Improving Text Mining with Controlled Natural Language: A Case Study for Protein Interactions

Kuhn, Tobias; Royer, Loïc; Fuchs, Norbert E.; Schröder, Michael

doi:10.1007/11799511_7

Cited by 20 publications

(15 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even though these abstracts are attached to narrative articles, the formally represented findings can be processed and interpreted independently from the narrative text. We proposed a similar approach in previous work with abstracts in controlled natural language [28].…”

Section: Related Workmentioning

confidence: 99%

Genuine semantic publishing

Kuhn

Dumontier

2017

Self Cite

View full text Add to dashboard Cite

Editor: Silvio Peroni (https://orcid.org/0000-0003-0530-4305) Solicited reviews: Sarven Capadisli (https://orcid.org/0000-0002-0880-9125); Francesco Osborne (https://orcid.org/0000-0001-6557-3131); Jodi Schneider (https://orcid.org/0000-0002-5098-5667) Closer inspection, however, reveals that these approaches are mostly not about publishing semantic representations, as the name seems to suggest. Rather, they take the processes and outcomes of the current narrative-based publishing system for granted and only work with already published papers. This includes approaches involving semantic annotations, semantic interlinking, semantic integration, and semantic discovery, but with the semantics coming into play only after the publication of the original article. While these are interesting and important approaches, they fall short of providing a vision to transcend the current publishing paradigm. We argue here for taking the term semantic publishing literally and work towards a vision of genuine semantic publishing, where computational tools and algorithms can help us with dealing with the wealth of human knowledge by letting researchers capture their research results with formal semantics from the start, as integral components of their publications. We argue that these semantic components should furthermore cover at least the main claims of the work, that they should originate from the authors themselves, and that they should be fine-grained and light-weight for optimized re-usability and minimized publication overhead. This paper is in fact not just advocating our concept, but is itself a genuine semantic publication, thereby demonstrating and illustrating our points.

show abstract

Section: Related Workmentioning

confidence: 99%

Genuine semantic publishing

Kuhn

Dumontier

2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Controlled languages define a restricted subset of English and are used for semantification in several fields, e.g. Attempto Controlled English (ACE) [10,11], Common Logic Controlled English (CLCE) [12], Biological Expression Language (BEL) [13]. Each one consists of: 1) a controlled vocabulary , which is its set of unique terms that each represent a specific concept in a domain; and 2) a controlled syntax , defining how users may combine the terms to construct the formal, controlled sentences of the language, that are recognised by a specific parser algorithm.…”

Section: Introductionmentioning

confidence: 99%

Jointly creating digital abstracts: dealing with synonymy and polysemy

Vercruysse

Kuiper

2012

BMC Res Notes

View full text Add to dashboard Cite

BackgroundIdeally each Life Science article should get a ‘structured digital abstract’. This is a structured summary of the paper’s findings that is both human-verified and machine-readable. But articles can contain a large variety of information types and contextual details that all need to be reconciled with appropriate names, terms and identifiers, which poses a challenge to any curator. Current approaches mostly use tagging or limited entry-forms for semantic encoding.FindingsWe implemented a ‘controlled language’ as a more expressive representation method. We studied how usable this format was for wet-lab-biologists that volunteered as curators. We assessed some issues that arise with the usability of ontologies and other controlled vocabularies, for the encoding of structured information by ‘untrained’ curators. We take a user-oriented viewpoint, and make recommendations that may prove useful for creating a better curation environment: one that can engage a large community of volunteer curators.ConclusionsEntering information in a biocuration environment could improve in expressiveness and user-friendliness, if curators would be enabled to use synonymous and polysemous terms literally, whereby each term stays linked to an identifier.

show abstract

“…In a similar approach [5] propose to annotate scientific publications with summaries in controlled natural languages, and point out that these annotations can be used for question answering and a number of additional reasoning tasks.…”

Section: Annotations In Controlled Natural Languagementioning

confidence: 99%

“…However, exclusively relying on description logic would considerably reduce the expressive power of the controlled natural language [5].…”

Section: Controlled Natural Languages For Web Annotationsmentioning

confidence: 99%

Web-Annotations for Humans and Machines

Fuchs

Schwitter

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. We propose to manually annotate web pages with computerprocessable controlled natural language. These annotations have well-defined formal properties and can be used as query relevant summaries to automatically answer questions expressed in controlled natural language, and as the basis for other forms of automated reasoning. Last, but not least, the annotations can also serve as human-readable summaries of the contents of the web pages. Arguably, annotations written in controlled natural language can bridge the gap between informal and formal notations and leverage true collaboration between humans and machines. This is a position paper that proposes a solution combining existing methods and techniques to achieve a highly relevant practical goal, namely how to effectively access information on the web. However, our solution introduces a "chicken and egg" problem: a critical mass of web annotations will be necessary that people perceive the value of these annotations and start annotating web pages themselves. Only the future will show whether this -basically non-technical -problem can be solved.

show abstract

Improving Text Mining with Controlled Natural Language: A Case Study for Protein Interactions

Cited by 20 publications

References 15 publications

Genuine semantic publishing

Genuine semantic publishing

Jointly creating digital abstracts: dealing with synonymy and polysemy

Web-Annotations for Humans and Machines

Contact Info

Product

Resources

About