Browsing by Subject "linked data"

Sort by: Order: Results:

Now showing items 1-5 of 5
  • Tamper, Minna; Oksanen, Arttu; Tuominen, Jouni; Hietanen, Aki; Hyvönen, Eero (2020)
    Texts referencing court decisions and statutes can be difficult to understand without context. It can be time consuming and expensive to find related statutes or to learn about context specific terminology. As a solution, we utilized a named entity linking tool for extracting information and tailored it into a service, Appi, that can automatically annotate legal documents to provide context to the readers. The service can identify and link named entities and references to legal texts to corresponding vocabularies and data sources by combining statistics- and rule-based named entity recognition with named entity linking. The results provide users with enhanced reading experience with contextual information and the possibility to access related materials, such as statutes and court decisions.
  • Tamper, Minna; Leal, Rafael; Sinikallio, Laura; Leskinen, Petri; Tuominen, Jouni; Hyvönen, Eero (, 2022)
    CEUR Workshop Proceedings
    This paper presents knowledge extraction and natural language processing methods used to enrich the knowledge graph of the plenary debates (textual transcripts of speeches) of the Parliament of Finland. This knowledge graph includes some 960 000 speeches (1907–2021) interlinked with a prosopographical knowledge graph about the politicians. A recent subset of the speeches was used to extract named entities and topical keywords for semantic searching and browsing the data and for data analysis. The process is based on linguistic analysis, named entity linking, and automatic subject indexing. The results were included into the ParliamentSampo knowledge graph in a SPARQL endpoint. This data can be used for studying parliamentary language and culture in Digital Humanities research and for developing applications, such as the ParliamentSampo portal.
  • Nevalainen, Terttu; Vartiainen, Turo; Säily, Tanja; Kesäniemi, Joonas; Dominowska, Agata; Öhman, Emily (2016)
    We introduce the Language Change Database (LCD), which provides access to the results of previous corpus-based research dealing with change in the English language. The LCD will be published on an open-access linked data platform that will allow users to enter information about their own publications into the database and to conduct searches based on linguistic and extralinguistic parameters. Both metadata and numerical data from the original publications will be available for download, enabling systematic reviews, meta-analyses, replication studies and statistical modelling of language change. The LCD will be of interest to scholars, teachers and students of English.
  • Kesäniemi, Joonas; Vartiainen, Turo; Säily, Tanja; Nevalainen, Terttu (, 2018)
    CEUR Workshop Proceedings
  • Rantala, Heikki; Ikkala, Esko; Jokipii, Ilkka; Hyvönen, Eero (2022)
    This article presents the semantic portal and Linked Open Data service WARVICTIMSAMPO 1914-1922 about the war victims, battles, and prisoner camps in the Finnish Civil and other wars in 1914-1922. The system is based on a database of the National Archives of Finland and additional related data created, compiled, and linked during the project. The system contains detailed information about some 40,000 deaths extracted from several data sources and data about over 1,000 battles of the Civil War. A key novelty of WARVICTIMSAMPO 1914-1922 is the integration of ready-to-use Digital Humanities visualizations and data analysis tooling with semantic faceted search and data exploration, which allows, e.g., studying data about wider prosopographical groups in addition to individual war victims. The article focuses on demonstrating how the tools of the portal, as well as the underlying SPARQL endpoint openly available on the Web, can be used to explore and analyze war history in flexible and visual ways. WARVICTIMSAMPO 1914-1922 is a new member in the series of "Sampo" model-based semantic portals. The portal is in use and has had 23,000 users, including both war historians and the general public seeking information about their deceased relatives.