Crowdsourced Data Collection Research Articles

Innovative Data Collection Methods for Advancing Coastal Management in Liberia

Effective coastal management is crucial for the sustainability of ecosystems and the livelihoods of communities that depend on coastal resources. In Liberia, current challenges such as outdated data collection methods, limited technological resources, and inadequate community engagement significantly hinder effective decision-making and coastal management efforts. This research explores innovative data collection methods, including remote sensing, crowdsourced data collection, swarm drones, and bioacoustic monitoring as potential solutions to these challenges. By integrating these technologies, Liberia can enhance data accuracy and cost-effectiveness while empowering local communities to actively participate in sustainable practices. The adoption of these methods will not only improve the management of coastal resources but also foster resilience against climate change and environmental degradation, paving the way for a sustainable future for Liberia’s coastal regions

Review of strategies to overcome the lack of data in landmine detection

This article reviews strategies to address the lack of data for training landmine recognition models. Since the outbreak of the war in Ukraine in 2014, the area of contaminated territories has gradually increased. However, after Russia's full-scale invasion on February 24, 2022, the problem of landmines in Ukraine has become much more severe, as the area of mined territories in the country has increased to 30%. It takes many years and efforts to clear such a large area. To overcome a problem of this scale, it is necessary to look for new methods of landmine detection that will allow demining to be conducted 24/7. Machine learning is one of the options for solving the problem of landmine detection. To train landmine recognition models, a large amount of data is required. However, the lack of diverse and large datasets creates significant obstacles to the development of effective detection systems. The safety concerns associated with conducting experiments with real landmines further exacerbate the problem. This article discusses three possible strategies to overcome the above problem: augmentation methods, the use of 3D printing technology, and crowdsourced data collection. Augmentation methods offer data generation to improve model performance. 3D printing allows for the creation of realistic replicas of landmines for safe experimentation. Crowdsourcing uses collective efforts to collect real-world data from conflict zones. Through the joint efforts of researchers, technology developers and humanitarian organizations, these approaches offer promising ways to improve landmine detection capabilities. The use of these approaches can address the data gap and ensure safe data collection.

Crowdsourced Data Collection Opens New Avenues for the Behavioral Sciences to Impact Real-World Applications

The behavioral sciences have had great success in their study of the mechanisms that drive behavior. However, they have had less impact on applied settings or policy. This gap results from the very adaptability that makes human behavior useful. Adaptability implies that behavior will be highly specific to the context in which it occurs. Thus, building a bridge between the lab and application requires testing in the specific applied setting, which runs afoul of the high cost of data collection. This cost has also led to a focus on simple paradigms that poorly match applied settings. However, crowdsourcing enables data collection at vastly reduced budgets and schedules. This new cost regime also enables paradigms better suited to applied settings. Behavioral science should now be used throughout applied- and policy-focused projects.

Methodological considerations for behavioral studies relying on response time outcomes through online crowdsourcing platforms

This perspective paper explores challenges associated with online crowdsourced data collection, particularly focusing on longitudinal tasks with time-sensitive outcomes like response latencies. Based on our research, we identify two significant sources of bias: technical shortcomings such as low, variable frame rates, and human factors, contributing to high attrition rates. We explored potential solutions to these problems, such as enforcing hardware acceleration and defining study-specific frame rate thresholds, as well as pre-screening participants and monitoring hardware performance and task engagement over each experimental session. With this discussion, we intend to provide recommendations on how to improve the quality and reliability of data collected via online crowdsourced platforms and emphasize the need for researchers to be cognizant of potential pitfalls in online research.

Reel reflections: the role of entertainment media narratives in coping among young adult cancer survivors

Introduction Young adult cancer survivors (YACS; ages 18–39) report a significant psychological burden. Entertainment media narratives (e.g., books, movies, shows that are produced for mass consumption) might be an effective tool for reducing this distress, although little is known about present use among YACS. Method YACS completed a survey about their use of entertainment media narratives to cope with cancer using an adapted version of the Brief COPE. Additionally, YACS reported their use of entertainment media narratives to start conversations about their experience with others, and they described features of entertainment media narratives that they found helpful in coping. Results We recruited 108 YACS from three recruitment sites. Most participants were White (n = 65), female (n = 54), and recruited from Prolific (n = 56), an Internet-based, crowdsourced data collection platform. Participants were, on average, 30 years old and 45 months from the completion of their primary cancer treatment. YACS who used entertainment media to cope with cancer (n = 32; 29.6%) were significantly younger and significantly closer to the end of their primary treatment. Compared to nonusers, users of entertainment media narratives to cope were also more likely to identify as Black; identifying as Black was associated with a 2.05-factor increase in using narrative entertainment media to cope with cancer even when controlling for other demographic differences. Additionally, compared to their peers, Black YACS reported greater use of narratives to start cancer-related conversations. Emotional and inspirational storylines were the most helpful story features. Discussion Some YACS, especially Black YACS, use stories to cope with their cancer experience. YACS patients could find stories useful in exploring their cancer-related emotions, although the exact benefits are still unknown.

EDIR: Efficient Distributed Image Retrieval of Novel Objects in Mobile Networks

Crowdsourcing data collection from a network of mobile devices is useful in various applications. Mobile devices store a large amount of visual data that can aid in different application scenarios. Trained Convolutional Neural Networks (CNNs) can be deployed on mobile devices to be used in searching for objects of interest. Querying for novel objects, for which models have not been trained yet, presents some unique challenges. When novel objects are queried, new models must be trained and distributed to all edge devices. In this paper, we propose an efficient method and a system, called EDIR, which enables answering these queries while taking into account the bandwidth limitations encountered in wireless networks, as well as the limited energy and computational power on mobile devices. Through extensive experimentation, we show that using distance-based classifiers, specifically those relying on the Cosine distance, leads to more efficient utilization of network resources by reducing the number of false positives. We perform analysis that enables the requester to tune the parameters of interest before issuing the query, and validate our theoretical results. EDIR reduces the amount of transferred data by more than 45% compared to other approaches while simultaneously achieving a good F1 score.

REVIEW OF GLOBAL PRACTICE OF USING CELL PHONE DATA TO MEASURE TRAVELLER DATA

JSC “Ukrzaliznytsia” is a national integrated railway company and the largest employer in the country, playing a crucial role in the Ukrainian economy and the labour market. The rail companies faces significant risks, mainly due to prolonged disruptions due to the ongoing armed conflict on Ukrainian territory [1]. In such conditions, the question of the need for a comprehensive solution to the problems of the functioning of passenger railway transport, where the number of travellers serves as a pivotal indicator of the demand for transportation services, offering essential information for planning and evaluations. Advancements in technology have enabled the collection of crowd-sourced traffic data [2]. Crowdsourced smartphone data can collect information about user locations, travel paths, route choices, travel times, and speeds. This analysis explores how the number of passengers in trains is measured, examining technologies and methods that use cell phone data globally to assess their potential application in Ukraine.

Examining the validity of the addictions neuroclinical assessment domains in a crowdsourced sample of adults with current alcohol use.

Several dimensional frameworks for characterizing heterogeneity in alcohol use disorder (AUD) have been proposed, including the Addictions Neuroclinical Assessment (ANA). The ANA is a framework for assessing individual variability within AUD across three domains corresponding to the proposed stages of the addiction cycle: reward (binge-intoxication stage), negative emotionality (withdrawal-negative affect stage), and cognitive control (preoccupation-anticipation stage). Recent work has evaluated the ANA's three-factor structure and construct validity, primarily in treatment-seekers with AUD. We extended this research by examining the factor structure, bias across alcohol use severity, longitudinal invariance, and concurrent and predictive validity of a novel assessment of the ANA domains in adults with past 12-month regular (10 + alcohol units/week) alcohol use. Participants recruited from Prolific (N = 732), a crowdsourced data collection platform, completed various self-report measures. A test-retest subsample (n = 234) completed these measures 30 days later. Split-half exploratory factor analysis and confirmatory factor analysis supported the three-factor structure of the ANA. The overall factor structure was invariant across 30 days. Concurrently and prospectively, ANA domains demonstrated convergent validity concerning theoretically aligned alcohol-related, psychological, and personality measures. However, there was evidence of poor discriminant validity, and several cognitive control and reward items demonstrated bias across alcohol use severity. Future research is needed to improve the measurement of ANA domains using multimodal indicators, examine longitudinal changes in domains and their relationship with alcohol use severity, characterize phenotypic subgroups based on relative levels of domains, and compare the utility of the ANA with other proposed frameworks for measuring AUD heterogeneity. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Crowd-Sourced Buildings Data Collection and Remote Training: New Opportunities to Engage Students in Seismic Risk Reduction

Crowd-Sourced Buildings Data Collection and Remote Training: New Opportunities to Engage Students in Seismic Risk Reduction

PACMAN: A Framework for Pulse Oximeter Digit Detection and Reading in a Low-Resource Setting

In light of the COVID-19 pandemic, patients were required to manually input their daily oxygen saturation (SpO2) and pulse rate (PR) values into a health monitoring system—unfortunately, such a process trend to be an error in typing. Several studies attempted to detect the physiological value from the captured image using optical character recognition (OCR). However, the technology has limited availability with high cost. Thus, this study aimed to propose a novel framework called PACMAN (Pandemic Accelerated Human-Machine Collaboration) with a low-resource deep learning-based computer vision. We compared state-of-the-art object detection algorithms (scaled YOLOv4, YOLOv5, and YOLOR), including the commercial OCR tools for digit recognition on the captured images from the pulse oximeter display. All images were derived from crowdsourced data collection with varying quality and alignment. YOLOv5 was the best-performing model against the given model comparison across all datasets, notably the correctly orientated image dataset. We further improved the model performance with the digits auto-orientation algorithm and applied a clustering algorithm to extract SpO2 and PR values. The accuracy performance of YOLOv5 with the implementations was approximately 81.0-89.5%, which was enhanced compared to without any additional implementation. Accordingly, this study highlighted the completion of the PACMAN framework to detect and read digits in real-world datasets. The proposed framework has been currently integrated into the patient monitoring system utilized by hospitals nationwide.

A Psychosocial Care Model for Families Affected by Congenital Heart Disease.

To develop a model of family-based psychosocial care for congenital heart disease (CHD). Qualitative study using crowdsourced data collected from parents of young children with CHD who received care across 42 hospitals. Yammer, a social networking platform used to facilitate online crowdsourcing and qualitative data collection. Geographically diverse sample of 100 parents (72 mothers and 28 fathers) of young children with CHD. None. Parents joined a private group on Yammer and responded to 37 open-ended study questions over a 6-month period. Qualitative data were coded and analyzed using an iterative process. Three broad themes corresponding to pillars of family-based psychosocial care were identified: pillar 1) parent partnership in family-integrated medical care, pillar 2) supportive interactions focused on parent and family wellbeing, and pillar 3) integrated psychosocial care and peer support for parents and families. Each pillar was supported by subthemes corresponding to specific intervention strategies. Most parents described the need for intervention strategies across multiple pillars, with almost half reporting needs across all three pillars of psychosocial care. Parents' preferences for psychosocial support changed over time with changes to their child's medical status and across care settings (e.g., hospital, outpatient clinic). Results support a model of family-based psychosocial care that is multidimensional and flexible to meet the needs of families affected by CHD. All members of the healthcare team play an important role in providing psychosocial support. Future research incorporating components of implementation science is needed to promote uptake of these findings, with the goal of optimizing family-based psychosocial support in the hospital setting and beyond.

Validation of the multiple disability multidimensional attitudes scale toward persons with disabilities.

This study investigated the psychometric properties of a Multiple Disability Multidimensional Attitudes Scale Toward Persons with Disabilities (MD-MAS), focusing on four types of disabilities: anxiety disorder, autism spectrum disorder (ASD), blindness, and schizophrenia. We developed new vignettes describing the situation of interacting with an individual with each type of disability. We recruited 991 participants from a crowdsourcing data collection tool (i.e., Prolific). Participants were randomly assigned to one of four online surveys depending on the disability type. Five MAS models were selected from the previous literature to conduct confirmatory factor analyses (CFA). CFA supported the German model of MAS with a four-factor structure (i.e., calm, negative affect, positive cognition, behavioral avoidance) as a good fit for the MD-MAS for four disability types. High internal consistency was found for four subscales across the disability types. This study modified the original MAS to assess attitudes toward persons with different types of disabilities. The adequate reliability and fit of the factor structure of the MD-MAS same across the four disability types allow researchers to compare attitudes based on disability types. This will provide significant implications for research and practice in understanding the nature of different attitudes by disability types. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

What people learn from death: exploring citizen scientists’ learning outcomes in Taiwan Roadkill Observation Network from an environmental education perspective

Citizen science is a key approach in scientific research that involves crowdsourced data collection for environmental monitoring and conservation. Participants in citizen science can gain scientific knowledge, change their pro-environmental attitude and behavior. Taiwan Roadkill Observation Network (TaiRON) was founded in 2011 and has a community of approximately 20,000 members. To collect data for further biological research, participants adopt a simple approach: taking pictures of roadkill and uploading them to a database. While many studies have explored citizen science learning outcomes, this study examines how a program with a unique participant task – photographing road kill – has an impact on participants pro-environmental learning. The learning outcome with the largest change was environmental attitude, suggesting that the participants’ feelings, awareness, and values regarding the environment changed after joining TaiRON. The reasons for the attitude changes could be attributed to the characteristics of TaiRON’s work: participants collect data from animal carcasses, prompting them to reflect deeply on the value of life. The second key learning outcome was behavioural intention, which means participants were driven to action to mitigate roadkill occurrence. Our results provide an example of evaluating learning outcomes in citizen science and also contribute to promoting environmental literacy.

Scalable and Efficient Clustering for Fingerprint-Based Positioning

Indoor positioning based on IEEE 802.11 wireless LAN (Wi-Fi) fingerprinting needs a reference data set, also known as a radio map, in order to match the incoming fingerprint in the operational phase with the most similar fingerprint in the data set and then estimate the device position indoors. Scalability problems may arise when the radio map is large, e.g., providing positioning in large geographical areas or involving crowdsourced data collection. Some researchers divide the radio map into smaller independent clusters, such that the search area is reduced to less dense groups than the initial database with similar features. Thus, the computational load in the operational stage is reduced both at the user devices and on servers. Nevertheless, the clustering models are machine-learning algorithms without specific domain knowledge on indoor positioning or signal propagation. This work proposes several clustering variants to optimize the coarse and fine-grained search and evaluates them over different clustering models and data sets. Moreover, we provide guidelines to obtain efficient and accurate positioning depending on the data set features. Finally, we show that the proposed new clustering variants reduce the execution time by half and the positioning error by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\approx 7$ </tex-math></inline-formula> % with respect to fingerprinting with the traditional clustering models.

Seeing the Forest, Not the Trees – Crowdsourced Data Collection Methods for Sector-Wide Research

ABSTRACT Research that involves a large and broad sample of museums can produce a representative picture of the entire museum sector and lead to global insights that may not be attainable through a more local lens. However, many museum research projects use a small sample of museums, meant to represent the entire field. We propose a research method that distributes data collection across a broad swath of museums to provide local detail that can be used to assemble a collective picture on a topic of interest to the field. This method, called crowdsourced data collection, was used in a yearlong study of zoos and aquariums in North America, in which 95 institutions were asked to collect data for one to two survey modules per month. We hoped this approach would produce data comparable to data gathered with conventional methods and reduce burden on participating institutions. We found the method replicated nationally representative studies with two validated scales. While only one third of the institutions completed all modules, institutions typically did 8-9 modules, with only slight decreases in the probability of completing the study over time. These results suggest researchers can use crowdsourced data collection to reliably study the museum sector. We also discuss the challenges of this method for researchers and institutions participating as data collection sites.

A national crowdsourced network of low-cost fine particulate matter and aerosol optical depth monitors: results from the 2021 wildfire season in the United States

A novel instrument designed for crowdsourced data collection on air quality and aerosol optical depth.

Assessing Associations Between COVID-19 Symptomology and Adverse Outcomes After Piloting Crowdsourced Data Collection: Cross-sectional Survey Study.

Crowdsourcing is a useful way to rapidly collect information on COVID-19 symptoms. However, there are potential biases and data quality issues given the population that chooses to participate in crowdsourcing activities and the common strategies used to screen participants based on their previous experience. The study aimed to (1) build a pipeline to enable data quality and population representation checks in a pilot setting prior to deploying a final survey to a crowdsourcing platform, (2) assess COVID-19 symptomology among survey respondents who report a previous positive COVID-19 result, and (3) assess associations of symptomology groups and underlying chronic conditions with adverse outcomes due to COVID-19. We developed a web-based survey and hosted it on the Amazon Mechanical Turk (MTurk) crowdsourcing platform. We conducted a pilot study from August 5, 2020, to August 14, 2020, to refine the filtering criteria according to our needs before finalizing the pipeline. The final survey was posted from late August to December 31, 2020. Hierarchical cluster analyses were performed to identify COVID-19 symptomology groups, and logistic regression analyses were performed for hospitalization and mechanical ventilation outcomes. Finally, we performed a validation of study outcomes by comparing our findings to those reported in previous systematic reviews. The crowdsourcing pipeline facilitated piloting our survey study and revising the filtering criteria to target specific MTurk experience levels and to include a second attention check. We collected data from 1254 COVID-19-positive survey participants and identified the following 6 symptomology groups: abdominal and bladder pain (Group 1); flu-like symptoms (loss of smell/taste/appetite; Group 2); hoarseness and sputum production (Group 3); joint aches and stomach cramps (Group 4); eye or skin dryness and vomiting (Group 5); and no symptoms (Group 6). The risk factors for adverse COVID-19 outcomes differed for different symptomology groups. The only risk factor that remained significant across 4 symptomology groups was influenza vaccine in the previous year (Group 1: odds ratio [OR] 6.22, 95% CI 2.32-17.92; Group 2: OR 2.35, 95% CI 1.74-3.18; Group 3: OR 3.7, 95% CI 1.32-10.98; Group 4: OR 4.44, 95% CI 1.53-14.49). Our findings regarding the symptoms of abdominal pain, cough, fever, fatigue, shortness of breath, and vomiting as risk factors for COVID-19 adverse outcomes were concordant with the findings of other researchers. Some high-risk symptoms found in our study, including bladder pain, dry eyes or skin, and loss of appetite, were reported less frequently by other researchers and were not considered previously in relation to COVID-19 adverse outcomes. We demonstrated that a crowdsourced approach was effective for collecting data to assess symptomology associated with COVID-19. Such a strategy may facilitate efficient assessments in a dynamic intersection between emerging infectious diseases, and societal and environmental changes.

SUNS: A User-Friendly Scheme for Seamless and Ubiquitous Navigation Based on an Enhanced Indoor-Outdoor Environmental Awareness Approach

Ubiquitous and seamless indoor-outdoor (I/O) localization is the primary objective for gaining more user satisfaction and sustaining the prosperity of the location-based services (LBS) market. Regular users, on the other hand, may be unaware of the impact of activating multiple localization sources on localization performance and energy consumption, or may lack experience deciding when to enable or disable localization sources in different environments. Consequently, an automatic handover mechanism that can handle these decisions on a user’s behalf can appreciably improve user satisfaction. This study introduces an enhanced I/O environmental awareness service that provides an automated handover mechanism for seamless navigation based on multi-sensory navigation integration schemes. Moreover, the proposed service utilizes low-power consumption sensor (LPCS) indicators to execute continuous detection tasks and invoke GNSS in confusion scenarios, and transition intervals to make the most firm decision on the credibility of the LPCS-triggered transition and compensate for indicator thresholds. In this manner, GNSS are used for short intervals that help reduce detection latency and power consumption. Consequently, the proposed service guarantees accurate and reliable I/O detection while preserving low power consumption. Leveraging the proposed service as an automated handover helped realize seamless indoor-outdoor localization with less switching latency, using an integrated solution based on extended Kalman filter. Furthermore, the proposed energy-efficient service was utilized to confine crowdsourced data collection to the required areas (indoors and semi-indoors) and prevent excess data collection outdoors, thereby reducing power drainage. Accordingly, the negative impact of data collection on the user’s device can be mitigated, participation can be encouraged, and crowdsourcing systems can be widely adopted.

End-to-End Data Quality Assessment Using Trust for Data Shared IoT Deployments

Continued development of communication technologies has led to widespread Internet-of-Things (IoT) integration into various domains, including health, manufacturing, automotive, and precision agriculture. This has further led to the increased sharing of data among such domains to foster innovation. Most of these IoT deployments, however, are based on heterogeneous, pervasive sensors, which can lead to quality issues in the recorded data. This can lead to sharing of inaccurate or inconsistent data. There is a significant need to assess the quality of the collected data, should it be shared with multiple application domains, as inconsistencies in the data could have financial or health ramifications. This article builds on the recent research on trust metrics and presents a framework to integrate such metrics into the IoT data cycle for real-time data quality assessment. Critically, this article adopts a mechanism to facilitate end-user parameterization of a trust metric tailoring its use in the framework. Trust is a well-established metric that has been used to determine the validity of a piece or source of data in crowd-sourced or other unreliable data collection techniques such as that in IoT. The article further discusses how the trust-based framework eliminates the requirement for a gold standard and provides visibility into data quality assessment throughout the big data model. To qualify the use of trust as a measure of quality, an experiment is conducted using data collected from an IoT deployment of sensors to measure air quality in which low-cost sensors were colocated with a gold standard reference sensor. The calculated trust metric is compared with two well-understood metrics for data quality, root mean square error (RMSE), and mean absolute error (MAE). A strong correlation between the trust metric and the comparison metrics shows that trust may be used as an indicative quality metric for data quality. The metric incorporates the additional benefit of its ability for use in low context scenarios, as opposed to RMSE and MAE, which require a reference for comparison.

Tracking decision-making of backcountry users using GPS tracks and participant surveys

Snow avalanches are a significant natural hazard representing the primary risk of death to backcountry travelers in many alpine countries. Careful use of backcountry terrain through effective decision making can mitigate the risk of dangerous snowpack conditions, but requires relevant knowledge and experience.We present the results from a large-scale crowd sourced data collection method from backcountry users. Using GPS tracking via a smartphone application, coupled with online surveys, we investigate the intersection of geographical complexity, backcountry experience, demographics and behavioral biases on decision-making while navigating hazardous winter terrain. We use data from 770 GPS tracks, representing almost 1.3 million GPS points, as a geographic expression of a group's resulting decisions, and use them to quantify and understand their decision-making process.Our analysis focuses on the change in terrain use as quantified using the Avalanche Terrain Exposure Scale (ATES), and time spent in avalanche terrain, as a function of experience, avalanche hazard and other group factors. We show that self-identified experts rate themselves as significantly more skilled and also had higher levels of avalanche education. Experts also had an increased exposure to avalanche terrain overall, and also more severe terrain, as represented by median time in class 3 ATES terrain.