scholarly journals Utilization of HFACS – Cognitive Map in Human Error Classification of Ship Collisions in Indonesia

2019 ◽  
Keyword(s):  
Human Error ◽  
Cognitive Map ◽  
Error Classification

scholarly journals On the Human Error Classification of Marine Collision Accidents

2002 ◽  
Vol 106 (0) ◽  
pp. 39-46 ◽  
Author(s):  
Takahiro TAKEMOTO ◽  
Yoshiharu SAKAMOTO ◽  
Hiroyuki SHIMADA ◽  
Masao FURUSHO
Keyword(s):  
Human Error ◽  
Error Classification

A Comparison of Rule-Based and Machine Learning Models for Classification of Human Factors Aviation Safety Event Reports

2020 ◽  
Vol 64 (1) ◽  
pp. 129-133
Author(s):  
Katherine Darveau ◽  
Daniel Hannon ◽  
Chad Foster
Keyword(s):  
Machine Learning ◽  
Human Factors ◽  
Human Error ◽  
Data Science ◽  
Aircraft Engine ◽  
Rule Based ◽  
Root Cause ◽  
Textual Data ◽  
Safety Event

There is growing interest in the study and practice of applying data science (DS) and machine learning (ML) to automate decision making in safety-critical industries. As an alternative or augmentation to human review, there are opportunities to explore these methods for classifying aviation operational events by root cause. This study seeks to apply a thoughtful approach to design, compare, and combine rule-based and ML techniques to classify events caused by human error in aircraft/engine assembly, maintenance or operation. Event reports contain a combination of continuous parameters, unstructured text entries, and categorical selections. A Human Factors approach to classifier development prioritizes the evaluation of distinct data features and entry methods to improve modeling. Findings, including the performance of tested models, led to recommendations for the design of textual data collection systems and classification approaches.

scholarly journals Automated Negative Lightning Return Strokes Classification System

2021 ◽  
Vol 2107 (1) ◽  
pp. 012022
Author(s):  
F. Abdul Haris ◽  
M.Z.A. Ab Kadir ◽  
S. Sudin ◽  
D. Johari ◽  
J. Jasni ◽  
...  
Keyword(s):  
Classification System ◽  
Automatic System ◽  
Human Error ◽  
Processing Time ◽  
Visual Inspection ◽  
Automated System ◽  
Graphic User Interface ◽  
Data Comparison ◽  
Good Agreement

Abstract Over the years, many studies have been conducted to measure and classify the lightning-generated electric field waveform for a better understanding of the lightning physics phenomenon. Through measurement and classification, the features of the negative lightning return strokes can be accessed and analysed. In most studies, the classification of negative lightning return strokes was performed using a conventional approach based on manual visual inspection. Nevertheless, this traditional method could compromise the accuracy of data analysis due to human error, which also required a longer processing time. Hence, this study developed an automated negative lightning return strokes classification system using MATLAB software. In this study, a total of 115 return strokes was recorded and classified automatically by using the developed system. The data comparison with the Tenaga Nasional Berhad Research (TNBR) lightning report showed a good agreement between the lightning signal detected from this study with those signals recorded from the report. Apart from that, the developed automated system was successfully classified the negative lightning return strokes which this parameter was also illustrated on Graphic User Interface (GUI). Thus, the proposed automatic system could offer a practical and reliable approach by reducing human error and the processing time while classifying the negative lightning return strokes.

scholarly journals Morphologische Verbfehler in elizitierten Narrativen bei russisch-deutsch bilingualen Kindern im Grundschulalter

2019 ◽  
Vol 62 ◽  
pp. 118-145
Author(s):  
Olga Vorobyeva
Keyword(s):  
Language Acquisition ◽  
Data Collection ◽  
Primary School ◽  
Assessment Instrument ◽  
Older Children ◽  
Error Classification

This paper focuses on morphological verb errors in elicited narratives of Russian-German primary school bilinguals. The data was collected from 37 children who were separated into four groups according to the age and language acquisition type (simultaneous and successive). The Multilingual Assessment Instrument for Narratives (MAIN) (Gagarina et al. 2012) was used for data collection. The narratives produced in mode telling after listening to a model story were analysed and morphological verb errors in Russian and German were classified. Therefore, the error classification of Gagarina (2008) for Russian monolingual children was expanded and for the classification of German errors an own classification was suggested. Errors in Russian typically produced by monolinguals and unique bilingual errors as well were documented. The results show that the language of the environment (German) increases with age. Older children make fewer errors than younger ones. Nevertheless, a strong heterogeneity between children within each group can be observed.  

scholarly journals The Australian Incident Monitoring Study: An Analysis of 2000 Incident Reports

1993 ◽  
Vol 21 (5) ◽  
pp. 520-528 ◽  
Author(s):  
R. K. Webb ◽  
M. Currie ◽  
C. A. Morgan ◽  
J. A. Williamson ◽  
P. Mackay ◽  
...  
Keyword(s):  
Patient Outcome ◽  
Human Error ◽  
Safety Margin ◽  
Monitoring Study ◽  
Set Up ◽  
Closed Claims ◽  
Voluntary Basis ◽  
Type Of Anaesthesia ◽  
Incident Monitoring

The Australian Patient Safety Foundation was formed in 1987; it was decided to set up and co-ordinate the Australian Incident Monitoring Study as a function of this Foundation; 90 hospitals and practices joined the study. Participating anaesthetists were invited to report, on an anonymous and voluntary basis, any unintended incident which reduced, or could have reduced, the safety margin for a patient. Any incident could be reported, not only those which were deemed “preventable” or were thought to involve human error. The Mark I AIMS form was developed which incorporated features and concepts from several other studies. All the incidents in this symposium were reported using this form, which contains general instructions to the reporter, key words and space for a narrative of the incident, structured sections for what happened (with subsections for circuitry incidents, circuitry involved, equipment involved, pharmacological incidents and airway incidents), why it happened (with subsections for factors contributing to the incident, factors minimising the incident and suggested corrective strategies), the type of anaesthesia and procedure, monitors in use, when and where the incident happened, the experience of the personnel involved, patient age and a classification of patient outcome. Enrolment, reporting and data-handling procedures are described. Data on patient outcome are presented; this is correlated with the stages at which the incident occurred and with the ASA status of the patients. The locations at which the incidents occurred and the types of procedures, the sets of incidents analysed in detail and a breakdown of the incidents due to drugs are also presented. The pattern and relative frequencies of the various categories of incidents are similar to those in “closed-claims” studies, suggesting that AIMS should provide information of relevance to those wishing to develop strategies to reduce the incidence and/or impact of incidents and accidents.

Abstract WP73: Automatic Classification of Clinical MRI Stroke Datasets With a Recurrent Convolutional Neural Network

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Yichuan Liu ◽  
Brandon L Hancock ◽  
Tri Hoang ◽  
Mark R Etherton ◽  
Steven J Mocking ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Human Error ◽  
Clinical Information ◽  
Stroke Care ◽  
Automated Classification ◽  
Aggregate Information ◽  
Clinical Mri ◽  
Testing Set

Background: Fundamental advances in stroke care will require pooling imaging phenotype data from multiple centers, to complement the current aggregation of genomic, environmental, and clinical information. Sharing clinically acquired MRI data from multiple hospitals is challenging due to inherent heterogeneity of clinical data, where the same MRI series may be labeled differently depending on vendor and hospital. Furthermore, the de-identification process may remove data describing the MRI series, requiring human review. However, manually annotating the MRI series is not only laborious and slow but prone to human error. In this work, we present a recurrent convolutional neural network (RCNN) for automated classification of the MRI series. Methods: We randomly selected 1000 subjects from the MRI-GENetics Interface Exploration study and partitioned them into 800 training, 100 validation and 100 testing subjects. We categorized the MRI series into 24 groups (see Table). The RCNN used a modified AlexNet to extract features from 2D slices. AlexNet was pretrained on ImageNet photographs. Since clinical MRI are 3D and 4D, a gated recurrent unit neural network was used to aggregate information from multiple 2D slices to make the final prediction. Results: We achieved a classification accuracy (correct/total cases) of 99.8%, 98.5% and 97.5% on the training, validation and testing set, respectively. The averaged F1-score (percent overlap between predicted cases and actual cases) over all categories were 99.8% 98.2% and 94.4% on the training, validation and testing set. Conclusion: We showed that automated annotation of MRI series by repurposing deep-learning techniques used for photographic image recognition tasks is feasible. Such methods can be used to facilitate high throughput curation of MRI data acquired across multiple centers and enable scientifically productive collaboration by researchers and, ultimately enhancing big data stroke research.

Sign in / Sign up

Export Citation Format

Share Document