Passenger Perceptions, Information Preferences, and Usability of Crowding Visualizations on Public Displays in Transit Stations and Vehicles

Large crowds in public transit stations and vehicles introduce obstacles for wayfinding, hygiene, and physical distancing. Public displays that currently provide on-site transit information could also provide critical crowdedness information. Therefore, we examined people's crowd perceptions and information preferences before and during the pandemic, and designs for visualizing crowdedness to passengers. We first report survey results with public transit users (n = 303), including the usability results of three crowdedness visualization concepts. Then, we present two animated crowd simulations on public displays that we evaluated in a field study (n = 44). We found that passengers react very positively to crowding information, especially before boarding a vehicle. Visualizing the exact physical spaces occupied on transit vehicles was most useful for avoiding crowded areas. However, visualizing the overall fullness of vehicles was the easiest to understand. We discuss design implications for communicating crowding information to support decision-making and promote a sense of safety. © 2023 ACM.

CCTwins: A Weakly-Supervised Transformer-based Crowd Counting Method with Adaptive Scene Consistency Attention

Recently, crowd counting has attracted significant attention, particularly in the context of the COVID-19 pandemic, due to its ability to automatically provide accurate crowd numbers in images. To address the challenges of location-level labeling, several transformer-based crowd counting methods have been proposed with only count-level supervision. However, these methods directly use the transformer as an encoder without considering the uneven crowd distribution. To address this issue, we propose CCTwins, a novel transformer-based crowd counting method with only count-level supervision. Specifically, we introduce an adaptive scene consistency attention mechanism to enhance the transformer-based model Twins-SVT-L for feature extraction in crowded scenes. Additionally, we design a multi-level weakly-supervised loss function that generates estimated crowd numbers in a coarse-to-fine manner, making it more appropriate for weakly-supervised settings. Moreover, intermediate features supervised by count-level labels are utilized to fuse multi-scale features. Experimental results on four public datasets demonstrate that our proposed method outperforms the state-of-the-art weakly-supervised methods, achieving up to a 16.6% improvement in MAE and up to a 13.8% improvement in RMSE across all evaluation settings. Moreover, the proposed CCTwins obtains competitive counting performance, even when compared to the state-of-the-art fully-supervised methods. IEEE

Is home advantage lost when football matches are played behind closed doors without spectators? Evidence from top European football leagues in the Covid-19 era

When the Covid-19 pandemic reached Europe in mid-March 2020, sport was one of the first activities to be impacted. Precautions taken to limit the spread of the virus resulted in professional football matches being played without spectators. This produced the conditions of a natural experiment enabling the empirical testing of related hypotheses. Using numerous observations from the top European leagues, this study analysed the role of spectators in one of the major phenomena of sports literature - the home advantage i.e., the home team's tendency to win more often than the away team. Strong evidence of the existence of a home advantage both in pre-Covid 19 and Covid-19 periods was found. However, the difference between points earned in favour of the home teams was found to decrease in the Covid period. This was found to be statistically significant when using the Difference-in-Difference (DiD) methodology found in many existing studies. However, alternative analyses 1) using each match as a single observation, rather than adding the away teams in as a control group and 2) taking into account the difference between the performances of the competing teams in previous matches, showed the differences in favour of the home teams with and without spectators to be statistically non-significant. Therefore, it is recommended that in future studies of this kind the most realistic and comprehensive measurement model possible needs to be applied if an accurate picture is to be gained. The conclusion of this study is that, although a decrease in the home advantage was observed when games were played without spectators, it was not sufficient to make a significant difference to that advantage.

A crowd clustering prediction and captioning technique for public health emergencies.

The COVID-19 pandemic has come to the end. People have started to consider how quickly different industries can respond to disasters due to this public health emergency. The most noticeable aspect of the epidemic regarding news text generation and social issues is detecting and identifying abnormal crowd gatherings. We suggest a crowd clustering prediction and captioning technique based on a global neural network to detect and caption these scenes rapidly and effectively. We superimpose two long convolution lines for the residual structure, which may produce a broad sensing region and apply our model's fewer parameters to ensure a wide sensing region, less computation, and increased efficiency of our method. After that, we can travel to the areas where people are congregating. So, to produce news material about the present occurrence, we suggest a double-LSTM model. We train and test our upgraded crowds-gathering model using the ShanghaiTech dataset and assess our captioning model on the MSCOCO dataset. The results of the experiment demonstrate that using our strategy can significantly increase the accuracy of the crowd clustering model, as well as minimize MAE and MSE. Our model can produce competitive results for scene captioning compared to previous approaches.

Oil in the age of coronavirus: Hysteria or appropriate fall?

We look at the oil price fall in the beginning of 2020 and the effects of coronavirus and the attention towards it on these prices. Such a fall was observed at multiple markets simultaneously with the spread of coronavirus and the panic around it, and oil market wasn't an exception. Using OLS time series models, we investigate - what was the main reason behind such a fall - the coronavirus pandemic itself or rather the attention towards it. We prove the absence of straight effects of the COVID-19 itself on oil prices. At the same time we find significant negative impact of the attention towards COVID-19 on the Internet search on the oil prices. We investigate the role of the OPEC in mitigating the negative impact of coronavirus and the attention towards it. We found that after the OPEC summit both the number of Covid cases and the attention towards the disease lost its influence on oil prices. Our paper is relevant for the behavioral finance researchers, as well as for those who look at the influence of informational shocks on different markets and particularly, on the oil market and at the effect of the COVID-19 on the economy.

Interactive Communication Robot Handling Crowd Management and Content Delivery in Museums Employing Crowd Counting

Through the last decade, and particularly after the Covid period (2020 - 2022), crowd counting and localization have attracted much attention of AI researchers due to its potential applicability in crowd monitoring and control, public safety, space design, interactive content delivery etc. Once delivery objectives for a system are envisaged and the premises are fixed, we can always construct manifold technology architecture that delivers the set goals. However, in the Indian context a solution of choice needs to be optimized on frugality and ease of adaptability. In this paper we report an economic and replicable application of crowd counting and interactive content delivery in museums through unbiased knowledge systems embedded in robotic museum assistants. We intend to demonstrate a robotic system that can deliver any gallery content to groups of visitors keeping special focus on the exhibits that are popular choices. Crowd counting is used here to enable the content presentation to a group of choice in an interactive way. There are some market-ready solutions available for interactive gallery demonstration by moveable robots but they require not only huge capital investment but are also of limited use within controlled environments. Our proposed design is to multiplex an existing infrastructure of surveillance system as a smart crowd counting and gallery demonstration system along with crowd management with minimum additional hardware infusion. © 2023 IEEE.

Wisdom of Two Crowds: Misinformation Moderation on Reddit and How to Improve this Process - -A Case Study of COVID-19

Past work has explored various ways for online platforms to leverage crowd wisdom for misinformation detection and moderation. Yet, platforms often relegate governance to their communities, and limited research has been done from the perspective of these communities and their moderators. How is misinformation currently moderated in online communities that are heavily self-governed? What role does the crowd play in this process, and how can this process be improved? In this study, we answer these questions through semi-structured interviews with Reddit moderators. We focus on a case study of COVID-19 misinformation. First, our analysis identifies a general moderation workflow model encompassing various processes participants use for handling COVID-19 misinformation. Further, we show that the moderation workflow revolves around three elements: content facticity, user intent, and perceived harm. Next, our interviews reveal that Reddit moderators rely on two types of crowd wisdom for misinformation detection. Almost all participants are heavily reliant on reports from crowds of ordinary users to identify potential misinformation. A second crowd - participants' own moderation teams and expert moderators of other communities - provide support when participants encounter difficult, ambiguous cases. Finally, we use design probes to better understand how different types of crowd signals - -from ordinary users and moderators - -readily available on Reddit can assist moderators with identifying misinformation. We observe that nearly half of all participants preferred these cues over labels from expert fact-checkers because these cues can help them discern user intent. Additionally, a quarter of the participants distrust professional fact-checkers, raising important concerns about misinformation moderation. © 2023 ACM.

Preferences for online grocery shopping during the COVID-19 pandemic - the role of fear-related attitudes.

In this study, we employ a choice experiment to analyze New York City residents' preferences for online grocery shopping at the beginning of the COVID-19 pandemic. We employ a latent class specification to identify three market segments and estimate consumers' willingness to pay for a variety of attributes of online grocery services related to the quality of the stock, delivery characteristics, and the cost of the online order. We characterize consumers in each segment by their observed characteristics as well as fear-related latent variables. On the one hand, we find that individuals who are actively protecting themselves against COVID-19 have a higher willingness to pay for almost all attributes. On the other hand, consumers who avoid crowds have a lower willingness to pay, but they assign relatively higher importance to no-contact delivery.

From Microscopic Droplets to Macroscopic Crowds: Crossing the Scales in Models of Short-Range Respiratory Disease Transmission, with Application to COVID-19.

Short-range exposure to airborne virus-laden respiratory droplets is an effective transmission route of respiratory diseases, as exemplified by Coronavirus Disease 2019 (COVID-19). In order to assess the risks associated with this pathway in daily-life settings involving tens to hundreds of individuals, the chasm needs to be bridged between fluid dynamical simulations and population-scale epidemiological models. This is achieved by simulating droplet trajectories at the microscale in numerous ambient flows, coarse-graining their results into spatio-temporal maps of viral concentration around the emitter, and coupling these maps to field-data about pedestrian crowds in different scenarios (streets, train stations, markets, queues, and street cafés). At the individual scale, the results highlight the paramount importance of the velocity of the ambient air flow relative to the emitter's motion. This aerodynamic effect, which disperses infectious aerosols, prevails over all other environmental variables. At the crowd's scale, the method yields a ranking of the scenarios by the risks of new infections, dominated by the street cafés and then the outdoor market. While the effect of light winds on the qualitative ranking is fairly marginal, even the most modest air flows dramatically lower the quantitative rates of new infections.

Prediction From Minimal Experience: How People Predict the Duration of an Ongoing Epidemic.

People are known for good predictions in domains they have rich experience with, such as everyday statistics and intuitive physics. But how well can they predict for problems they lack experience with, such as the duration of an ongoing epidemic caused by a new virus? Amid the first wave of COVID-19 in China, we conducted an online diary study, asking each of over 400 participants to predict the remaining duration of the epidemic, once per day for 14 days. Participants' predictions reflected a reasonable use of publicly available information but were meanwhile biased, subject to the influence of negative affect and future time perspectives. Computational modeling revealed that participants neither relied on prior distributions of epidemic durations as in inferring everyday statistics, nor on mechanistic simulations of epidemic dynamics as in computing intuitive physics. Instead, with minimal experience, participants' predictions were best explained by similarity-based generalization of the temporal pattern of epidemic statistics. In two control experiments, we further confirmed that such cognitive algorithm is not specific to the epidemic scenario and that minimal and rich experience do lead to different prediction behaviors for the same observations. We conclude that people generalize patterns in recent history to predict the future under minimal experience.

The one meter rail station Implementing physical distanciation at Paris Nord and North Shenzen railway stations

Since the pandemy, the carrying capacity of major railway stations has been restricted to respect the distance. In Asian metropolises, a tech-driven flow management system reinforces the surveillance of travellers and help the maintenance of a top affluence in station's buildings. In France, an attempt is made to space out travelers with less intrusive processes, such as signage and boarding protocol adaptation. A comparison between these two responses leads us to question what physical distancing does, here and there, to the practices and places of transit and to the economic injonction to densify flows in and around stations until now. Articulating the approaches of crowding science and health regulation in transit environments, the article presents a transcontinental comparison, and then questions the status of major station as post pandemic urbanism showroom.

Social Media Driven Big Data Analysis for Disaster Situation Awareness: A Tutorial

Situational awareness tries to grasp the important events and circumstances in the physical world through sensing, communication, and reasoning. Tracking the evolution of changing situations is an essential part of this awareness and is crucial for providing appropriate resources and help during disasters. Social media, particularly Twitter, is playing an increasing role in this process in recent years. However, extracting intelligence from the available data involves several challenges, including (a) filtering out large amounts of irrelevant data, (b) fusion of heterogeneous data generated by the social media and other sources, and (c) working with partially geo-tagged social media data in order to deduce the needs of the affected people. Spatio-temporal analysis of the data plays a key role in understanding the situation, but is available only sparsely because only a small fraction of people post relevant text and of those very few enable location tracking. In this paper, we provide a comprehensive survey on data analytics to assess situational awareness from social media big data.

Crowd management in public transport to ensure social distancing for prevention of spread of COVID-19

There was global shock from COVID-19 epidemic. Social isolation is becoming more crucial as this delicate condition spreads swiftly. Public transit must be enhanced to stop the spread. This paper proposes an IoT method using LoRa technology that might reduce overcrowding and disease transmission in public buses. In the proposed, buses shall have LoRa transmitters and receivers. It is shown on the bus stop's LCD screen and announced over a speaker if the bus is within range of the receiver. An automatic door mechanism limits the number of people inside the vehicle. In the mobile app, the bus occupancy data is sent to Google Firebase. The app also indicates nearby buses, their occupancy, and their estimated arrival time. In certain cases, authorities may utilise this data to analyse and act. This simple technique would improve bus safety and contain COVID-19.

Crowd intervention strategies considering indoor respiratory exposure risks

In order to explore impacts of crowd intervention strategies on indoor respiratory exposure risks during major pandemics, a variety of crowd motion scenarios were established in general indoor conditions based on improved pedestrian dynamics model and respiratory infection probability model. Then, multi-agent simulation technology was utilized to simulate impacts of strategies, including protection optimization, pedestrian flow optimization and route optimization, on the exposure risks. The results show that indoor respiratory exposure risks are mainly determined by total pedestrian flow, individuals' stay length, movement route planning and duration of stay in contaminated areas. The carryover effect will be formed due to pedestrians' obedience behavior of social distancing, which will further increase exposure time to contaminated areas. The lower pathogen permeability of masks, and the greater space ventilation are, the lower infection probability the crowd will face. © 2022 China Safety Science Journal. All rights reserved.

Design and Implementation of a Real-Time Crowd Monitoring System Based on Public Wi-Fi Infrastructure: A Case Study on the Sri Chiang Mai Smart City

The COVID-19 pandemic has caused significant changes in many aspects of daily life, including learning, working, and communicating. As countries aim to recover their economies, there is an increasing need for smart city solutions, such as crowd monitoring systems, to ensure public safety both during and after the pandemic. This paper presents the design and implementation of a real-time crowd monitoring system using existing public Wi-Fi infrastructure. The proposed system employs a three-tiered architecture, including the sensing domain for data acquisition, the communication domain for data transfer, and the computing domain for data processing, visualization, and analysis. Wi-Fi access points were used as sensors that continuously monitored the crowd and uploaded data to the server. To protect the privacy of the data, encryption algorithms were employed during data transmission. The system was implemented in the Sri Chiang Mai Smart City, where nine Wi-Fi access points were installed in nine different locations along the Mekong River. The system provides real-time crowd density visualizations. Historical data were also collected for the analysis and understanding of urban behaviors. A quantitative evaluation was not feasible due to the uncontrolled environment in public open spaces, but the system was visually evaluated in real-world conditions to assess crowd density, rather than represent the entire population. Overall, the study demonstrates the potential of leveraging existing public Wi-Fi infrastructure for crowd monitoring in uncontrolled, real-world environments. The monitoring system is readily accessible and does not require additional hardware investment or maintenance. The collected dataset is also available for download. In addition to COVID-19 pandemic management, this technology can also assist government policymakers in optimizing the use of public space and urban planning. Real-time crowd density data provided by the system can assist route planners or recommend points of interest, while information on the popularity of tourist destinations enables targeted marketing.

Covid-19 crowd detection and alert system using image processing

In this paper, we aim to help in identifying the people that are violating social distancing norms set by the government (necessary during the COVID-19 pandemic in public places), by providing an efficient real-time deep learning-based framework to automate the process of monitoring the social distancing via object detection and tracking approaches. Our system is divided into two subsystems: one that deals with crowd detection and control, and the other that sends information to the police authorities. Our system technologies, including as IoT, image processing, web cams, BLE, OpenCV, and Cloud, are being considered for inclusion in the proposed framework. The image processing is divided into two sections, the first of which is the extraction of frames from real-time movies, and the second of which is the processing of the frame to determine the number of individuals in the crowd. Even in a crowd, dissemination may be restricted if people adhere to social distancing standards. As a result, the image processing model primarily targets the number of people who do not adhere to social distancing norms and stand too close together. © 2023 IEEE.

Patient–Robot Co-Navigation of Crowded Hospital Environments

Intelligent multi-purpose robotic assistants have the potential to assist nurses with a variety of non-critical tasks, such as object fetching, disinfecting areas, or supporting patient care. This paper focuses on enabling a multi-purpose robot to guide patients while walking. The proposed robotic framework aims at enabling a robot to learn how to navigate a crowded hospital environment while maintaining contact with the patient. Two deep reinforcement learning models are developed;the first model considers only dynamic obstacles (e.g., humans), while the second model considers static and dynamic obstacles in the environment. The models output the robot's velocity based on the following inputs;the patient's gait velocity, which is computed based on a leg detection method, spatial and temporal information from the environment, the humans in the scene, and the robot. The proposed models demonstrate promising results. Finally, the model that considers both static and dynamic obstacles is successfully deployed in the Gazebo simulation environment.

Estimating the effect of attendance on home advantage in the National Basketball Association

Using COVID-19 safety protocols as a natural experiment, we are able to delineate three distinct attendance categories in the NBA: 1) unrestricted games played prior to the pandemic, 2) attendance-restricted games played with socially distanced fans, and 3) ‘ghost games' played without fans. Further, since attendance at restricted games was exogenously determined by local COVID-19 protocols that were in turn driven by changes in COVID-19 case counts, we are able to estimate whether the ‘marginal fan' contributes to home advantage. Taken together, our results indicate that the presence of fans matters to home team performance;in fact, ‘ghost games' eliminated home advantage in totality. With a relatively small number of socially distanced fans, however, the entirety of home advantage was retained. Interestingly, since the size of socially distanced crowds had a statistically insignificant impact on home advantage, we find no evidence of a ‘marginal fan' effect. Finally, since researchers have found that officiating is influenced by fans in international soccer (e.g. Anders and Rotthoff, 2014), we explore whether NBA officiating behaviour was altered due to changes in attendance conditions. Our results indicate that NBA officials were not measurably influenced by the presence or quantity of fans.

Emerging data inputs for infectious diseases surveillance and decision making.

Infectious diseases create a significant health and social burden globally and can lead to outbreaks and epidemics. Timely surveillance for infectious diseases is required to inform both short and long term public responses and health policies. Novel data inputs for infectious disease surveillance and public health decision making are emerging, accelerated by the COVID-19 pandemic. These include the use of technology-enabled physiological measurements, crowd sourcing, field experiments, and artificial intelligence (AI). These technologies may provide benefits in relation to improved timeliness and reduced resource requirements in comparison to traditional methods. In this review paper, we describe current and emerging data inputs being used for infectious disease surveillance and summarize key benefits and limitations.

Crowd work in STEM‐related fields: A window of opportunity from a gender perspective?

After the Global Financial Crisis (2008) many people found new job opportunities on crowd platforms. The COVID‐19 crisis reinforced this trend and virtual work is expected to increase. Although the working conditions of individuals engaged on these platforms is an emerging topic, of research, the existing literature tends to overlook the gendered dimension of the gig economy. Following a quantitative approach, based on the statistical analysis of 444 profiles (platform Freelancer.com in Spain and Argentina), we examine the extent to which the gig economy reproduces gender inequalities such as the underrepresentation of women in STEM‐related tasks and the gender pay gap. While the findings reveal lower participation of women than men, this gap is not higher in Argentina than in Spain. Moreover, gender variations in hourly wages are not as marked as expected, and such differences disappear once STEM skill levels are controlled for. Asymmetry in individuals' STEM skill level provides a better explanation than gender of the hourly wage differences. This finding opens a window of opportunity to mitigate the classical gender discrimination that women face in technological fields in traditional labor markets. Finally, the paper identifies some issues concerning the methodological bias entailed by the use of an application programming interface in cyber‐research, when analyzing gender inequalities.