1,720,989 research outputs found
Griesbaum, J.; Mandl, T.; Womser-Hacker, C. (Hrsg.) Information und Wissen: global, sozial und frei?: Proceedings des 12. Internationalen Symposiums für Informationswissenschaft (ISI 2011) Hildesheim, 9.–11. März 2011. Boizenburg: Hülsbusch, 2011. 532 S. ISBN 978-3-940317-91-9.
Get PDFReview of the proceedings volume "Information
und Wissen: global, sozial und frei? Proceedings des 12. Internationalen
Symposiums für Informationswissenschaft (ISI 2011) Hildesheim, 9.–11. März 2011", edited by Griesbaum, J., Mandl, T. and Womser-Hacker, C
An evaluation resource for geographic information retrieval
Get PDFIn this paper we present an evaluation resource for geographic information retrieval developed within the Cross Language Evaluation
Forum (CLEF). The GeoCLEF track is dedicated to the evaluation of geographic information retrieval systems. The resource
encompasses more than 600,000 documents, 75 topics so far, and more than 100,000 relevance judgments for these topics. Geographic
information retrieval requires an evaluation resource which represents realistic information needs and which is geographically
challenging. Some experimental results and analysis are reported
Towards ease of building legos in assessing ehealth language technologies:A RESTful laboratory for data and software
No full textMore and more scientific literature, care guidelines, health records, social media, and other textual eHealth information are electronically available. Language technologies provide a way to analyse these documents for the benefit of both individuals and populations. In order to catalyse the development of eHealth language technologies, we propose a virtual laboratory with a standardised platform for easy building and assessment of the systems from the "lego" bricks of shared data, resources, and software. Our aim is to address specific needs in eHealth: governance and sharing of private data; provenance and sharing of resources and software; systematic benchmarking and quality control of systems and their components; and collaboration of eHealth language technology developers and users across healthcare services, academia, industry, and government. The Epicure virtual laboratory is intended to be used for software and re-source evaluation and development as well as for data analysis if data subjects' privacy is ensured. Epicure is a meta-framework in the sense of abstracting over existing frameworks. Its five roles for clients are data or resource provider, ap-plication assembler, application user, software developer, and system administrator. We have implemented Epicure based on publicly available software. Its control layer is a Glassfish JavaEE server, providing a RESTful (REpresentational State Transfer) application programming interface; web interface for accessing and installing third-party platforms; and easy operation via standard web commands. After proper user authentication and authorisation of incoming requests, it builds applications, analyses data and assesses outcomes by orchestrating storage and execution layers. The storage layer of Epicure uses a CouchDB-based repository for centralised storage of data, resources, and software. It enables controlling document access on the level of documents; tracking all changes; recording these revisions; storing all analysis outcomes; and associating the outcomes with the data, resources and software used in their generation. The execution layer of Epicure provides a runtime environment for executing data analysis tasks and installing third party platforms. It invokes tools as simple commands. A tool must be specify its input format, output formats, parameters, and their possible values as a file and be executable on a command line. Tools do not need to be installed within Epicure itself but instead be accessed via a network interface and wrapper, which provides access from Epicure to this re-mote service.</p
Overview of the INEX 2012 Social Book Search Track
No full textThe goal of the INEX 2012 Social Book Search Track is to evaluate approaches for supporting users in reading, searching, and navigating book metadata and full texts of digitised books as well as associated user-generated content. The investigation is focused around two tasks: 1) the Social Book Search task investigates the complex nature relevance in book search and the role of user information and traditional and user-generated book metadata for retrieval, 2) the Prove It task evaluates focused retrieval approaches for searching pages in books that support or refute a given factual claim. There are two additional tasks that did not run this year. The Structure Extraction task tests automatic techniques for deriving structure from OCR and layout information, and the Active Reading Task aims to explore suitable user interfaces for eBooks enabling reading, annotation, review, and summary across multiple books. We report on the setup and the results of the two search tasks
Using collaborative filtering in social book search
No full textIn this paper we describe our participation in INEX 2012 in the Social Book Search Track and the Linked Data Track. For the Social Book Search Track we compare the impact of query- and user-independent popularity measures and recommendations based on user profiles. Book suggestions are more than just topical relevance judgements and may include personal factors such as interestingness, fun and familiarity and book-related aspects such as quality and popularity. Our aim is to understand to what extent book suggestions are related to user-dependent and -independent aspects of relevance. Our findings are that evidence that is both query- and user-independent is not effective for improving a standard retrieval model using blind feedback. User-dependent evidence, on the contrary, is highly effective, leading to significant improvements. For the Linked Data Track we compare different methods of weighted result aggregation using the DBpedia ontology relations as facets and values. Facets and values are aggregated using either document counts or retrieval scores. The reason to use retrieval scores for facet ranking is that we want the top retrieved results to be summarised by the top ranked facets and values. In addition, we look at the impact of taking overlap in aggregation into account. Facet values that give access to many of the same documents have high overlap. Selecting facet values that have low overlap may avoid frustrating the user
Overview of RepLab 2012: Evaluating Online Reputation Management Systems
No full textThis paper summarizes the goals, organization and results of the first RepLab competitive evaluation campaign for Online Reputation Management Systems (RepLab 2012). RepLab focused on the reputation of companies, and asked participant systems to annotate different types of information on tweets containing the names of several companies. Two tasks were proposed: a profiling task, where tweets had to be annotated for relevance and polarity for reputation, and a monitoring task, where tweets had to be clustered thematically and clusters had to be ordered by priority (for reputation management purposes). The gold standard consisted of annotations made by reputation management experts, a feature which turns the RepLab 2012 test collection in a useful source not only to evaluate systems, but also to reach a better understanding of the notions of polarity and priority in the context of reputation management
Overview of the INEX 2012 Linked Data Track
No full textThis paper provides an overview of the Linked Data Track that was newly introduced to the set of INEX tracks in 2012
Using collaborative filtering in social book search
Get PDFIn this paper we describe our participation in INEX 2012 in the Social Book Search Track and the Linked Data Track. For the Social Book Search Track we compare the impact of query- and user-independent popularity measures and recommendations based on user profiles. Book suggestions are more than just topical relevance judgements and may include personal factors such as interestingness, fun and familiarity and book-related aspects such as quality and popularity. Our aim is to understand to what extent book suggestions are related to user-dependent and -independent aspects of relevance. Our findings are that evidence that is both query- and user-independent is not effective for improving a standard retrieval model using blind feedback. User-dependent evidence, on the contrary, is highly effective, leading to significant improvements. For the Linked Data Track we compare different methods of weighted result aggregation using the DBpedia ontology relations as facets and values. Facets and values are aggregated using either document counts or retrieval scores. The reason to use retrieval scores for facet ranking is that we want the top retrieved results to be summarised by the top ranked facets and values. In addition, we look at the impact of taking overlap in aggregation into account. Facet values that give access to many of the same documents have high overlap. Selecting facet values that have low overlap may avoid frustrating the user
From Sentiment to Reputation: ILPS at RepLab 2012
No full textWe report on our participation in the profiling task of the first edition of the CLEF RepLab evaluation initiative. We assume that a statement - such as a tweet - that carries negative sentiment can have a positive impact on the reputation of the entity it talks about (and vice versa). Our model directly captures this impact by observing the reactions - such as replies - the statement solicits. We present the assumptions behind our model and the model itself. We find that given the current setting, results on the test set are strongly entity-dependent and that the test data is very different from the trial data. We conclude with a proposal on how to create a task that avoids such dataset dependent problems
- …
