A Hybrid Approach to Semi-supervised Named Entity Recognition in Health, Safety and Environment Reports

Sari, Yunita; Hassan, Mohd Fadzil; Zamin, Norshuhani

doi:10.1109/icfcc.2009.52

Cited by 7 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to achieve the aim of this paper, the NLP method needs to successfully identify and label terms within the text as hazard, cause or consequence. Phrase matching methodologies and Named Entity Recognition (NER) are strong contenders as a solution being that it seeks to semantically recognize and identify the occurrences of a given, predefined phrase in an annotated text [11].…”

Section: Application Of Nlpmentioning

confidence: 99%

“…Previous applications of NER to safety incident and accident reports include research into using a link grammar parser and basilisk bootstrapping algorithm to recognize entities in health and safety reports [11]. While NER was used by Razavi et al [12] to identify features such as time and date from text in order to determine risk within the maritime domain.…”

Section: Application Of Nlpmentioning

confidence: 99%

See 1 more Smart Citation

An NLP framework for extracting causes, consequences, and hazards from occurrence reports to validate a HAZOP study

Ricketts

Pelham

Barry

et al. 2022

2022 IEEE/AIAA 41st Digital Avionics Systems Conference (DASC)

View full text Add to dashboard Cite

A substantial amount of effort and resource is applied to the design of aircraft systems to reduce risk to life and improve safety. This is often applied through a variety of safety assessment methods, one of which being Hazard and Operability (HAZOP) Studies. Once an air system is in-service, it is common for flight data to be collected and analysed to validate the original safety assessment. However, the operator of the air system generates and stores a substantial amount of safety knowledge within free-text occurrence reports. These allow maintainers and aircrew to report occurrences, often describing hazards and associated detail revealing consequences and causes. A lack of resource means it is difficult for safety professionals to manually review these occurrences and although occurrences are classified against a set taxonomy (e.g., birdstrike, technical failure) this lacks the granularity to apply to a specific safety analysis. To resolve this, the paper presents the development of a novel Natural Language Processing (NLP) framework for extracting causes, consequences, and hazards from free-text occurrence reports in order to validate and inform an aircraft sub-system HAZOP study. Specifically using a combination of rule-based phrase matching with a spaCy Named Entity Recognition (NER) model. It is suggested that the framework could form a continual improvement process whereby the findings drive updates to the HAZOP, in turn updating the rules and model, therefore improving accuracy and hazard identification over time.

show abstract

Section: Application Of Nlpmentioning

confidence: 99%

Section: Application Of Nlpmentioning

confidence: 99%

An NLP framework for extracting causes, consequences, and hazards from occurrence reports to validate a HAZOP study

Ricketts

Pelham

Barry

et al. 2022

2022 IEEE/AIAA 41st Digital Avionics Systems Conference (DASC)

View full text Add to dashboard Cite

show abstract

“…In semi-supervised approach, a model is trained on an initial set of labelled data and true labels, then, predictions are made on a separate set of unlabeled data, and then improved models are created iteratively using predictions of previously developed models. [16]. For example, a system aimed at "disease names" could prompt the user to give a small number of example names.…”

Section: Semi Supervised Learningmentioning

confidence: 99%

Name Entity Recognition by New Framework Using Machine Learning Algorithm

Kaur¹,

Verma²

2014

IOSRJCE

View full text Add to dashboard Cite

The amount of textual information available electronically has made it difficult for many users to find and access the right information within acceptable time. Research communities in the natural language processing (NLP) field are developing tools and techniques to alleviate these problems and help users in exploiting these vast resources. These techniques include Information Retrieval (IR) and Information Extraction (IE). The work described in this thesis concerns IE and more specifically, named entity extraction in English. The English language is of significant interest to the NLP community mainly due to its political and economic significance, but also due to its interesting characteristics. Text usually contains all kinds of names such as person names, company names, city and country names, sports teams, chemicals and lots of other names from specific domains. These names are called Named Entities (NE) and Named Entity Recognition (NER), one of the main tasks of IE systems, seeks to locate and classify automatically these names into predefined categories. NER systems are developed for different applications and can be beneficial to other information management technologies as it can be built over an IR system or can be used as the base module of a Data Mining application. In this thesis we propose an efficient and effective framework for extracting Arabic NEs from text using a rule based approach. Our approach makes use of English contextual and morphological information to extract named entities. The context is represented by means of words that are used as clues for each named entity type. Morphological information is used to detect the part of speech of each word given to the morphological analyzer. Subsequently we developed and implemented our rules in order to recognize each position of the named entity. Finally, our system implementation, evaluation metrics and experimental results are presented. We Present our Methodology by this Paper. Which use Hybrid approach of NlP and Machine Learning. This paper is a Review paper and Introduce Our Methodlogy.

show abstract

“…It takes a small set of labeled data referred as the 'seeds' to initiate the learning process. Some successful semi supervised NERs using the 'bootstrapping' technique are demonstrated by Sari, Hassan and Zamin (2009), Pasca, Lin, Bigham, Lifchits & Jain (2006) and Heng & Grishman (2006). Unsupervised learning is a method where the machine learns from unlabeled data.…”

Section: Ner Approachesmentioning

confidence: 99%

Projecting Named Entity Tags From a Resource Rich Language to a Resource Poor Language

Zamin

Oxley

Bakar

2013

Journal of Information and Communication Technology

Self Cite

View full text Add to dashboard Cite

Named Entities (NE) are the prominent entities appearing in textual documents. Automatic classification of NE in a textual corpus is a vital process in Information Extraction and Information Retrieval research. Named Entity Recognition (NER) is the identification of words in text that correspond to a pre-defined taxonomy such as person, organization, location, date, time, etc. This article focuses on the person (PER), organization (ORG) and location (LOC) entities for a Malay journalistic corpus of terrorism. A projection algorithm, using the Dice Coefficient function and bigram scoring method with domain-specific rules, is suggested to map the NE information from the English corpus to the Malay corpus of terrorism. The English corpus is the translated version of the Malay corpus. Hence, these two corpora are treated as parallel corpora. The method computes the string similarity between the English words and the list of available lexemes in a pre-built lexicon that approximates the best NE mapping. The algorithm has been effectively evaluated using our own terrorism tagged corpus; it achieved satisfactory results in terms of precision, recall, and F-measure. An evaluation of the selected open source NER tool for English is also presented.

show abstract

A Hybrid Approach to Semi-supervised Named Entity Recognition in Health, Safety and Environment Reports

Cited by 7 publications

References 10 publications

An NLP framework for extracting causes, consequences, and hazards from occurrence reports to validate a HAZOP study

An NLP framework for extracting causes, consequences, and hazards from occurrence reports to validate a HAZOP study

Name Entity Recognition by New Framework Using Machine Learning Algorithm

Projecting Named Entity Tags From a Resource Rich Language to a Resource Poor Language

Contact Info

Product

Resources

About