Using contextual data to predict risky driving events: A novel methodology from explainable artificial intelligence.

Usage-based insurance has allowed insurers to dynamically tailor insurance premiums by understanding when and how safe policyholders drive. However, telematics information can also be used to understand the driving contexts experienced by the driver within each trip (e.g., road types, weather, traffic). Since different combinations of these conditions affect exposure to accidents, this understanding introduces predictive opportunities in driving risk assessment. This paper investigates the relationships between driving context combinations and risk using a naturalistic driving dataset of 77,859 km. In particular, XGBoost and Random Forests are used to determine the predictive significance of driving contexts for near-misses, speeding and distraction events. Moreover, the most important contextual factors in predicting these risky events are identified and ranked through Shapley Additive Explanations. The results show that the driving context has significant power in predicting driving risk. Speed limit, weather temperature, wind speed, traffic conditions and road slope appear in the top ten most relevant features for most risky events. Analysing contextual feature variations and their influence on risky events showed that low-speed limits increase the predicted frequency of speeding and phone unlocking events, whereas high-speed limits decrease harsh accelerations. Low temperatures decrease the expected frequency of harsh manoeuvres, and precipitations increase harsh acceleration, harsh braking, and distraction events. Furthermore, road slope, intersections and pavement quality are the most critical factors among road layout attributes. The methodology presented in this study aims to support road safety stakeholders and insurers by providing insights to study the contextual risk factors that influence road accident frequency and driving risk.

On the impact of advanced driver assistance systems on driving distraction and risky behaviour: An empirical analysis of irish commercial drivers.

Advanced driver assistance systems (ADAS) present promising benefits in mitigating road collisions. However, these benefits are limited when risky drivers continue engaging in distraction events. While there is evidence that real-time warnings help improve driving behaviour, the sustained benefits of warning-based ADAS on reducing driving distraction in light commercial vehicle (LCV) drivers remain unclear. This research determines the effect of receiving instant distraction warnings over two years using a naturalistic driving dataset comprising around one million trips from 373 LCV drivers in the Republic of Ireland. Furthermore, the study applies Association Rule Mining (ARM) to find the contextual variables (e.g., speed limit, road type, traffic conditions) that increase the likelihood of distraction events. The results show that warning-based ADAS providing real-time warnings helps reduce distraction events triggering driver inattention, forward collision, and lane departure warnings. Over half of the studied fleet reduced these warnings by at least 50% - lane departure after two months and driver inattention and forward collision after six months. It is found that both passive and active monitoring systems, coupled with coaching and rewards, significantly reduce aggressive driving behaviours tied to harsh acceleration (by 76%) and harsh braking (by 65%). The results of ARM show that the driving context introduces explanatory information for road safety programs. Low-speed urban roads and the summer season increase the likelihood of driver inattention and forward collision warnings. In contrast, high-speed rural roads increase the likelihood of lane departure warnings. These research findings support road safety stakeholders in developing risk assessments based on warning-based ADAS, targeted campaigns to reduce driving distraction, and driving coaching programs.

Cyber risk and cybersecurity: a systematic review of data availability.

Cybercrime is estimated to have cost the global economy just under USD 1 trillion in 2020, indicating an increase of more than 50% since 2018. With the average cyber insurance claim rising from USD 145,000 in 2019 to USD 359,000 in 2020, there is a growing necessity for better cyber information sources, standardised databases, mandatory reporting and public awareness. This research analyses the extant academic and industry literature on cybersecurity and cyber risk management with a particular focus on data availability. From a preliminary search resulting in 5219 cyber peer-reviewed studies, the application of the systematic methodology resulted in 79 unique datasets. We posit that the lack of available data on cyber risk poses a serious problem for stakeholders seeking to tackle this issue. In particular, we identify a lacuna in open databases that undermine collective endeavours to better manage this set of risks. The resulting data evaluation and categorisation will support cybersecurity researchers and the insurance industry in their efforts to comprehend, metricise and manage cyber risks. Supplementary Information: The online version contains supplementary material available at 10.1057/s41288-022-00266-6.

Regulatory and Technical Constraints: An Overview of the Technical Possibilities and Regulatory Limitations of Vehicle Telematic Data.

A telematics device is a vehicle instrument that comes preinstalled by the vehicle manufacturer or can be added later. The device records information about driving behavior, including speed, acceleration, and turning force. When connected to vehicle computers, the device can also provide additional information regarding the mechanical usage and condition of the vehicle. All of this information can be transmitted to a central database via mobile networks. The information provided has led to new services such as Usage Based Insurance (UBI). A range of consultants, industry commentators and academics have produced an abundance of projections on how telematics information will allow the introduction of services from personalized insurance, bespoke entertainment and advertise and vehicle energy optimization, particularly for Electric Vehicles (EVs). In this paper we examine these potential services against a backdrop of nascent regulatory limitations and against the technical capacity of the devices. Using a case study approach, we examine three applications that can use telematics information. We find that the expectations of service providers will be significantly tempered by regulatory and technical hurdles. In our discussion we detail these limitations and suggest a more realistic rollout of ancillary services.

Immune or at-risk? Stock markets and the significance of the COVID-19 pandemic.

The closure of borders and traditional commerce due to the COVID-19 pandemic is expected to have a lasting financial impact. We determine whether the growth in COVID-19 affected index prices by examining equity markets in five regional epicentres, along with a 'global' index. We also investigate the impact of COVID-19 after controlling for investor sentiment, credit risk, liquidity risk, safe-haven asset demand and the price of oil. Despite controlling for these traditional market drivers, the daily totals of COVID-19 cases nevertheless explained index price changes in Spain, Italy, the United Kingdom and the United States. Similar results were not observed in China, the origin of the virus, nor in the 'global' index (MSCI World). Our results suggest that early interventions (China) and the spatiotemporal nature of pandemic epicentres (World) should be considered by governments, regulators and relevant stakeholders in the event of future COVID-19 'waves' or further extreme societal disruptions.

Associations between mobility patterns and COVID-19 deaths during the pandemic: A network structure and rank propagation modelling approach.

BACKGROUND: From February 2020, both urban and rural Ireland witnessed the rapid proliferation of the COVID-19 disease throughout its counties. During this period, the national COVID-19 responses included stay-at-home directives issued by the state, subject to varying levels of enforcement. METHODS: In this paper, we present a new method to assess and rank the causes of Ireland COVID-19 deaths as it relates to mobility activities within each county provided by Google while taking into consideration the epidemiological confirmed positive cases reported per county. We used a network structure and rank propagation modelling approach using Personalised PageRank to reveal the importance of each mobility category linked to cases and deaths. Then a novel feature-selection method using relative prominent factors finds important features related to each county's death. Finally, we clustered the counties based on features selected with the network results using a customised network clustering algorithm for the research problem. FINDINGS: Our analysis reveals that the most important mobility trend categories that exhibit the strongest association to COVID-19 cases and deaths include retail and recreation and workplaces. This is the first time a network structure and rank propagation modelling approach has been used to link COVID-19 data to mobility patterns. The infection determinants landscape illustrated by the network results aligns soundly with county socio-economic and demographic features. The novel feature selection and clustering method presented clusters useful to policymakers, managers of the health sector, politicians and even sociologists. Finally, each county has a different impact on the national total.

Application of Bayesian networks in determining nanoparticle-induced cellular outcomes using transcriptomics.

Inroads have been made in our understanding of the risks posed to human health and the environment by nanoparticles (NPs) but this area requires continuous research and monitoring. Machine learning techniques have been applied to nanotoxicology with very encouraging results. This study deals with bridging physicochemical properties of NPs, experimental exposure conditions and in vitro characteristics with biological effects of NPs on a molecular cellular level from transcriptomics studies. The bridging is done by developing and implementing Bayesian Networks (BNs) with or without data preprocessing. The BN structures are derived either automatically or methodologically and compared. Early stage nanotoxicity measurements represent a challenge, not least when attempting to predict adverse outcomes and modeling is critical to understanding the biological effects of exposure to NPs. The preprocessed data-driven BN showed improved performance over automatically structured BN and the BN with unprocessed datasets. The prestructured BN captures inter relationships between NP properties, exposure condition and in vitro characteristics and links those with cellular effects based on statistic correlation findings. Information gain analysis showed that exposure dose, NP and cell line variables were the most influential attributes in predicting the biological effects. The BN methodology proposed in this study successfully predicts a number of toxicologically relevant cellular disrupted biological processes such as cell cycle and proliferation pathways, cell adhesion and extracellular matrix responses, DNA damage and repair mechanisms etc., with a success rate >80%. The model validation from independent data shows a robust and promising methodology for incorporating transcriptomics outcomes in a hazard and, by extension, risk assessment modeling framework by predicting affected cellular functions from experimental conditions.

Hazard Screening Methods for Nanomaterials: A Comparative Study.

Hazard identification is the key step in risk assessment and management of manufactured nanomaterials (NM). However, the rapid commercialisation of nano-enabled products continues to out-pace the development of a prudent risk management mechanism that is widely accepted by the scientific community and enforced by regulators. However, a growing body of academic literature is developing promising quantitative methods. Two approaches have gained significant currency. Bayesian networks (BN) are a probabilistic, machine learning approach while the weight of evidence (WoE) statistical framework is based on expert elicitation. This comparative study investigates the efficacy of quantitative WoE and Bayesian methodologies in ranking the potential hazard of metal and metal-oxide NMs-TiO2, Ag, and ZnO. This research finds that hazard ranking is consistent for both risk assessment approaches. The BN and WoE models both utilize physico-chemical, toxicological, and study type data to infer the hazard potential. The BN exhibits more stability when the models are perturbed with new data. The BN has the significant advantage of self-learning with new data; however, this assumes all input data is equally valid. This research finds that a combination of WoE that would rank input data along with the BN is the optimal hazard assessment framework.

Application of Bayesian networks for hazard ranking of nanomaterials to support human health risk assessment.

In this study, a Bayesian Network (BN) was developed for the prediction of the hazard potential and biological effects with the focus on metal- and metal-oxide nanomaterials to support human health risk assessment. The developed BN captures the (inter) relationships between the exposure route, the nanomaterials physicochemical properties and the ultimate biological effects in a holistic manner and was based on international expert consultation and the scientific literature (e.g., in vitro/in vivo data). The BN was validated with independent data extracted from published studies and the accuracy of the prediction of the nanomaterials hazard potential was 72% and for the biological effect 71%, respectively. The application of the BN is shown with scenario studies for TiO2, SiO2, Ag, CeO2, ZnO nanomaterials. It is demonstrated that the BN may be used by different stakeholders at several stages in the risk assessment to predict certain properties of a nanomaterials of which little information is available or to prioritize nanomaterials for further screening.

A Tractable Method for Measuring Nanomaterial Risk Using Bayesian Networks.

While control banding has been identified as a suitable framework for the evaluation and the determination of potential human health risks associated with exposure to nanomaterials (NMs), the approach currently lacks any implementation that enjoys widespread support. Large inconsistencies in characterisation data, toxicological measurements and exposure scenarios make it difficult to map and compare the risk associated with NMs based on physicochemical data, concentration and exposure route. Here we demonstrate the use of Bayesian networks as a reliable tool for NM risk estimation. This tool is tractable, accessible and scalable. Most importantly, it captures a broad span of data types, from complete, high quality data sets through to data sets with missing data and/or values with a relatively high spread of probability distribution. The tool is able to learn iteratively in order to further refine forecasts as the quality of data available improves. We demonstrate how this risk measurement approach works on NMs with varying degrees of risk potential, namely, carbon nanotubes, silver and titanium dioxide. The results afford even non-experts an accurate picture of the occupational risk probabilities associated with these NMs and, in doing so, demonstrated how NM risk can be evaluated into a tractable, quantitative risk comparator.

Interaction of adrenomedullin and calcitonin gene-related peptide with the periodontal pathogen Porphyromonas gingivalis.

The nature of the interaction between Porphyromonas gingivalis and the multifunctional peptides adrenomedullin and calcitonin gene-related peptide (CGRP) was investigated. Growth of P. gingivalis was not inhibited in the presence of either of these peptides [minimal inhibitory concentration (MIC)>250 microg mL(-1)]. The ability of the arginine- and lysine-specific proteases from P. gingivalis to breakdown these peptides was investigated. Adrenomedullin and CGRP were incubated with culture supernatants from wild-type and protease gene knockout strains. No significant effect on antimicrobial activity against the indicator organism Escherichia coli BUE55 was found (MIC=6.25 microg mL(-1) in all cases). The role of anionic components on the surface of P. gingivalis, which may alter binding of these cationic peptides, was also investigated in relation to adrenomedullin. Growth of gene knockout strains lacking surface polysaccharide and capsule components was not inhibited (MIC>250 microg mL(-1)). It is suggested that a lack of sensitivity to adrenomedullin and CGRP may enable P. gingivalis to persist in the oral cavity and cause disease.

Antimicrobial activity does not predict cytokine response to adrenomedullin or its shortened derivatives.

The aim of this study was to investigate cytokine release from oral keratinocytes and fibroblasts in response to AM and shortened derivatives previously characterised in terms of their antimicrobial activities. Cells were incubated with AM or its fragments (residues 1-12, 1-21, 13-52, 16-21, 16-52, 22-52, 26-52, and 34-52), and culture supernatants collected after 1, 2, 4, 8, and 24 hours. A time-dependant increase in production of interleukin1-alpha and interleukin 1-beta from keratinocytes in response to all peptides was demonstrated. However, exposure to fragments compared to whole AM resulted in reduced production of these cytokines (60% mean reduction at 24 hours, P<.001). No consistent differences were shown between the cytokine response elicited by antimicrobial and nonantimicrobial fragments. The production of interleukin-6 and interleukin-8 did not change significantly with time or peptide used. Fibroblast cells were relatively unresponsive to all treatments. This study demonstrates that antimicrobial activity does not predict cytokine response to adrenomedullin or its shortened derivatives.

Mechanisms of adrenomedullin antimicrobial action.

The mechanism of antimicrobial action of the multifunctional peptide adrenomedullin (AM) against Escherichia coli and Staphylococcus aureus was investigated. AM (52 residues) and AM fragments (1-12, 1-21, 13-52, 16-21, 16-52, 22-52, 26-52 and 34-52 residues) were tested for activity. Carboxy-terminal fragments were shown to be up to 250-fold more active than the parent molecule. Minimum inhibitory concentration values of the most active fragments (13-52 and 16-52) and the parent molecule were 4.9 x 10(-2) and 12.5 microg/ml, respectively, with E. coli. Ultrastructural analyses of AM treated cells demonstrated marked cell wall disruption with E. coli within 0.5 h. Abnormal septum formation with no apparent peripheral cell wall disruption was observed with S. aureus after 2 h. Outer membrane permeabilisation assays with E. coli confirmed that the C-terminal fragments were significantly (P < 0.05) more active. It is suggested that postsecretory processing may generate multiple AM congeners that have enhanced antimicrobial activities against a range of potential targets.