An Analysis of Sound Event Detection under Acoustic Degradation Using Multi-Resolution Systems

The Sound Event Detection task aims to determine the temporal locations of acoustic events in audio clips. In recent years, the relevance of this field is rising due to the introduction of datasets such as Google AudioSet or DESED (Domestic Environment Sound Event Detection) and competitive evaluations like the DCASE Challenge (Detection and Classification of Acoustic Scenes and Events). In this paper, we analyze the performance of Sound Event Detection systems under diverse artificial acoustic conditions such as high- or low-pass filtering and clipping or dynamic range compression, as well as under an scenario of high overlap between events. For this purpose, the audio was obtained from the Evaluation subset of the DESED dataset, whereas the systems were trained in the context of the DCASE Challenge 2020 Task 4. Our systems are based upon the challenge baseline, which consists of a Convolutional-Recurrent Neural Network trained using the Mean Teacher method, and they employ a multiresolution approach which is able to improve the Sound Event Detection performance through the use of several resolutions during the extraction of Mel-spectrogram features. We provide insights on the benefits of this multiresolution approach in different acoustic settings, and compare the performance of the single-resolution systems in the aforementioned scenarios when using different resolutions. Furthermore, we complement the analysis of the performance in the high-overlap scenario by assessing the degree of overlap of each event category in sound event detection datasets.

Economic ripple resonance from consecutive weather extremes amplifies consumption losses

<p>Weather extremes such as heat waves, tropical cyclones, and river floods are likely to intensify with increasing global mean temperature. In a globally connected supply and trade network such extreme weather events cause economic shocks that may interfere with each other potentially amplifying their overall economic impact.</p><p>Here we analyze the economic resonance of consecutive extreme events, that is the overlapping of economic response dynamics of more than one extreme event category both spatially and temporally. In our analysis we focus on the event categories heat stress, river floods, and tropical cyclones. We simulate, via an agent-based anomaly model with more than 7,000 economic agents and 1.8 million connections, the regional (direct) and inter-regional (indirect via supply chains) economic losses and gains for each extreme event category individually as well as for their concurrent occurrence for the next two decades (2020-2039). Thus we compare the outcome of the sum of the three single simulations to the outcome of the concurrent simulation. We show that the global welfare losses due to concurrent weather extremes are increased by more than 18% due to market effects compared to the summation of the losses of each single event category. Overall, this economic resonance yields a non-linearly enhanced price effect, which leads to a stronger economic impact. As well as a highly heterogeneous distribution of the amplification of regional welfare losses among countries.</p><p>Our analysis is based on the climate models of the CMIP5 ensemble which have been bias-corrected within the ISIMIP2b project towards an observation-based data set using a trend-preserving method. From these we use RCP2.6 and 6.0 for future climate projections. Thus we compute for each of the three extreme weather event category regional, and sectoral production failure on a daily time scale. Our agent-based dynamic economic loss-propagation model <em>A</em><em>cclimate</em> then uses these local production failures to compute the immediate response dynamics within the global supply chain as well as the subsequent trade adjustments. The Acclimate model thereby depicts a highly interconnected network of firms and consumers, which maximize their profits by choosing the optimal production level and corresponding upstream demand as well as the optimal distribution of this demand among its suppliers; transport and storage inventories act as buffers for supply shocks. The model accounts for local price changes; supply and demand mismatches are resolved explicitly over time.</p><p><span>Our results suggest that economic impacts of weather extremes are larger than can be derived from conventional single event analysis. Consequently the societal cost of climate change are likely to be underestimated in studies focusing on single extreme categories.</span></p>

Responsive Leadership in Preventing Transmission of Covid-19 in the Indonesia-Malaysia Border Area

The spread of the COVID-19 is increasingly globalized to have an impact not only on human health but also on all lines of life that requires rapid responses with anticipatory and preventive steps before its effects become worse. All countries feel the impact of outbreak, which moves so fast, including border communities. Today, COVID-19 travels through the human respiratory system to one another and from one area to another including Sambas area. The purpose of this study was to investigate the leadership responses and efforts to prevent the spread of this global disease on the Malaysia-Indonesia border area, precisely in the city of Sambas. We divided the findings into two. First, the spread of COVID-19 in Sambas was declared an extraordinary event category, because Sambas is located in the border area between Malaysia and Indonesia, where the community's largest livelihood is working in Malaysia. Second, when Sambas was declared an extraordinary category of the spread of that deadly virus, the local government responded to this incident by setting specific policies, including not carrying out religious or other rituals that invited crowds. On one hand. the Sambas community responded well to the policy. However, for certain reason the local Sambal are still carried out daily activities with the excuse for daily needs.

F-HMTC: Detecting Financial Events for Investment Decisions Based on Neural Hierarchical Multi-Label Text Classification

The share prices of listed companies in the stock trading market are prone to be influenced by various events. Performing event detection could help people to timely identify investment risks and opportunities accompanying these events. The financial events inherently present hierarchical structures, which could be represented as tree-structured schemes in real-life applications, and detecting events could be modeled as a hierarchical multi-label text classification problem, where an event is designated to a tree node with a sequence of hierarchical event category labels. Conventional hierarchical multi-label text classification methods usually ignore the hierarchical relationships existing in the event classification scheme, and treat the hierarchical labels associated with an event as uniform labels, where correct or wrong label predictions are assigned with equal rewards or penalties. In this paper, we propose a neural hierarchical multi-label text classification method, namely F-HMTC, for a financial application scenario with massive event category labels. F-HMTC learns the latent features based on bidirectional encoder representations from transformers, and directly maps them to hierarchical labels with a delicate hierarchy-based loss layer. We conduct extensive experiments on a private financial dataset with elaborately-annotated labels, and F-HMTC consistently outperforms state-of-art baselines by substantial margins. We will release both the source codes and dataset on the first author's repository.

Single vs. concurrent extreme events: Economic resonance of weather extremes increases impact on societal welfare loss

<p>Weather extremes such as heat waves, tropical cyclones and river floods are likely to intensify with increasing global mean temperature. In a globally connected supply and trade network such extreme weather events cause economic shocks that may interfere with each other potentially amplifying their overall economic impact.</p><p>Here we analyze the economic resonance of concurrent extreme events, that is the overlapping of economic response dynamics of more than one extreme event category both spatially and temporally. In our analysis we focus on the event categories heat stress, river floods and tropical cyclones. We simulate the regional (direct) and global (indirect via supply chains) economic losses and gains for each extreme event category individually as well as for their concurrent occurrence for the next two decades. Thus we compare the outcome of the sum of the three single simulations to the outcome of the concurrent simulation. Here we show that the global welfare loss due to concurrent weather extremes is increased by more than 17% due to market effects compared to the summation of the losses of each single event category. Overall, this economic resonance yields a non-linearly enhanced price effect, which leads to a stronger economic impact. As well as a highly heterogeneous distribution of the amplification of regional welfare losses among countries.</p><p>Our analysis is based on the climate models of the CMIP5 ensemble which have been bias-corrected within the ISIMIP2b project towards an observation-based data set using a trend-preserving method. From these we use RCP2.6 and 6.0 for future climate projections. We transfer the three extreme weather event categories to a daily, regional and sectoral production failure. Our agent-based dynamic economic loss-propagation model <em>Acclimate</em> then uses these local production failures to compute the immediate response dynamics within the global supply chain as well as the subsequent trade adjustments. The <em>Acclimate</em> model thereby depicts a highly interconnected network of firms and consumers, which maximize their profits by choosing the optimal production level and corresponding upstream demand as well as the optimal distribution of this demand among its suppliers; transport and storage inventories act as buffers for supply shocks. The model accounts for local price changes, and supply and demand mismatches are resolved explicitly over time.</p><p>Our results suggest that economic impacts of weather extremes are larger than can be derived from conventional single event analysis. Consequently the societal cost of climate change are likely to be underestimated in studies focusing on single extreme categories.</p>

Assessing safety of Ontario’s publicly funded MMR and MMRV immunization programs, 2012 to 2016

Background The combined measles, mumps, rubella (MMR) and measles, mumps, rubella, and varicella (MMRV) vaccines are part of Ontario’s routine immunization schedule. Objective To assess adverse events following immunization (AEFIs) reported in Ontario following administration of MMR and MMRV vaccines between 2012 and 2016. Methods Reports of AEFIs were extracted from the provincial surveillance database on May 9, 2017. Events were grouped by provincial surveillance definitions. Reporting rates were calculated using provincial population estimates or net doses distributed as the denominator. A serious AEFI is defined as an AEFI that resulted in an in-patient hospitalization or death. Results Overall, 289 AEFIs were reported following administration of MMR (n=246) or MMRV (n=43) vaccines, for annualized reporting rates of 16.6 and 8.8 reports per 100,000 distributed doses, respectively. The highest age-specific reporting rate was in children aged 1 to 3 years for MMR (7.7 per 100,000 population) and children aged 4 to 9 years for MMRV (0.8 per 100,000 population). Systemic reactions were the most frequently reported event category, while rash was the most frequently reported event for both vaccines. There were 22 serious AEFIs, 19 following MMR and 3 following MMRV (1.3 and 0.6 per 100,000 doses distributed, respectively). Conclusions Our assessment found a low reporting rate of adverse events following MMR and MMRV vaccines in Ontario. No safety concerns were identified. Our findings are consistent with the safety profiles of these vaccines. Continued monitoring of vaccine safety is necessary to maintain timely detection of unusual postvaccine events and public confidence in vaccine safety.

Tourism and recurring sport events

Purpose Both groups are profiled in terms of travel-related and socio-demographic characteristics. Furthermore, the purpose of this paper is to address determinants of spending for each spectator group. Data collection was conducted using an on-site questionnaire. Analysis of variance between profile characteristics is based on χ2 and Wilcoxon-Mann-Whitney tests, whereas the analysis of determinants of spending builds on the Tobit model. Design/methodology/approach Recurring sport events that do not count among the mega sport event category have become a popular means of attracting tourists to a destination. Thus, research on different spectator groups attending such events is very relevant, yet surprisingly scarce. This study helps filling this void by a comparative analysis of two types of spectators present at the Professional Windsurf Association Windsurf World Cup on the German island of Sylt: travellers who come to Sylt solely for the event (event tourists) and travellers whose motivation to visit the island was not primarily driven by the event (regular tourists). Findings The results show that the two examined groups are clearly distinguishable, both in terms of profile characteristics and determinants of spending. This indicates that specific strategies seem advisable for sport event and tourism destination managers at mature tourist destinations. Originality/value The study’s major contribution to both tourism and event management literature is that it exposes key characteristics of and differences between both groups within a specific setting at a non-mega sport event at a mature tourist destination.