Semantic Annotation for Improved Safety in Construction Work

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 1990–1999 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 1990 Semantic Annotation for Improved Safety in Construction Work Paul Thompson1, Tim Yates2, Emrah Inan1 and Sophia Ananiadou1 1National Centre for Text Mining, Department of Computer Science, University of Manchester, UK, 2Health & Safety Executive, HSE Science & Research Centre, Buxton, UK {paul.thompson,emrah.inan,sophia.ananiadou}@manchester.ac.uk, tim.yates@hse.gov.uk Abstract Risk management is a vital activity to ensure employee safety in construction projects. Various documents provide important supporting evidence, including details of previous incidents, consequences and mitigation strategies. Potential hazards may depend on a complex set of project-specific attributes, including activities undertaken, location, equipment used, etc. However, finding evidence about previous projects with similar attributes can be problematic, since information about risks and mitigations is usually hidden within and may be dispersed across a range of different free text documents. Automatic named entity recognition (NER), which identifies mentions of concepts in free text documents, is the first stage in structuring knowledge contained within them. While developing NER methods generally relies on annotated corpora, we are not aware of any such corpus targeted at concepts relevant to construction safety. In response, we have designed a novel named entity annotation scheme and associated guidelines for this domain, which covers hazards, consequences, mitigation strategies and project attributes. Four health and safety experts used the guidelines to annotate a total of 600 sentences from accident reports; an average inter-annotator agreement rate of 0.79 F-Score shows that our work constitutes an important first step towards developing tools for detailed semantic analysis of construction safety documents. Keywords: Semantic Annotation, Named Entities, Construction Safety 1. Introduction The health and safety of workers is of paramount impor- tance in any construction project. As such, risk manage- ment is a vital activity, whose aims are to identify the poten- tial risks to workers at each stage of the project and to de- termine appropriate safety measures that can mitigate these risks as effectively as possible. An important source of evidence for risk management comes from existing document repositories. Reports on construction-related injuries and illnesses, site inspections and prosecutions, along with health and safety guidelines, can collectively provide details about which accidents have occurred during previous projects, together with possible reasons (e.g., failure to put adequate protection measures in place) and recommendations of how incidents can be min- imised or avoided in future. Such document repositories usually contain a large num- ber of (possibly lengthy) documents. However, information about previous projects is only likely to be relevant to risk management for new projects if there is a match between at least some of their attributes. These attributes may include location type (e.g., the terrain or type of construction), con- struction activities undertaken (e.g., demolition, digging), equipment used (e.g., ladders, drills), etc. Furthermore, in- formation about incidents, potential reasons and possible mitigations may be fragmented both within and across a range of different documents. As such, the processes of …

Content truncated to 500 words