Truth Discovery in Crowdsourced Detection of Spatial Events

Abstract

Abstract-The ubiquity of smartphones has led to the emergence of mobile crowdsourcing tasks such as the detection of spatial events when smartphone users move around in their daily lives. However, the credibility of those detected events can be negatively impacted by unreliable participants with low-quality data. Consequently, a major challenge in mobile crowdsourcing is truth discovery, i.e., to discover true events from diverse and noisy participants' reports. This problem is uniquely distinct from its online counterpart in that it involves uncertainties in both participants' mobility and reliability. Decoupling these two types of uncertainties through location tracking will raise severe privacy and energy issues, whereas simply ignoring missing reports or treating them as negative reports will significantly degrade the accuracy of truth discovery. In this paper, we propose two new unsupervised models, i.e., Truth finder for Spatial Events (TSE) and Personalized Truth finder for Spatial Events (PTSE), to tackle this problem. In TSE, we model location popularity, location visit indicators, truths of events, and three-way participant reliability in a unified framework. In PTSE, we further model personal location visit tendencies. These proposed models are capable of effectively handling various types of uncertainties and automatically discovering truths without any supervision or location tracking. Experimental results on both real-world and synthetic datasets demonstrate that our proposed models outperform existing state-of-the-art truth discovery approaches in the mobile crowdsourcing environment.

Debiasing crowdsourced quantitative characteristics in local businesses and services

“…Although there are other truth finding methods proposed for crowdsourced image classification [17], social sensing [21] and crowdsourced detection of spatial events [12], they are mainly designed for binary truth and extending them to multinomial truth and quantitative truth is problematic. We thus do not include them for comparison.…”

“…Third, the quantitative truth needs to be found in an unsupervised manner as it is difficult to manually collect and annotate ground truth for supervised model training in reality. Fourth, existing truth finding algorithms [12,15,21,28,30] mostly focus on aggregating conflicting categorical information (e.g., which option is most likely to be correct) and they are not directly applicable or are not efficient for aggregating noisy crowdsourced quantitative information.…”

A Survey on Security, Privacy, and Trust in Mobile Crowdsourcing

“…Based on this idea, Quyang et al studied the process of how a crowdsourced report is generated [76]. In order to make a report, a worker must physically present at a certain location to observe whether there is any target event.…”

“…Two mobility-aware schemes were proposed in [78] and [79], which take into account the context or mobility trajectory of workers to decide the likelihood that a worker has actually generated the sensing report it uploads. Just like the work in [76], only low-level DT is offered, and these schemes lack the consideration of security and privacy issues.…”

Security and Privacy Dimensions in Next Generation DDDAS/Infosymbiotic Systems: A Position Paper

“…Recent works [12] have used graphical probabilistic models for truth discovery in crowdsourced detection of spatial events. However, they only consider the static case where the input is the user report at a given time stamp.…”