Individual Crowd Research Articles

Quantifying crowding perception at large events using beacons and surveys

Given the increasing urban population and frenetic mobility, understanding how individuals perceive crowding at large-scale events is crucial for effective crowd management and safety. This study focuses on Tokyo Big Sight in Japan exhibitions to examine participants’ perceptions of peak crowding times, locations, and local density, and compare them with the actual measurements. Our methodology integrated questionnaires with beacon tag data. The results showed that participants’ perceptions of crowded times and locations aligned well with the actual (measured) crowding data, demonstrating that people can pick crowded moments and locations with better accuracy. However, when asked to select images that closely reflect crowding density conditions within the facility, the association was mostly relative (i.e., context-depending) and lacked absolute accuracy. Furthermore, perceptions of time tend to be biased towards exit times. This study underscores the necessity for event organizers and urban planners to account for the subjective and relative nature of individual crowding experiences, thereby emphasizing the need for adaptive management strategies that consider personal perceptions.

How behavioural changes in social groups affect evacuation efficiency of crowds

Consider an incident of acute emergency—such as a violent attack—necessitating swift evacuation of a crowd primarily made up of social groups. Is it possible to identify simple strategies that the groups can implement to facilitate collective movement efficiency? There has been some research, though not extensive, aimed at understanding unique evacuation dynamics of crowds of social groups, but mostly from a descriptive or observational perspective. Here, we introduce and explore a prescriptive or interventional approach. The aim is to establish effective behavioural interventions. Experiments are conducted with crowds of solo individuals as well as crowds composed of groups of three or groups of five. The results demonstrate a clear decrease in evacuation efficiency as crowd composition shifts from solo to groups, with a more pronounced effect observed with larger groups. We demonstrate how this inefficiency is attributable to the larger spatial footprint associated with groups. The findings also reveal that instructing groups to hold hands does not enhance evacuation efficiency. However, instructing them to form a line (i.e., platooning) leads to significant improvements, reducing evacuation times to levels comparable to crowds of solo individuals. These findings suggest that simple and easily communicable behavioural modifications for social groups can have substantial impacts on their evacuation efficiency. Making such strategies common knowledge has the potential of helping more people survive acute crises, an important aspect of collective resilience.

Crowdsourcing Affective Annotations Via fNIRS-BCI

Affective annotation refers to the process of labeling media content based on the emotions they evoke. Since such experiences are inherently subjective and depend on individual differences, the central challenge is associating digital content with its affective, interindividual experience. Here, we present a first-of-its-kind methodology for affective annotation directly from brain signals by monitoring the affective experience of a crowd of individuals via functional near-infrared spectroscopy (fNIRS). An experiment is reported in which fNIRS was recorded from 31 participants to develop a brain-computer interface (BCI) for affective annotation. Brain signals evoked by images were used to draw predictions about the affective dimensions that characterize the stimuli. By combining annotations, the results show that monitoring crowd responses can draw accurate affective annotations, with performance improving significantly with increases in crowd size. Our methodology demonstrates a proof-of-concept to source affective annotations from a crowd of BCI users without requiring any auxiliary mental or physical interaction.

Fish evacuate smoothly respecting a social bubble

Crowd movements are observed among different species and on different scales, from insects to mammals, as well as in non-cognitive systems, such as motile cells. When forced to escape through a narrow opening, most terrestrial animals behave like granular materials and clogging events decrease the efficiency of the evacuation. Here, we explore the evacuation behavior of macroscopic, aquatic agents, neon fish, and challenge their gregarious behavior by forcing the school through a constricted passage. Using a statistical analysis method developed for granular matter and applied to crowd evacuation, our results clearly show that, unlike crowds of people or herds of sheep, no clogging occurs at the bottleneck. The fish do not collide and wait for a minimum waiting time between two successive exits, while respecting a social distance. When the constriction becomes similar to or smaller than their social distance, the individual domains defined by this cognitive distance are deformed and fish density increases. We show that the current of escaping fish behaves like a set of deformable 2D-bubbles, their 2D domain, passing through a constriction. Schools of fish show that, by respecting social rules, a crowd of individuals can evacuate without clogging, even in an emergency situation.

Multi-objective structural profile optimization of ships based on improved Artificial Bee Colony Algorithm and structural component library

This paper addresses the multi-objective, design variables discrete problem of river boat profile structure optimization. The study considers not only the economic performance but also the stability of the boat to a certain extent when selecting the optimization objectives. First, the standard component library of the vessels is established and the discrete structural model is converted into a continuous optimization model based on the component library index. Second, to balance the global search and local search capability of the optimization algorithm, an adaptive Artificial Bee Colony Algorithm (I-ABC) is proposed. I-ABC uses a modified chaotic mapping to improve the diversity of nectar sources, and adaptive search and following mechanisms are adopted for the leading and following bee stages, respectively. Then, a Dominating Artificial Bee Colony Algorithm (D-ABC) is proposed for multi-objective optimization by introducing Pareto domination and considering the degree of individual crowding in terms of hypersphere neighborhood. Then, numerical simulations and comparative analysis of the proposed algorithm are performed using different test functions, and the results show that I-ABC and D-ABC have excellent performance in both single-objective and multi-objective optimization. Finally, using the component library established in this paper and the proposed improved Bee Colony Algorithm to optimize the profile structure of a chemical-liquid tanker, the optimization results show that the maximum reduction of the ship profile area is 5.08% and the maximum reduction of the vertical center height is 7.427%. The proposed optimization algorithm is also applicable to other river boats.

A Complexity-based Framework for Social Product Development

The development of new products using social technologies among defined crowds of individuals or groups, commonly known as Social Product Development (SPD), has become a growing trend in engineering design and manufacturing literature. This paper proposes an approach to studying the actors involved in SPD from a complexity science point of view. Specifically, we suggest that organizational efforts towards SPD must be concentrated on human and technological elements to achieve success in New Product Development (NPD). The paper explores the differences and similarities between the creation of new products and SPD and proposes a differentiated approach, based on complexity science for the different stages of NPD and SPD. Our findings show that SPD requires the integration of the complexity approach due to the combination of the knowledge, skills, and abilities of those involved during the development of new products. In addition, it is recommended that engineering design managers consider the importance of leadership styles, work structure, and organizational culture to achieve the expected outcomes of SPD.

The motor Wisdom of the Crowd.

Wisdom of the Crowd is the aggregation of many individual estimates to obtain a better collective one. Because of its enormous social potential, this effect has been thoroughly investigated, but predominantly on tasks that involve rational thinking (such as estimating a number). Here we tested this effect in the context of drawing geometrical shapes, which still enacts cognitive processes but mainly involves visuomotor control. We asked more than 700 school students to trace five patterns shown on a touchscreen and then aggregated their individual trajectories to improve the match with the original pattern. Our results show the characteristics of the strongest examples of Wisdom of the Crowd. First, the aggregate trajectory can be up to 5 times more accurate than the individual ones. Second, this great improvement requires aggregating trajectories from different individuals (rather than trials from the same individual). Third, the aggregate trajectory outperforms more than 99% of individual trajectories. Fourth, while older individuals outperform younger ones, a crowd of young individuals outperforms the average older one. These results demonstrate for the first time Wisdom of the Crowd in the realm of motor control, opening the door to further studies of human and also animal behavioural trajectories and their mechanistic underpinnings.

Task design in complex crowdsourcing experiments: Item assignment optimization

Crowdsourcing is the practice of outsourcing a task that would otherwise be performed by one or a few experts to a crowd of individuals. It is often used to collect large amounts of manually created labels that form datasets for training and evaluating supervised machine learning models. When designing a (micro-task) crowdsourcing experiment, it is often desirable to combine different types of items (e.g., images of both cats and dogs for an image classification task) together into the same task and to have diverse contributors (e.g., male and female crowd workers) providing labels for the same item so as to mitigating existing bias in the collected labels. Setting up such a crowdsourcing experiment, i.e., constructing a human intelligence task containing the right mix of items and assigning it to the right mix of crowd workers, becomes a complex problem.In this paper we address the problem of optimizing human intelligence task construction by providing its formal definition and by applying a local search method initialized by a greedy construction to solve it. We experimentally show the flexibility of the proposed solution in addressing different type of constraints, both on the item and on the worker side, and demonstrate its scalability to complex crowdsourcing setups. The proposed solution enables the design of tools for crowdsourcing requesters to set up complex task deployments in platforms like Amazon MTurk or Prolific.

Taking a Disagreeing Perspective Improves the Accuracy of People’s Quantitative Estimates

Many decisions rest on people’s ability to make estimates of unknown quantities. In these judgments, the aggregate estimate of a crowd of individuals is often more accurate than most individual estimates. Remarkably, similar principles apply when multiple estimates from the same person are aggregated, and a key challenge is to identify strategies that improve the accuracy of people’s aggregate estimates. Here, we present the following strategy: Combine people’s first estimate with their second estimate, made from the perspective of someone they often disagree with. In five preregistered experiments (N = 6,425 adults; N = 53,086 estimates) with populations from the United States and United Kingdom, we found that such a strategy produced accurate estimates (compared with situations in which people made a second guess or when second estimates were made from the perspective of someone they often agree with). These results suggest that disagreement, often highlighted for its negative impact, is a powerful tool in producing accurate judgments.

A Convex Formulation for Learning from Crowds

Recently crowdsourcing services are often used to collect a large amount of labeled data for machine learning, since they provide us an easy way to get labels at very low cost and in a short period. The use of crowdsourcing has introduced a new challenge in machine learning, that is, coping with the variable quality of crowd-generated data. Although there have been many recent attempts to address the quality problem of multiple workers, only a few of the existing methods consider the problem of learning classifiers directly from such noisy data. All these methods modeled the true labels as latent variables, which resulted in non-convex optimization problems. In this paper, we propose a convex optimization formulation for learning from crowds without estimating the true labels by introducing personal models of the individual crowd workers. We also devise an efficient iterative method for solving the convex optimization problems by exploiting conditional independence structures in multiple classifiers. We evaluate the proposed method against three competing methods on synthetic data sets and a real crowdsourced data set and demonstrate that the proposed method outperforms the other three methods.

Fundraising activities and digitalization: defining risk indicators for evaluating equity crowdfunding campaigns

PurposeThe wide development of digital platforms permitted the birth of new financing modalities, namely, crowdfunding, where the crowd of individuals and investors can supply the necessary financial resources for venture creation and growth. While the extant literature has focused on analyzing the dynamics and features of crowdfunding campaigns, few studies have focused on understanding how crowd investors decide which ventures to invest in and which factors influence their decision-making process. Due to this gap, the purpose of this paper is to analyze the factors influencing the choice to invest in an equity crowdfunding campaign, by defining a set of indicators useful to evaluate the risk of the campaign.Design/methodology/approachAn empirical research study of Italian equity crowdfunding campaigns has been conducted to identify quantitative indicators useful for evaluating the risk in a crowdfunding campaign.FindingsFindings demonstrate that the risk indicators proposed to represent important gauges that investors can usefully consider ex ante to assess the degree of riskiness of the investment in the equity crowdfunding campaign.Research limitations/implicationsThe limitations of the study regarding the size of the sample that is small due to the necessity to extract enough information in pre and post-equity campaigns. Also, the lack of historical data is another limitation.Originality/valueThe originality of the studies relies on the proposal of quantitative indicators for the evaluation of the risk in equity crowdfunding campaigns for “crowd” investors to reduce information asymmetries.

A modular hierarchical array camera

Array cameras removed the optical limitations of a single camera and paved the way for high-performance imaging via the combination of micro-cameras and computation to fuse multiple aperture images. However, existing solutions use dense arrays of cameras that require laborious calibration and lack flexibility and practicality. Inspired by the cognition function principle of the human brain, we develop an unstructured array camera system that adopts a hierarchical modular design with multiscale hybrid cameras composing different modules. Intelligent computations are designed to collaboratively operate along both intra- and intermodule pathways. This system can adaptively allocate imagery resources to dramatically reduce the hardware cost and possesses unprecedented flexibility, robustness, and versatility. Large scenes of real-world data were acquired to perform human-centric studies for the assessment of human behaviours at the individual level and crowd behaviours at the population level requiring high-resolution long-term monitoring of dynamic wide-area scenes.

Tracking multiple perspectives: Spontaneous computation of what individuals in high entitative groups see.

Perspective-taking ability is crucial for supporting social interactions. It has been widely suggested that the calculation of an individual's perspective is spontaneous. Nevertheless, people typically engage with more than one individual, and computing what individuals in a crowd see is important. The current study explored whether people spontaneously compute the perspectives of individuals displayed in a crowd. The classic visual perspective-taking task was adopted, but the picture of the room was presented with four human avatars facing two walls. The results showed that if the crowd of individuals was treated as a high entitative group, when none of the perspectives of the individuals contained the same number of discs as that from the perspective of the participant, the judgment of the participant's perspective was slower than when a proportion of the perspectives of the individuals displayed in the crowd were consistent with the participant's perspective, even if the perspectives of the multiple individuals in a crowd were not explicitly noticed. This altercentric intrusion effect was not present when the crowd had low entitativity. These findings were replicated by using different methods to operationalize group entitativity. Hence, this study demonstrates that spontaneously tracking the perspectives of individuals displayed in a crowd has a boundary condition and that people can spontaneously compute what individuals in high entitative groups see.

Investors’ Behavior in Reward Based Crowdfunding: Do Followers Actually Follow?

In reward-based crowdfunding, would-be entrepreneurs raise consumer- based finance for their prototypes from a crowd of small individuals or backers. This is a type of financing characterized by high informational asymmetries and sequential financing where backers (investors that provide financing) can observe the decisions of other backers before pledging their money. Such a setting allows testing the dynamics of informational cascades and herding behavior by investors in crowdfunding. This paper addresses this issue by investigating how backers make their funding decisions and to what extent they rely on the decisions made by early backers. We use a panel data from 3,923 projects crawled from Kickstarter to analyze this issue. We find evidence that followers make use of the information provided by early backers’ funding decisions, but they also engage in active observational learning over the quality of the project and the entrepreneur to complement the previous information. In terms of managerial implications, our findings indicate that entrepreneurs will be more successful in project financing if they target their products to the demands of early backers, rather than to the tastes of the median consumer/backer. This strategy of focusing on early backers allows creating a positive informational cascade ending in a successful crowdfunding.

Experimental study on pedestrian behavior in a mixed crowd of individuals and groups

The influences of groups on pedestrian behavior in a mixed crowd that comprises individuals and groups were investigated experimentally. The crowd comprised 20 pedestrians: some were composed into groups of three or four pedestrians (3-pedestrian and 4-pedestrian groups), and the others walked as individuals The walking speed and the walking behavior of pedestrians in the crowds were the main focus. Results showed that the walking speed of pedestrians – the​ individuals and group members in a mixed crowd – basically decreased compared with their walking speed in none-group crowd, and the decreasing of pedestrians’ walking speed became more obvious with the size of groups inside the mixed crowd. Basically, the decrease in the degree of walking speed was larger for female pedestrians than for the male pedestrians. However, in the mixed crowd conditions with the 3-pedestrian group, the walking speed of individuals with an initial location near the group might possibly be higher than that in the none-group crowd condition, which might be caused by the accelerating process for overtaking the group in front of them. This speed accelerating behavior was more remarkable among the male pedestrians in this study. In the mixed crowd conditions with the 4-pedestrian group, the overtaking of the group became more difficult for the individuals. The walking speed of a group in a crowd was lower than that in the free-walking environment because the group members had to pay more attention to maintain group relationships and not separate from each other in the crowd.

Enabling Strong Privacy Preservation and Accurate Task Allocation for Mobile Crowdsensing

Mobile crowdsensing engages a crowd of individuals to use their mobile devices to cooperatively collect data about social events and phenomena for special interest customers. It can reduce the cost on sensor deployment and improve data quality with human intelligence. To enhance data trustworthiness, it is critical for service provider to recruit mobile users based on their personal features, e.g., mobility pattern and reputation, but it leads to the privacy leakage of mobile users. Therefore, how to resolve the contradiction between user privacy and task allocation is challenging in mobile crowdsensing. In this paper, we propose SPOON, a strong privacy-preserving mobile crowdsensing scheme supporting accurate task allocation from geographic information and credit points of mobile users. In SPOON, the service provider enables to recruit mobile users based on their locations, and select proper sensing reports according to their trust levels without invading user privacy. By utilizing proxy re-encryption and BBS+ signature, sensing tasks are protected and reports are anonymized to prevent privacy leakage. In addition, a privacy-preserving credit management mechanism is introduced to achieve decentralized trust management and secure credit proof for mobile users. Finally, we show the security properties of SPOON and demonstrate its efficiency on computation and communication.

Providing Task Allocation and Secure Deduplication for Mobile Crowdsensing via Fog Computing

Mobile crowdsensing enables a crowd of individuals to cooperatively collect data for special interest customers using their mobile devices. The success of mobile crowdsensing largely depends on the participating mobile users. The broader participation, the more sensing data are collected; nevertheless, the more replicate data may be generated, thereby bringing unnecessary heavy communication overhead. Hence it is critical to eliminate duplicate data to improve communication efficiency, a.k.a., data deduplication. Unfortunately, sensing data is usually protected, making its deduplication challenging. In this paper, we propose a fog-assisted mobile crowdsensing framework, enabling fog nodes to allocate tasks based on user mobility for improving the accuracy of task assignment. Further, a fog-assisted secure data deduplication scheme (Fo-SDD) is introduced to improve communication efficiency while guaranteeing data confidentiality. Specifically, a BLS-oblivious pseudo-random function is designed to enable fog nodes to detect and remove replicate data in sensing reports without exposing the content of reports. To protect the privacy of mobile users, we further extend the Fo-SDD to hide users’ identities during data collection. In doing so, Chameleon hash function is leveraged to achieve contribution claim and reward retrieval for anonymous mobile users. Finally, we demonstrate that both schemes achieve secure, efficient data deduplication.

Finite-Time Influence Systems and the Wisdom of Crowd Effect

Recent contributions have studied how an influence system may affect the wisdom of crowd phenomenon. In the so-called naive learning setting, a crowd of individuals holds opinions that are statisti...

Reducing Uncertainty of Schema Matching via Crowdsourcing with Accuracy Rates

Schema matching is a central challenge for data integration systems. Inspired by the popularity and the success of crowdsourcing platforms, we explore the use of crowdsourcing to reduce the uncertainty of schema matching. Since crowdsourcing platforms are most effective for simple questions, we assume that each Correspondence Correctness Question (CCQ) asks the crowd to decide whether a given correspondence should exist in the correct matching. Furthermore, members of a crowd may sometimes return incorrect answers with different probabilities. Accuracy rates of individual crowd workers can be attributes of CCQs as well as evaluations of individual workers. We prove that uncertainty reduction equals to entropy of answers minus entropy of crowds and show how to obtain lower and upper bounds for it. We propose frameworks and efficient algorithms to dynamically manage the CCQs to maximize the uncertainty reduction within a limited budget of questions. We develop two novel approaches, namely “Single CCQ” and “Multiple CCQ”, which adaptively select, publish, and manage questions. We verify the value of our solutions with simulation and real implementation.

Towards understanding socio-cognitive behaviors of crowds from visual surveillance data

The problem of understanding socio-cognitive aspects of crowd behavior is a challenging yet critical task particularly for human-computer interaction applications. This issue is considered an important component of both current surveillance systems and futuristic interactions between intelligent agents and crowds of humans. In this paper, a probabilistic formulation of different categories of socio-cognitive crowd behavior is proposed. This framework models in a hierarchical manner relationships between the movements of entities within the crowd to differentiate between various crowd behaviors. Functionally, the framework can be considered a mid-level layer between purely visual analysis (i.e. detection and tracking of crowd components) and detailed semantics of crowd behaviors. In the presented study, the proposed framework extends a Gaussian Mixture Model of Dynamic Textures detection (GMM-of-DT) technique using Kalman filtering for the hierarchical motion representation of the crowds, simultaneously at micro (individual crowd members) and macro (groups of individuals) levels. Thereon, socio-cognitive crowd behaviors are categorized into individual, group, leader-follower and social interaction types. The proposed approach is validated on exemplary sequences from a benchmark PETS dataset (and, preliminarily, on other publicly available datasets), where the actual detection/tracking results are employed to evaluate probabilities of various socio-cognitive behaviors, and to compare the outputs of the proposed models to manually annotated ground-truth data.