Iterative Approach for Information Extraction and Ontology Learning from Textual Aviation Safety Reports

Saeeda, Lama

doi:10.1007/978-3-319-58451-5_18

Cited by 4 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another example of the benefits of accessing unmapped ontological data is ontology-based classification. In the aviation safety reporting tool, events can be classified using RDFS classes from a predefined vocabulary [57]. However, the application does not need to map these classes in order to use them if the OTM library allows access to these unmapped types.…”

Section: Fully Satisfied Ifmentioning

confidence: 99%

A comparison of object-triple mapping libraries

Ledvinka

Křemen

2020

View full text Add to dashboard Cite

Domain-independent information systems like ontology editors provide only limited usability for non-experts when domain-specific linked data need to be created. On the contrary, domain-specific applications require adequate architecture for data authoring and validation, typically using the object-oriented paradigm. So far, several software libraries mapping the RDF model (representing linked data) to the object model have been introduced in the literature. In this paper, we develop a novel framework for comparison of object-triple mapping solutions in terms of features and performance. For feature comparison, we designed a set of qualitative criteria reflecting object-oriented application developer's needs. For the performance comparison, we introduce a benchmark based on a real-world information system that we implemented using one of the compared objecttriple mapping (OTM) solutions -JOPA. We present a detailed evaluation of a selected set of OTM libraries, showing how they differ in terms of features. We further narrow down the set of selected OTM libraries to contain only RDF4J-compatible ones and use the benchmark to measure their time and memory efficiency.

show abstract

Section: Fully Satisfied Ifmentioning

confidence: 99%

A comparison of object-triple mapping libraries

Ledvinka

Křemen

2020

View full text Add to dashboard Cite

show abstract

“…The body of work on computational analysis of text, contains examples from the health and medical domain (for example Toyabe, 2012;Chase, Mitrani, Lu and Fulgieri, 2017) aviation domain (for example Saeeda, 2017), to identify information from highway accidents (for example Mannering, Shankar and Bhat, 2016) and to obtain information from social media (for example Proctor, Vis and Vos, 2013). Popping (2000) developed a categorisation of computational text analysis approaches, describing them as being either thematic, semantic, or network.…”

Section: Computational Analysis Of Safety Incident Reportsmentioning

confidence: 99%

Extracting safety information from multi-lingual accident reports using an ontology-based approach

et al. 2019

View full text Add to dashboard Cite

This paper describes an approach to extract meaning from multilingual free-text safety incident reports. A sample of 5065 safety incident reports from the Swiss Federal Office of Transport were used in the study. Each report was written in either German, French or Italian natural language. An interactive learning approach between a human and computer software was undertaken to identify key terms in the text that are relevant to discovering meaning. A multilingual ontology was created to join meaningful semantic patterns and identify specific classes of safety incident on the railway, including injuries occurring whilst passengers were boarding or alighting from vehicles, falling down stairs, struck by closing doors, or struck by objects such as suitcases. A graph database was used to query the text records via the ontology and identify reports of incidents in each class, regardless of the language used in the report. Fluent speakers of each language-German, French and Italian-reviewed the results to confirm true positive results and detect false positives. The performance of the process varied across languages and incident types, however the overall true positive rate was determined by the fluent speakers to be 98.5%.

show abstract

“…It has been demonstrated, however, that the effort required can be reduced by applying semi-automated techniques [12,13]. Many researchers have considered the task of extracting information from text data, such as in the health and medical domain (for example [14,15]) aviation domain (for example [16]), the automotive safety (for example [17]) as well as to obtain information from social media (for example [18]). Applying a thematic approach, Hughes [19] categorised source text according to the occurrence of terms in the text, where a term is either a single word or sequence of consecutive words (also known as an n-gram).…”

Section: Introductionmentioning

confidence: 99%

Observed failures at railway turnouts: Failure analysis, possible causes and links to current and future research

Grossoni

Hughes

Bezin

et al. 2021

Engineering Failure Analysis

View full text Add to dashboard Cite

Railway turnouts are a necessary part of railway infrastructure, used throughout the network to direct trains to the desired path, thus ensuring the smooth running of every train. Their reliability and availability is therefore an essential element for the efficiency of the entire network. However, the combination of movable parts, some of which with reducing cross sections and structural integrity, discontinuities in the rail geometry and variability in the system support stiffness leads to high failure rates compared with plain line track and to very high maintenance and renewal costs. In this paper, the database of failures occurred in the GB network between 2011 and 2017 has been statistically investigated with the support of free-text analysis techniques to interrogate the comments provided by the on-track maintenance staff. The key failures and possible causes have been identified for both the switch and crossing panels as well as areas for future research to support the development of the next generation of switches and crossings.

show abstract

Iterative Approach for Information Extraction and Ontology Learning from Textual Aviation Safety Reports

Cited by 4 publications

References 16 publications

A comparison of object-triple mapping libraries

A comparison of object-triple mapping libraries

Extracting safety information from multi-lingual accident reports using an ontology-based approach

Observed failures at railway turnouts: Failure analysis, possible causes and links to current and future research

Contact Info

Product

Resources

About