Steffen Metzger

QBEES: query by entity examples

Structured knowledge bases are an increasingly important way for storing and retrieving information. Within such knowledge bases, an important search task is finding similar entities based on one or more example entities. We present QBEES, a novel framework for defining entity similarity based only on structural features, so-called aspects, of the entities, that includes query-dependent and query-independent entity ranking components. We present evaluation results with a number of existing entity list completion benchmarks, comparing to several state-of-the-art baselines.

S3K: seeking statement-supporting top-K witnesses

Traditional information retrieval techniques based on keyword search help to identify a ranked set of relevant documents, which often contains many documents in the top ranks that do not meet the users intention. By considering the semantics of the keywords and their relationships, both precision and recall can be improved. Using an ontology and mapping keywords to entities/concepts and identifying the relationship between them that the user is interested in, allows for retrieving documents that actually meet the users intention. In this paper, we present a framework that enables semantic-aware document retrieval. User queries are mapped to semantic statements based on entities and their relationships. The framework searches for documents expressing these statements in different variations, e.g., synonymous names for entities or different textual expressions for relations between them. The size of potential result sets makes ranking documents according to their relevance to the user an essential component of such a system. The ranking model proposed in this paper is based on statistical language-models and considers aspects such as the authority of a document and the confidence in the textual pattern representing the queried information.

Aspect-Based Similar Entity Search in Semantic Knowledge Graphs with Diversity-Awareness and Relaxation

Structured knowledge bases are an increasingly important way for storing and retrieving information. Within such knowledge bases, an important search task is finding similar entities based on one or more example entities. We present QBEES, a novel framework for defining entity similarity based only on structural features, so-called aspects, of the entities, that naturally model potential interest profiles of a user submitting an ambiguous query. The aspect model provides natural diversity-awareness and includes query-dependent and query-independent entity ranking components. We present evaluation results with a number of existing entity list completion benchmarks, comparing to several state-of-the-art baselines.

ROXXI: Reviving witness dOcuments to eXplore eXtracted Information

In recent years, there has been considerable research on information extraction and constructing RDF knowledge bases. In general, the goal is to extract all relevant information from a corpus of documents, store it into an ontology, and answer future queries based only on the created knowledge base. Thus, the original documents become dispensable. On the one hand, an ontology is a convenient and non-redundant structured source of information, based on which specific queries can be answered efficiently. On the other hand, many users doubt the correctness of facts and ontology subgraphs presented to them as query results without proof. Instead, users often wish to verify the obtained facts or subgraphs by reading about them in context, i.e., in a document relating the facts and providing background information. In this demo, we present ROXXI, a system operating on top of an existing knowledge base and reviving the abandoned witness documents. In doing so, it goes the opposite way of information extraction approaches -- starting with ontological facts and tracing their way back to the documents they were extracted from. ROXXI offers interfaces for expert users (SPARQL) as well as for non-experts (ontology browser) and provides a ranked list of documents each associated with a content snippet highlighting the queried facts in context. At the demonstration site, we will show the advantages of this novel approach towards document retrieval and illustrate the benefits of reviving the documents that information extraction approaches neglect.In recent years, there has been considerable research on information extraction and constructing RDF knowledge bases. In general, the goal is to extract all relevant information from a corpus of documents, store it into an ontology, and answer future queries based only on the created knowledge base. Thus, the original documents become dispensable. On the one hand, an ontology is a convenient and non-redundant structured source of information, based on which specific queries can be answered efficiently. On the other hand, many users doubt the correctness of facts and ontology subgraphs presented to them as query results without proof. Instead, users often wish to verify the obtained facts or subgraphs by reading about them in context, i.e., in a document relating the facts and providing background information. In this demo, we present ROXXI, a system operating on top of an existing knowledge base and reviving the abandoned witness documents. In doing so, it goes the opposite way of information extraction approaches -- starting with ontological facts and tracing their way back to the documents they were extracted from. ROXXI offers interfaces for expert users (SPARQL) as well as for non-experts (ontology browser) and provides a ranked list of documents each associated with a content snippet highlighting the queried facts in context. At the demonstration site, we will show the advantages of this novel approach towards document retrieval and illustrate the benefits of reviving the documents that information extraction approaches neglect.

QBEES: Query-by-Example Entity Search in Semantic Knowledge Graphs Based on Maximal Aspects, Diversity-awareness and Relaxation

Structured knowledge bases are an increasingly important way for storing and retrieving information. Within such knowledge bases, an important search task is finding similar entities based on one or more example entities. We present QBEES, a novel framework for defining entity similarity based on structural features, so-called aspects and maximal aspects of the entities, that naturally model potential interest profiles of a user submitting an ambiguous query. Our approach based on maximal aspects provides natural diversity awareness and includes query-dependent and query-independent entity ranking components. We present evaluation results with a number of existing entity list completion benchmarks, comparing to several state-of-the-art baselines.

Heterogeneous resource federation with a centralized security model for information extraction

With the continuous growth of data generated in various scientific and commercial endeavors and the rising need for interdisciplinary studies and applications in e-Science easy exchange of information and computation resources capable of processing large amounts of data to allow ad-hoc co-operation becomes ever more important. Unfortunately different communities often use incompatible resource management systems. In this work we try to alleviate the difficulties occurring on bridging the gap between different research eco-systems by federating resources and thus unifying resource access.To this end, our solution presented in this paper outlines a secure, simple, yet highly interoperable and flexible architecture using RESTful Web services and WebDAV. While, first and foremost in the Grid computing domain, there are already standards and solutions in place addressing related problems, our solution differs from those approaches by allowing to federate data storage systems that are not aware of being federated. Access to these is enabled by our federation layer using storage system specific connectors. Hence, our federation approach is intended as an abstraction layer on top of existing storage or middleware solutions, allowing for a more uniform access mechanism.Additionally, our solution also allows for submission and management of computational jobs on said data, thereby federating not only data but also computational resources. Once resource access is unified, information from different data formats can be semantically unified by information extraction methods. It is our belief that the work in this paper can complement existing Grid computing efforts by facilitating access to data storage system not inherently available via commonly used Grid computing standards.

Colledge: a vision of collaborative knowledge networks

More and more semantic information has become available as RDF data recently, with the linked open data cloud as a prominent example. However, participating in the Semantic Web is cumbersome. Typically several steps are involved in using semantic knowledge. Information is first acquired, e.g. by information extraction, crowd sourcing or human experts. Then ontologies are published and distributed. Users may apply reasoning and otherwise modify their local ontology instances. However, currently these steps are treated separately and although each involves human effort, nearly no synergy effect is used and it is also mostly a one way process, e.g. user feedback hardly flows back into the main ontology version. Similarly, user cooperation is low. While there are approaches alleviating some of these limitations, e.g. extracting information at query time, personalizing queries, and integration of user feedback, this work combines all the pieces envisioning a social knowledge network that enables collaborative knowledge generation and exchange. Each aforementioned step is seen as a particular implementation of a network node responding to knowledge queries in its own way, e.g. by extracting it, applying reasoning or asking users, and learning from knowledge exchanged with neighbours. Original knowledge as well as user feedback is distributed over the network based on similar trust and provenance mechanisms. The extended query language we call for also allows for personalization.

Interactive Information Extraction based on Distributed Data Management for German Grid Projects

c Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike Licence. The current infrastructure proviced and maintained by the German Grid Initiative (D-Grid) primarily covers resource management and exchange at the data level supporting mainly technical resources such as computational capacity, data transport networks, storage resources, and management software. The WisNetGrid project (www.wisnetgrid.org) aims to broaden the focus of resource sharing towards the actual content, such as research and production data, to enable in-terdisciplinary usage. To achieve this goal, resource sharing is supported on different abstraction layers. First, we create an information layer by providing a universal interface to access data on the grid independent of the underlying grid storage system. Second, at the knowledge layer, we offer interactive knowledge extraction and management tools that can also take advantage of a communitys grid resources. These tools enable the user to formulate the domain specific knowledge in different ways to ease the interaction with the knowledge extraction process and to provide input for automatic extraction workflow. Within this project, we work together with use groups from the humanities and from landscaping as disparate use cases to evaluate which advantages can be gained by using semi-automatic extraction tools to gather and manage knowledge content.

LUKe and MIKe: learning from user knowledge and managing interactive knowledge extraction

Semantic recognition and annotation of unqiue enities and their relations is a key in understanding the essence contained in large text corpora. It typically requires a combination of efficient automatic methods and manual verification. Usually, both parts are seen as consecutive steps. In this demo we present MIKE, a user interface enabling the integration of user feedback into an iterative extraction process. We show how an extraction system can directly learn from such integrated user supervision. In general, this setup allows for stepwise training of the extraction system to a particular domain, while using user feedback early in the iterative extraction process improves extraction quality and reduces the overall human effort needed.

Bericht zur Herbstschule Information Retrieval 2010

Seit 1998 bietet die GI-Fachgruppe eine Herbstschule fur Information Retrieval an. Sie richtet sich an Anwender, Entwickler sowie Studierende aus Wissenschaft und Praxis, die im Bereich Informationssysteme arbeiten und vermittelt Erkenntnisse aus einem breiten Themenspektrum des Information Retrieval (IR). Im September 2010 war es wieder soweit. Dieses Mal wurde die Herbstschule unter Leitung von Ralf Schenkel organisiert. Austragungsort war das renommierte Informatikzentrum Schloss Dagstuhl bei Wadern im Saarland, das fur seine zuruckgezogene, entspannte Atmosphare bekannt ist. Etwa 24 Teilnehmer, zumeist Doktoranden, jedoch ebenso einige Studenten und Promovierte, nahmen die Einladung an und beschaftigten sich eine Woche lang mit verschiedenen Themen aus dem Bereich Information Retrieval (Abb. 1). Dazu prasentierten insgesamt acht Vortragende in jeweils zweieinhalbstundigen Vortragen verschiedene Themenbereiche. Zudem erhielten die Teilnehmer die Gelegenheit, ihre aktuelle Arbeit vorzustellen, und konnten auf diese Weise wertvolles Feedback von anderen Teilnehmern erhalten. Daruber hinaus lies das Programm, insbesondere in den Abendstunden, genugend Freiraum fur vertiefende Diskussionen und das Knupfen von Kontakten.