Exploring entity recognition and disambiguation for cultural heritage collections

Abstract

Unstructured metadata fields such as 'description' offer tremendous value for users to understand cultural heritage objects. However, this type of narrative information is of little direct use within a machine-readable context due to its unstructured nature. This paper explores the possibilities and limitations of Named-Entity Recognition (NER) and Term Extraction (TE) to mine such unstructured metadata for meaningful concepts. These concepts can be used to leverage otherwise limited searching and browsing operations, but they can also play an important role to foster Digital Humanities research. In order to catalyze experimentation with NER and TE, the paper proposes an evaluation of the performance of three third-party entity extraction services through a comprehensive case study, based on the descriptive fields of the Smithsonian Cooper-Hewitt National Design Museum in New York. In order to cover both NER and TE, we first offer a quantitative analysis of named-entities retrieved by the services in terms of precision and recall compared to a manually annotated gold-standard corpus, then complement this approach with a more qualitative assessment of relevant terms extracted. Based on the outcomes of this double analysis, the conclusions present the added value of entity extraction services, but also indicate the dangers of uncritically using NER and/or TE, and by extension Linked Data principles, within the Digital Humanities. All metadata and tools used within the paper are freely available, making it possible for researchers and practitioners to repeat the methodology. By doing so, the paper offers a significant contribution towards understanding the value of entity recognition and disambiguation for the Digital Humanities.

Developing Onomastic Gazetteers and Prosopographies for the Ancient World Through Named Entity Recognition and Graph Visualization: Some Examples from Trismegistos People

“…NER was originally developed by computational linguists in the 1990s, but quickly spread to other fields, such as biology and genetics [3] and is now gaining momentum in the Digital Humanities [4]. The problem with NER-systems, however, is that techniques designed for one genre or field do not necessarily work for others, due to specific text properties (some follow strict writing constraints, e.g.…”

HistoGraph -- A Visualization Tool for Collaborative Analysis of Networks from Historical Social Multimedia Collections

“…Frequently, textual collections are searched for occurrences of specific entities such as individuals or institutions in order to study their interaction with other entities: e.g. as in the analysis of the Smithsonian Cooper-Hewitt National Design Museum in New York [20], historical newspapers [9] or literary fiction [13]. Beyond text documents, multimedia objects (e.g.…”

A comparison of NER tools w.r.t. a domain-specific vocabulary

“…Recent work of Aldo Gangemi compares the performances of such tools in their recognition of general purpose entities in a New York Times news article [2]. In the cultural heritage context, [3] have compared some tools in their ability to link to the Linked Data cloud.…”