitivity and 94.79% specificity, a statistically significant improvement over using any single source of information alone.

The ability to reason over large scale data and return responsive query results is widely seen as a critical step to achieving the Semantic Web vision. We describe an approach for partitioning OWL Lite datasets and then propose a strategy for parallel reasoning about concept instances and role instances on each partition. The partitions are designed such that each can be reasoned on independently to find answers to each query sub goal, and when the results are unioned together, a complete set of results are found for that sub goal. Our partitioning approach has a polynomial worst case time complexity in the size of the knowledge base. In our current implementation, we partition semantic web datasets and execute reasoning tasks on partitioned data in parallel on independent machines. We implement a master-slave architecture that distributes a given query to the slave processes on different machines. All slaves run in parallel, each performing sound and complete reasoning to execute each sub goal of its query on its own set of partitions. As a final step, master joins the results computed by the slaves. We study the impact of our parallel reasoning approach on query performance and show some promising results on LUBM data.

Without semantically enriched content, the Web cannot reach its full potential. The authors discuss tools and techniques for generating and processing such content, thus setting a foundation upon which to build the Semantic Web. The authors put a Semantic Web language through its paces and answer questions about how people can use it, such as: how do authors generate semantic descriptions; how do agents discover these descriptions; how can agents integrate information from different sites; and how can users query the Semantic Web.

In this paper we investigate the keyword based class retrieval problem, which we define as how to identify ontological classes that best match a keyword based query. Most previous applications use simple syntactic matching approaches on the class labels and/or comments, or expand the keyword query by using lexicons such as Word Net, but fail to retrieve relevant resources in many scenarios. Instead of relying on external sources, we investigate this problem by using the annotations of instances associated with classes in the knowledge base. We propose a general framework of this approach, which consists of two phases: the keyword query is first used to locate relevant instances, then we induce the classes given this list of weighted matched instances. If we identify sufficient text for the instances, then the first phase can be solved by a traditional information retrieval (IR) query, however the second phase might be cast in different ways: as an additive value function, as an IR problem with instance as queries, or as an instance-based ontology alignment problem. With many applicable strategies initiated from different viewpoints, we find that some of them are mathematically equivalent or very similar. In the experiments we compare our proposed framework to simple syntactic approaches and evaluate different strategies.

In this paper we present the infrastructure of the contextual tag cloud system which can execute large volumes of queries about the number of instances that use particular ontological terms. The contextual tag cloud system is a novel application that helps users explore a large scale RDF dataset: the tags are ontological terms (classes and properties), the context is a set of tags that defines a subset of instances, and the font sizes reflect the number of instances that use each tag. It visualizes the patterns of instances specified by the context a user constructs. Given a request with a specific context, the system needs to quickly find what other tags the instances in the context use, and how many instances in the context use each tag. The key question we answer in this paper is how to scale to Linked Data; in particular we use a dataset with 1.4 billion triples and over 380,000 tags. This is complicated by the fact that the calculation should, when directed by the user, consider the entailment of taxonomic and/or domain/range axioms in the ontology. We combine a scalable preprocessing approach with a specially-constructed inverted index and use three approaches to prune unnecessary counts for faster intersection computations. We compare our system with a state-of-the-art triple store, examine how pruning rules interact with inference and analyze our design choices.

Intrusion Detection Systems are tools used to detect attacks against networks. Many of these attacks are a sequence of multiple simple attacks. These complex attacks are more difficult to identify because (a) they are difficult to predict, (b) almost anything could be an attack, and (c) there are a huge number of possibilities. The problem is that the expertise of what constitutes an attack lies in the tacit knowledge of experienced network engineers. By providing an ontological representation of what constitutes a network attack human expertise to be codified and tested. The details of this representation are explained. An implementation of the representation has been developed. Lastly, the use of the representation in an Intrusion Detection System for complex attack detection has been demonstrated using use cases.

To better support information integration on Semantic Web data with varying degrees of quality, this paper proposes an approach to detect triples which reflect some sort of error. In particular, erroneous triples may occur due to factual errors in the original data source, misuse of the ontology by the original data source, or errors in the integration process. Although diagnosing such errors is a difficult problem, we propose that the degree to which a triple deviates from similar triples can be an important heuristic for identifying errors. We detect such “abnormal triples” by learning probabilistic rules from the reference data and checking to what extent these rules agree with the triples. The system consists of two components for two types of abnormal relational descriptions that a Semantic Web statement could have, whether accidentally or maliciously: a statement could relate two resources that are unlikely to have anything in common or an inappropriate predicate could be used to describe the relation between the two resources. The classification technique is adopted to learn statistical characteristics for detecting a suspect resource pair, i.e. there is no significant relation between the subject and the object in the statement. For the suspect usages of a predicate, the system learns semantic patterns for each predicate from indirect semantic connections between the subject / object pairs.

ISBN : 9780819484116; International audience; Making datasets available for peer reviewing of published document analysis methods or distributing large commonly used document corpora for benchmarking are extremely useful and sound practices and initiatives. This paper shows that they cover only a very tiny segment of the uses shared and commonly available research data may have. We develop a completely new paradigm for sharing and accessing common data sets, benchmarks and other tools that is based on a very open and free community based contribution model. The model is operational and has been implemented so that it can be tested on a broad scale. The new interactions that will arise from its use may spark innovative ways of conducting document analysis research on the one hand, but create very challenging interactions with other research domains as well.

Data quality issues arise in the Semantic Web because data is created by diverse people and/or automated tools. In particular, erroneous triples may occur due to factual errors in the original data source, the acquisition tools employed, misuse of ontologies, or errors in ontology alignment. We propose that the degree to which a triple deviates from similar triples can be an important heuristic for identifying errors. Inspired by functional dependency, which has shown promise in database data quality research, we introduce value-clustered graph functional dependency to detect abnormal data in RDF graphs. To better deal with Semantic Web data, this extends the concept of functional dependency on several aspects. First, there is the issue of scale, since we must consider the whole data schema instead of being restricted to one database relation. Second, it deals with multi-valued properties without explicit value correlations as specified as tuples in databases. Third, it uses clustering to consider classes of values. Focusing on these characteristics, we propose a number of heuristics and algorithms to efficiently discover the extended dependencies and use them to detect abnormal data. Experiments have shown that the system is efficient on multiple data sets and also detects many quality problems in real world data.

One challenge for Linked Data is scalably establishing highquality owl:sameAs links between instances (e.g., people, geographical locations, publications, etc.) in different data sources. Traditional approaches to this entity coreference problem do not scale because they exhaustively compare every pair of instances. In this paper, we propose a candidate selection algorithm for pruning the search space for entity coreference. We select candidate instance pairs by computing a character-level similarity on discriminating literal values that are chosen using domain-independent unsupervised learning. We index the instances on the chosen predicates' literal values to efficiently look up similar instances. We evaluate our approach on two RDF and three structured datasets. We show that the traditional metrics don't always accurately reflect the relative benefits of candidate selection, and propose additional metrics. We show that our algorithm frequently outperforms alternatives and is able to process 1 million instances in under one hour on a single Sun Workstation. Furthermore, on the RDF datasets, we show that the entire entity coreference process scales well by applying our technique. Surprisingly, this high recall, low precision filtering mechanism frequently leads to higher F-scores in the overall system.

In this paper, we present a novel entity coreference algorithm for Semantic Web instances. The key issues include how to locate context information and how to utilize the context appropriately. To collect context information, we select a neighborhood (consisting of triples) of each instance from the RDF graph. To determine the similarity between two instances, our algorithm computes the similarity between comparable property values in the neighborhood graphs. The similarity of distinct URIs and blank nodes is computed by comparing their outgoing links. To provide the best possible domain-independent matches, we examine an appropriate way to compute the discriminability of triples. To reduce the impact of distant nodes, we explore a distance-based discounting approach. We evaluated our algorithm using different instance categories in two datasets. Our experiments show that the best results are achieved by including both our triple discrimination and discounting approaches.

In recent years, there has been an explosion of publicly available RDF and OWL data sources. In order to effectively and quickly answer queries in such an environment, we present an approach to identifying the potentially relevant Semantic Web data sources using query rewritings and a term index. We demonstrate that such an approach must carefully handle query goals that lack constants; otherwise the algorithm may identify many sources that do not contribute to eventual answers. This is because the term index only indicates if URIs are present in a document, and specific answers to a subgoal cannot be calculated until the source is physically accessed - an expensive operation given disk/network latency. We present an algorithm that, given a set of query rewritings that accounts for ontology heterogeneity, incrementally selects and processes sources in order to maintain selectivity. Once sources are selected, we use an OWL reasoner to answer queries over these sources and their corresponding ontologies. We present the results of experiments using both a synthetic data set and a subset of the real-world Billion Triple Challenge data.

Researchers have found that Word Sense Disambiguation (WSD) is useful for tasks such as ontology alignment. Many other Semantic Web applications could also be enhanced with WSD results of Semantic Web documents. A system that can provide reusable intermediate WSD results is desirable. Compared to the top sense or a rank of senses, an output of meaningful scores of each possible sense informs subsequent processes of the certainty in results, and facilitates the application of other knowledge in choosing the correct sense. We propose that probabilistic models, which have proved successful in many other fields, can also be applied to WSD. Based on such observations, we focus on the problem of calculating probability distributions of senses for terms. In this paper we propose our novel WSD approach with our probability model, derive the problem formula into small computable pieces, and propose ways to estimate the values of these pieces.

In recent years, there has been an explosion of publicly available RDF and OWL web pages. Typically, these pages are small, heterogeneous and prone to change frequently. In order to effectively integrate them, we propose to adapt a query reformulation algorithm and combine it with an information retrieval inspired index in order to select all sources relevant to a query. We treat each RDF document as a bag of URIs and literals and build an inverted index. Our system first reformulates the user’s query into a set of sub goals and then translates these into Boolean queries against the index in order to determine which sources are relevant. Finally, the selected data sources and the relevant ontology mappings are used in conjunction with a description logic reasoner to provide an efficient query answering solution for the Semantic Web. We have evaluated our system using ontology mappings and ten million real world data sources.

In order to effectively and quickly answer queries in environments with distributed RDF/OWL, we present a query optimization algorithm to identify the potentially relevant Semantic Web data sources using structural query features and a term index. This algorithm is based on the observation that the join selectivity of a pair of query triple patterns is often higher than the overall selectivity of these two patterns treated independently. Given a rule goal tree that expresses the reformulation of a conjunctive query, our algorithm uses a bottom-up approach to estimate the selectivity of each node. It then prioritizes loading of selective nodes and uses the information from these sources to further constrain other nodes. Finally, we use an OWL reasoner to answer queries over the selected sources and their corresponding ontologies. We have evaluated our system using both a synthetic data set and a subset of the real-world Billion Triple Challenge data.

Internet business-to-business transactions present great challenges in merging information from different sources. In this paper we describe a project to integrate four representative commercial classification systems with the Federal Cataloging System (FCS). The FCS is used by the US Defense Logistics Agency to name, describe and classify all items under inventory control by the DoD. Our approach uses the ECCMA Open Technical Dictionary (eOTD) as a common vocabulary to accommodate all different classifications. We create a semantic bridging ontology between each classification and the eOTD to describe their logical relationships in OWL DL. The essential idea is that since each classification has formal definitions in a common vocabulary, we can use subsumption to automatically integrate them, thus mitigating the need for pairwise mappings. Furthermore our system provides an interactive interface to let users choose and browse the results and more importantly it can translate catalogs that commit to these classifications using compiled mapping results.

We present an efficient search approach for selecting all potentially relevant data sources for a conjunctive Semantic Web query. We use map ontologies to align heterogeneous domain ontologies. This allows us to select data sources that may be relevant to the query but generally do not describe their data directly in terms of the ontology of the query. The "goal node search" algorithm is a significant improvement on our original source selection algorithm. The new algorithm allows a more expressive knowledge representation language to describe domain ontologies and it is about three times more efficient than the original source selection algorithm when performing similar tasks.

A true semantic Web repository must scale both in terms of number of ontologies and quantity of data. It should also support reasoning using different points of view about the meanings and relationships of concepts and roles. Our DLDB2 system has these features. Our system is sound and complete on a sizable subset of description horn logic when answering extensional conjunctive queries, but more importantly also computes many entailments from OWL DL. By delegating TBox reasoning to a DL reasoner, we focus on the design of the table schema, database views, and algorithms that achieve essential ABox reasoning over an RDBMS. We evaluate the system using synthetic benchmarks as well as real-world data and queries.

In this work we adapt an efficient information integration algorithm to identify the minimal set of potentially relevant Semantic Web data sources for a given query. The vast majority of these sources are files written in RDF or OWL format, and must be processed in their entirety. Our adaptation includes enhancing the algorithm with taxonomic reasoning, defining and using a mapping language for the purpose of aligning heterogeneous Semantic Web ontologies, and introducing a concept of source relevance to reduce the number of sources that we need to consider for a given query. After the source selection process, we load the selected sources into a Semantic Web reasoner to get a sound and complete answer to the query. We have conducted an experiment using synthetic ontologies and data sources which demonstrates that our system performs well over a wide range of queries. A typical response time for a substantial work load of 50 domain ontologies, 80 map ontologies and 500 data sources is less than 2 seconds. Furthermore,our system returned correct answers to 200 randomly generated queries in several workload configurations. We have also compared our adaptation with a basic implementation of the original information integration algorithm that does not do any taxonomic reasoning. In the most complex configuration with 50 domain ontologies, 100 map ontologies and 1000 data sources our system returns complete answers to all the queries whereas the basic implementation returns complete answers to only 28% of the queries.

We discuss our DLDB knowledge base system and evaluate its capability in processing a very large set of real-world Semantic Web data. Using DLDB, we have constructed the Hawkeye knowledge base, in which we have loaded more than 166 million facts from a diverse set of real-world data sources. We use this knowledge base to demonstrate realistic integration queries in e-government and academic scenarios. In order to support Hawkeye, we extended DLDB with additional reasoning capabilities. At present, the Semantic Web consists of numerous independent ontologies.We demonstrate that OWL can be used to integrate these ontologies and thereby integrate the data sources that commit to them. In terms of performance, we show that the load time of our system is linear on the number of triples loaded. Furthermore, we show that many complex queries have response times under one minute, and that simple queries can be answered in seconds.

In this paper, we propose document-centric query answering, a novel form of query answering for the Semantic Web. We discuss how we have built a knowledge base system to support the new queries. In particular, we describe the key techniques used in the system in order to address scalability issues. In addition, we show encouraging experimental results.

A distributed, end-to-end information integration system that is based on the Semantic Web architecture is of considerable interest to both commercial and government organizations. However, there are a number of challenges that have to be resolved to build such a system given the currently available Semantic Web technologies. We describe here the ISENS prototype system we designed, implemented, and tested (on a small scale) to address this problem. We discuss certain system limitations (some coming from underlying technologies used) and future ISENS development to resolve them and to enable an extended set of capabilities.

We present a method for rapid development of benchmarks for Semantic Web knowledge base systems. At the core, we have a synthetic data generation approach for OWL that is scalable and models the real world data. The data-generation algorithm learns from real domain documents and generates benchmark data based on the extracted properties relevant for benchmarking. We believe that this is important because relative performance of systems will vary depending on the structure of the ontology and data used. However, due to the novelty of the Semantic Web, we rarely have sufficient data for benchmarking. Our approach helps overcome the problem of having insufficient real world data for benchmarking and allows us to develop benchmarks for a variety of domains and applications in a very time efficient manner. Based on our method, we have created a new Lehigh BibTeX Benchmark and conducted an experiment on four Semantic Web knowledge base systems. We have verified our hypothesis about the need for representative data by comparing the experimental result to that of our previous Lehigh University Benchmark. The difference in both experiments has demonstrated the influence of ontology and data on the capability and performance of the systems and thus the need of using a representative benchmark for the intended application of the systems.

In this paper, we investigate the (in)dependence among OWL documents with respect to the logical consequence when they are combined, in particular the inference of concept and role assertions about individuals. On the one hand, we present a systematic approach to identifying those documents that affect the inference of a given fact. On the other hand, we consider ways for fast detection of independence. First, we demonstrate several special cases in which two documents are independent of each other. Secondly, we introduce an algorithm for checking the independence in the general case. In addition, we describe two applications in which the above results have allowed us to develop novel approaches to overcome some difficulties in reasoning with large scale OWL data. Both applications demonstrate the usefulness of this work for improving the scalability of a practical Semantic Web system that relies on the reasoning about individuals.

Emerging Semantic Web technology such as the DARPA Agent Markup Language (DAML) will support advanced semantic interoperability in the next generation of aerospace architectures. The basic idea of DAML is to mark up artifacts (e.g. documents sensors databases legacy software) so that software agents can interpret and reason with the information DAML will support the representation of ontologies (which include taxonomies of terms and semantic relations) via extensions to XML, XML alone is not sufficient for agents because it provides only syntactic interoperability that depends on implicit semantic agreements DAML is the official starting point for the Web Ontology Language an emerging standard from the World Wide Web Consortium. This paper will cover promising aerospace applications and significant challenges for Semantic Web technologies Potential applications include higher-level information fusion collaboration in both operational and engineering environments and rapid systems integration. The challenges that will be discussed include the complexity of ontology development automation of markup semantic mismatch between current object-oriented models and Semantic Web ontologies scalability issues related to reasoning with large knowledge bases and technology transition issues. The paper will explain ongoing research that is focused on addressing these challenges.

In this paper, we present an evaluation of four knowledge base systems (KBS) with respect to use in large OWL applications. To our knowledge, no experiment has been done with the scale of data used here. The smallest dataset used consists of 15 OWL files totaling 8MB, while the largest dataset consists of 999 files totaling 583MB. We evaluated two memory-based systems (OWL Jess KB and memory-based Sesame) and two systems with persistent storage (database-based Sesame and DLDB-OWL). We describe how we have performed the evaluation and what factors we have considered in it. We show the results of the experiment and discuss the performance of each system. In particular, we have concluded that existing systems need to place a greater emphasis on scalability.

We present a benchmark that facilitates the evaluation of DAML+OIL repositories in a standard and systematic way. This benchmark is intended to evaluate the performance of DAML+OIL repositories with respect to extensional queries over a large data set that commits to a single realistic ontology. It consists of the ontology, customizable synthetic data, a set of test queries, and several performance metrics. Main features of the benchmark include a plausible ontology for the university domain, a repeatable data set that can be scaled to an arbitrary size, and an approach for measuring the degree to which a repository returns complete query answers. We also show a benchmark experiment for the evaluation of DLDB, a DAML+OIL repository that extends a relational database management system with description logic inference capabilities.

XML will have a profound impact on the way data is exchanged on the Internet. An important feature of thislanguage is the separation of content from presentation, which makes it easier to select and/or reformat the data.However, due to the likelihood of numerous industry and domain specific DTDs, those who wish to integrateinformation will still be faced with the problem of semantic interoperability. In this paper we discuss why thisproblem is not solved by XML, and then discuss why the Resource Description Framework is only a partial solution.We then present the SHOE language, which we feel has many of the features necessary to enable a semantic web,and describe an existing set of tools that make it easy to use the language.Jeff Heflin, University of MarylandJames Hendler, University of Maryland

inants2015a; International audience; Algebras of relations were shown useful in managing ontology alignments. They make it possible to aggregate alignments disjunctively or conjunctively and to propagate alignments within a network of ontologies. The previously considered algebra of relations contains taxonomical relations between classes. However, compositional inference using this algebra is sound only if we assume that classes which occur in alignments have nonempty extensions. Moreover, this algebra covers relations only between classes. Here we introduce a new algebra of relations, which, first, solves the limitation of the previous one, and second, incorporates all qualitative taxonomical relations that occur between individuals and concepts, including the relations "is a" and "is not". We prove that this algebra is coherent with respect to the simple semantics of alignments.

Research track: Ontology Mapping; International audience; Ontology mappings are often assigned a weight or confidence factor by matchers. Nonetheless, few semantic accounts have been given so far for such weights. This paper presents a formal semantics for weighted mappings between different ontologies. It is based on a classificational interpretation of mappings: if O1 and O2 are two ontologies used to classify a common set X, then mappings between O1 and O2 are interpreted to encode how elements of X classified in the concepts of O1 are re-classified in the concepts of O2, and weights are inter- preted to measure how precise and complete re-classifications are. This semantics is justifiable by extensional practice of ontology matching. It is a conservative ex- tension of a semantics of crisp mappings. The paper also includes properties that relate mapping entailment with description logic constructors.