User and task analysis for interface design

We present a system for the detection of the stance of headlines with regard to their corresponding article bodies The approach can be applied in fake news, especially clickbait detection scenarios The component is part of a larger platform for the curation of digital content; we consider veracity and relevancy an increasingly important part of curating online information We want to contribute to the debate on how to deal with fake news and related online phenomena with technological means, by providing means to separate related from unrelated headlines and further classifying the related headlines On a publicly available data set annotated for the stance of headlines with regard to their corresponding article bodies, we achieve a (weighted) accuracy score of 8959

/pdf/from-clickbait-to-fake-news-detection-an-approach-based-on-58q05qsy1s.pdf

From Clickbait to Fake News Detection: An Approach based on Detecting the Stance of Headlines to Articles

This paper describes an approach at Named Entity Recognition (NER) in German language documents from the legal domain. For this purpose, a dataset consisting of German court decisions was developed. The source texts were manually annotated with 19 semantic classes: person, judge, lawyer, country, city, street, landscape, organization, company, institution, court, brand, law, ordinance, European legal norm, regulation, contract, court decision, and legal literature. The dataset consists of approx. 67,000 sentences and contains 54,000 annotated entities. The 19 fine-grained classes were automatically generalised to seven more coarse-grained classes (person, location, organization, legal norm, case-by-case regulation, court decision, and legal literature). Thus, the dataset includes two annotation variants, i.e., coarse- and fine-grained. For the task of NER, Conditional Random Fields (CRFs) and bidirectional Long-Short Term Memory Networks (BiLSTMs) were applied to the dataset as state of the art models. Three different models were developed for each of these two model families and tested with the coarse- and fine-grained annotations. The BiLSTM models achieve the best performance with an 95.46 F\(_1\) score for the fine-grained classes and 95.95 for the coarse-grained ones. The CRF models reach a maximum of 93.23 for the fine-grained classes and 93.22 for the coarse-grained ones. The work presented in this paper was carried out under the umbrella of the European project LYNX that develops a semantic platform that enables the development of various document processing and analysis applications for the legal domain.

/pdf/fine-grained-named-entity-recognition-in-legal-documents-6ezznufbll.pdf

Fine-Grained Named Entity Recognition in Legal Documents

Online media and digital communication technologies have an unprecedented, even increasing level of social, political and also economic relevance. This article proposes an infrastructure to address phenomena of modern online media production, circulation and manipulation by establishing a distributed architecture for automatic processing and human feedback.

/pdf/an-infrastructure-for-empowering-internet-users-to-handle-2qskr6u3h4.pdf

An Infrastructure for Empowering Internet Users to Handle Fake News and Other Online Media Phenomena

The Entity Linking (EL) task identifies entity mentions in a text corpus and associates them with corresponding entities in a given knowledge base. While traditional EL approaches have largely focused on English texts, current trends are towards language-agnostic or otherwise multilingual approaches that can perform EL over texts in many languages. One of the obstacles to ongoing research on multilingual EL is a scarcity of annotated datasets with the same text in different languages. In this work we thus propose VoxEL: a manually-annotated gold standard for multilingual EL featuring the same text expressed in five European languages. We first motivate and describe the VoxEL dataset, using it to compare the behaviour of state of the art EL (multilingual) systems for five different languages, contrasting these results with those obtained using machine translation to English. Overall, our results identify how five state-of-the-art multilingual EL systems compare for various languages, how the results of different languages compare, and further suggest that machine translation of input text to English is now a competitive alternative to dedicated multilingual EL configurations.

VoxEL: A benchmark dataset for multilingual entity linking

In an attempt to put a Semantic Web-layer that provides linguistic analysis and discourse information on top of digital content, we develop a platform for digital curation technologies. The platform offers language-, knowledge- and data-aware services as a flexible set of workflows and pipelines for the efficient processing of various types of digital content. The platform is intended to enable human experts (knowledge workers) to get a grasp and understand the contents of large document collections in an efficient way so that they can curate, process and further analyse the collection according to their sector-specific needs.

Towards a Platform for Curation Technologies: Enriching Text Collections with a Semantic-Web Layer

We present an approach at identifying a specific class of events, movement action events (MAEs), in a data set that consists of ca. 2,800 personal letters exchanged by the German architect Erich Mendelsohn and his wife, Luise. A backend system uses these and other semantic analysis results as input for an authoring environment that digital curators can use to produce new pieces of digital content. In our example case, the human expert will receive recommendations from the system with the goal of putting together a travelogue, i.e., a description of the trips and journeys undertaken by the couple. We describe the components and architecture and also apply the system to news data.

Event Detection and Semantic Storytelling: Generating a Travelogue from a large Collection of Personal Letters

Digital content and online media have reached an unprecedented level of relevance and importance. In the context of a research and technology transfer project on Digital Curation Technologies for online content we develop a platform that provides curation services that can be integrated into concrete curation or content management systems. In this project, the German Research Center for Artificial Intelligence (DFKI) collaborates with four Berlin-based SMEs that work with and on digital content in four different sectors. The curation services comprise several semantic text and document analytics processes as well as knowledge technologies that can be applied to document collections. The key objective of this set of curation services is to support knowledge workers and digital curators in their daily work, i.e., to automate or to semi-automate processes that the human experts are normally required to do intellectually and without tool support. The goal is to help this group of information and knowledge workers to become more efficient and more effective as well as to enable them to produce high-quality content in their respective sectors. In this article we concentrate on the current state of a user interface that is currently under development at ART+COM, one of the SME partners in the project. A second, more generic, i.e., not domain-specific user interface is under development at DFKI. In this article we describe the technology platform and the two different interfaces. We also take a look at the different requirements for ART+COM’s domain-specific and DFKI’s generic user interface.

Designing User Interfaces for Curation Technologies

This paper introduces the current state of the FREME framework. The paper puts FREME into the context of linguistic linked data and related approaches of multilingual and semantic processing. In addition, we focus on two specific aspects of FREME: the FREME NER e-Service, and chaining of FREME e-Services. We believe that the flexible and distributed combination of e-Services bears a potential for their mutual improvement. The FREME framework is an open source software available for free download (https://github.com/freme-project/).

Chainable and Extendable Knowledge Integration Web Services

The FREME framework bridges Language Technologies (LT) and Linked Data (LD). It establishes workflows between LT and LD in a well defined, coherent way. FREME addresses common challenges that both researchers and industry face when integrating LT and LD: interoperability, ”silo” solutions and the lack of adequate tooling. Usability, reusability and interoperability are often attributes of frameworks and toolkits for LT and LD. In this paper, we take a novel approach: We define user types and user levels and describe how they influence design decisions in a LT and LD processing framework. In this way, we combine research outcomes from various communities: language technology, linked data and software interface engineering. This paper explains the different user types and how FREME addresses the specific needs of each user type. Core attributes of FREME are usability, reusability and interoperability.

Jan Nehring

Papers

Towards a Platform for Curation Technologies: Enriching Text Collections with a Semantic-Web Layer

Event Detection and Semantic Storytelling: Generating a Travelogue from a large Collection of Personal Letters

Designing User Interfaces for Curation Technologies

Chainable and Extendable Knowledge Integration Web Services

A Framework for the Needs of Different Types of Users in Multilingual Semantic Enrichment.