LL(O)D and NLP Perspectives on Semantic Change for Humanities Research

Tracking #: 2848-4062

Florentina Armaselu
Elena-Simona Apostol
Fahad Khan
Chaya Liebeskind
Barbara McGillivray
Ciprian-Octavian Truica
Andrius Utka
Giedrė Valūnaitė Oleškevičienė
Marieke van Erp

Responsible editor: 
Philipp Cimiano

Submission type: 
Survey Article
The paper presents an overview of the LL(O)D and NLP methods, tools and data for detecting and representing semantic change, with main application in humanities research. Its aim is to provide the starting points for the construction of a workflow and set of multilingual diachronic ontologies within the humanities use case of the COST Action \textit{Nexus Linguarum, European network for Web-centred linguistic data science}, CA18209. The survey focuses on the essential aspects needed to understand the current trends and to build applications in this area of study.
Full PDF Version: 


Solicited Reviews:
Click to Expand/Collapse
Review #1
By Enrico Daga submitted on 21/Sep/2021
Review Comment:

The authors have performed significant changes to the content, which significantly improve the article with respect to the previous submission.

Following the four SWJ criteria for survey articles, the current version is indeed a good (1) introductory text, targeted at researchers, PhD students, or practitioners, to get started on the covered topic.

The addition of a methodology section gives reasonable justification for (2) how comprehensive and how balanced are the presentation and coverage. However, a few more details about the sources of the survey would be useful, especially if mentioning keywords and phrases used in the search (Scopus? Google Scholar? Microsoft Academia? …). In addition, it is still a bit opaque what is intended with "refining and balancing the structure of the covered areas" - end of Section 2. However, I consider these minor issues that can be fixed during the preparation of the camera-ready.

Finally, the article is readable and clear (3) and the content is relevant to the community (4).

Review #2
By Julia Bosque submitted on 04/Oct/2021
Review Comment:

I reviewed a previous version of this manuscript, for which I recommended a major revision based on the need for a clearer motivation, scope and limitations of this effort, as well as on the structure and flow of the paper at that time. In this new version the authors have addressed all my highlighted concerns:

- The motivation, scope and limitations are clearly defined

- The interplay between the different sections is elaborated and illustrated with a workflow diagram that facilitates reading. There are numerous references to this workflow and interlinks between the sections, resulting in a cohesive document.

- More context is provided in the introductory paragraphs of each section, and the project in which this effort is carried out is clearly introduced. The relation of each section/topic with respect to the overall topic of the survey is now explicit.

- The authors have improved the categorization of tools and approaches.

- The tables in the appendix summarize the main approaches, tools and resources surveyed according to the proposed classification.

Taking into account these modifications, I maintain the reasons upon which I based the recommendation for acceptance in terms of the criteria for surveys:

- The topic of the paper, at the intersection of humanities and the Semantic Web, is interesting and relevant for the advancement in a line of research which poses numerous challenges.

- The quality of writing is good and the survey is well balanced, with a broad coverage encompassing theoretical standpoints and approaches, tools, repositories and datasets.

- The granularity and length are also appropriate for the text to serve as an introductory text.

Minor comments for improvement:

- The authors have provided details on the methodology for the survey, indicating the different stages in the generation and keywords used in literature search. There is no explicit reference to a filtering process after those keyword-based search results, was there any filtering step? If so, which criteria were applied?

- In table 3, the included resources diverge in their nature, so the current list groups together LLOD Cloud, Lila Etymological Lexicon, LingHub, and Diachronic semantic lexicon of Dutch, etc. for example. I suggest including a mark here to distinguish which resources are particularly relevant for diachronic analysis, in contrast to general LLOD resources (e.g. Lila Etymological Lexicon vs. LLOD cloud and LingHub).

- The authors of [12], referenced on p. 5, mention Lemon (Lexicon Model For Ontologies), and in their diagrams (in Github) they seem to be using OntoLex-Lemon, not its ancestor. Throughout this survey "OntoLex-Lemon" is the term used to refer to the 2016 Specification as the outcome of the W3C Ontology-Lexica Community Group, so for that bib. reference I would recommend to replace the mention of "Lemon" with "OntoLex-Lemon" for consistency in the whole document.


l.19, .p. 19, right column, "A combined resource like this, allows..." → remove comma

p. 20, l. 1, right column → remove "(linguistic)", already covered by the first L in LLOD

Appendix tables, Table 4. → word embeddings (add pl. "s")

Review #3
By Thierry Declerck submitted on 05/Nov/2021
Review Comment:

Not really a lot to add. I see that the revised version of the submission was taking good care of former comments and suggestions.
Just a minor point: 1) Ensure that footnotes are always placed after the punctuation signs (for consistency across the paper, se fn 2 and 3 which are not placed consistently). So, very few corrections to do.