Visualizing and Analyzing Networks of Named Entities in Biographical Dictionaries for Digital Humanities Research

Abstract

This paper shows how named entity extraction and network analysis can be used to examine biographies individually and in groups to aid historians in biographical and prosopographical research. For this purpose a reference network of 13100 biography entries in the collections of the Biographical Centre of the Finnish Literature Society was created, based on links between the biographies as well as automatically extracted named entities found in the texts. The data was published in a SPARQL endpoint as a Linked Data knowledge graph on top of which network analytic tools were created and analysis were done showing the usefulness of the approach in Digital Humanities. The reference graph has been utilized for network analysis to examine egocentric networks of individual persons as well as networks among groups of people in prosopography. The data and tools presented are in use in the new national semantic portal BiographySampo published in the autumn of 2018.

Automatic Annotation Service APPI: Named Entity Linking in Legal Domain

“…However, the tool does not have a powerful disambiguation system like other named entity linking tools, e.g., DBpedia Spotlight 14 [8] and Gate Cloud 15 [7]. For this purpose, Nelli was created, and based on it, a contextual reader was implemented for the BiographySampo portal [13] for Finnish biographical texts. In BiographySampo, the entities are not extracted with Nelli in real time but in a preprocessing phase that ensures robust semantic disambiguation similarly to [10,2].…”

“…In order to automatically annotate the legal texts of Semantic Finlex, the Nelli tool [13] was developed further. Nelli is a combination of NER (FiNER [11,4], LINFER [13]) and NEL (ARPA [5]) tools and it disambiguates entities using a scoring scheme where popularity of the interpretation of the named entity type, the string length, and successful linking are taken into account. Initially, Nelli was a command line tool that could be only used for annotating text documents.…”

“…To support end users in close reading and to enable linking of legal texts, we created a service called Appi 5 . It utilizes a named entity linking tool, Nelli [13], for identifying domain specific information and to enable named entity linking of legal texts. As a result, the Appi service can identify and link named entities, terminology, and references to legal texts to corresponding vocabularies and data sources by combining statistics-and rule-based named entity recognition (NER) with named entity linking (NEL).…”

Extracting Genealogical Networks of Linked Data from Biographical Texts

“…It contains 13 144 biographies of notable Finns that can be browsed through a faceted search application and using tools for Digital Humanities research. [6] In addition to the genealogical network discussed in this paper, the data been a source for reference network extraction [7,8].…”

“…The years of birth, death, or marriage are easily separated from the text sequence. To separate occupational descriptions from the proper names, we used the ARPA service 7 together with vocabularies of Finnish female, male, and family names 8 .…”