Academia.eduAcademia.edu

Outline

Title
Abstract
Introduction
Related Work
Participant Reliability
Model Analysis
Model Design
Estimating Participant Reliability
Participant's Location Visit Indicator
Methods in Comparison
Data Analysis
Future Work
Conclusion
References
All Topics
Computer Science
Machine Learning

Truth Discovery in Crowdsourced Detection of Spatial Events

Profile image of alice tonioloalice toniolo

2014, Proceedings of the 23rd ACM International Conference on Conference on Information and Knowledge Management

https://doi.org/10.1145/2661829.2662003
visibility

…

description

14 pages

link

1 file

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.

Related papers

A Probabilistic Approach to Socio-Geographic Reality Mining
Katayoun Farrahi

ACM SIGMultimedia Records, 2011

As we live our daily lives, our surroundings know about it. Our surroundings consist of people, but also our electronic devices. Our mobile phones, for example, continuously sense our movements and interactions. This socio-geographic data could be continuously captured by hundreds of millions of people around the world and promises to reveal important behavioral clues about humans in a manner never before possible. Mining patterns of human behavior from large-scale mobile phone data has deep potential impact ...

View PDFchevron_right
Crowdsourcing with Smartphones
Demetrios Zeinalipour-Yazti, Andreas Konstantinidis

IEEE Internet Computing, 2000

Smartphones can unfold the full potential of crowdsourcing, allowing users to transparently contribute to complex and novel problem solving. We present the intrinsic characteristics of smartphones, a taxonomy that classifies the emerging field of mobile crowdsourcing and three in-house applications that optimize location-based search and similarity services over data generated by a crowd: (i) SmartTrace + enables similarity matching between a given pattern and the trajectories of smartphone users, keeping the target trajectories private; (ii) Crowdcast enables location-based interaction by efficiently calculating the k nearest neighbors for each user at all times; (iii) SmartP2P optimizes energy, time and recall of search in a mobile social community for objects generated by a crowd. We show how these applications can be deployed on SmartLab, a novel cloud of 40+ Android devices deployed at University of Cyprus, providing an open testbed that facilitates research and development of applications on smartphones at a massive scale.

View PDFchevron_right
Social Event Detection in Massive Mobile Phone Data Using Probabilistic Location Inference
Frédéric Morlot

2011

The unprecedented amount of data from mobile phones creates new possibilities to analyze various aspects of human behavior. Over the last few years, much effort has been devoted to studying the mobility patterns of humans. In this paper we will focus on unusually large gatherings of people, i.e. unusual social events. We introduce the methodology of detecting such social events

View PDFchevron_right
Crowdsourcing (in) Voluntary Citizen Geospatial Data from Google Android Smartphones
Mark Lindquist

2018

By 2016 it is projected that there will be over 2 billion smartphone users worldwide with Google’s Android operating system installed on the vast majority of these devices (82.8% market share as of Q2 2015). The majority of Android smartphones sold in the last three years have GPS functionality and, unless actively disabled by the user, have a Location History feature that records the phone's geographic coordinates at frequent intervals whenever the phone is powered on. User interaction with the device or an active display is not required for the recording of these data and as such users may be unaware of the data being collected. Google has introduced Google Maps Timeline that facilitates mapping and interacting with a user location history data via Google Maps, and also provides a service called Takeout with the explicit goal of facilitating user access to personal data held by Google, including Location History. The Location History data can be downloaded and analysed using G...

View PDFchevron_right
Efficient Algorithms for Accuracy Improvement in Mobile Crowdsensing Vehicular Applications
Emanuele Lattanzi

2015

Mobile crowdsensing is emerging as a cost-effective solution to conduct extensive monitoring campaigns by exploiting the potential of mobile terminals with unprecedented communication, processing, and sensing capabilities. On the other hand, data provided by a large number of end-users equipped with heterogeneous devices pose trust and accuracy issues that might impair the overall reliability and usability of the system. Trust management techniques developed in the field of online social networks can be effectively used to detect and isolate cheating users, but they cannot avoid the risk of inaccurate data provided by trustworthy agents because of the inherent limitations of their devices or of the adverse conditions in which they operate. This work presents efficient algorithms for compensating the inaccuracy of crowdsensing geospatial data to be reported on a road map. The paper illustrates on a representative case study the main issues of map matching and the effectiveness of the...

View PDFchevron_right
Demo: Distilling likely truth from noisy streaming data with Apollo
Hieu Le

Proceedings of the 9th ACM Conference on Embedded Networked Sensor Systems, 2011

At CPSWeek 2011, the authors presented a demonstration of Apollo, a fact-finder for participatory sensing that ranks archived human-centric and sensor data by credibility. The current demonstration significantly extends our previous work by allowing Apollo to operate on live streaming data; in this case, live Twitter feeds. As the role of humans as sensors increases in emerging sensing applications, a principled approach becomes necessary to address the problem of ascertaining the veracity of sources and observations made by them. Participatory and social sensing applications may use potentially unreliable or unverified sources, such as a phone-based sensing application that grows virally in a large un-vetted population, a disaster-response application, where conflicting damage assessment reports may come from large numbers of different volunteers, or a military application, where friendly observers at a remote location may make hard-to-verify claims about local events. Apollo analyzes noisy data that increasingly plagues human-centric sensing to determine which items of information are more likely to be true.

View PDFchevron_right
Reality mining in urban space
Symeon Papadopoulos, Spiros Nikolopoulos

The widespread adoption of smartphones and other sensor-capable devices and the availability of infrastructure capable of storing and sharing sensor information have been the main enablers for the intriguing and ambitious concept of Reality Mining. Reality mining is concerned with the collection and analysis of sensed data pertaining to different aspects of human social behavior, allowing to observe the societal dynamics and predict what individuals will do next, model the behavior of large organizations, detect trends, spot emerging phenomena and events, mine public opinion, etc. The concept of reality mining is even more intriguing when applied in the context of urban space, since user-generated content can be analyzed to reveal the citizens' perspective on various aspects of the city reality. In this paper, after elaborating on the concept of reality mining in urban space, we present three cases where user-generated content is employed to discover the citizens' perspective about: a) the city attractions, b) the city issues/problems, and c) major events in the city.

View PDFchevron_right
Valid Statements by the Crowd: Statistical Measures for Precision in Crowdsourced Mobile Measurements
Stefan Wunderer

Network

Crowdsourced network measurements (CNMs) are becoming increasingly popular as they assess the performance of a mobile network from the end user’s perspective on a large scale. Here, network measurements are performed directly on the end-users’ devices, thus taking advantage of the real-world conditions end-users encounter. However, this type of uncontrolled measurement raises questions about its validity and reliability. The problem lies in the nature of this type of data collection. In CNMs, mobile network subscribers are involved to a large extent in the measurement process, and collect data themselves for the operator. The collection of data on user devices in arbitrary locations and at uncontrolled times requires means to ensure validity and reliability. To address this issue, our paper defines concepts and guidelines for analyzing the precision of CNMs; specifically, the number of measurements required to make valid statements. In addition to the formal definition of the aspect...

View PDFchevron_right
Trustworthiness of Context-Aware Urban Pollution Data in Mobile Crowd Sensing
Lucia Vaira

IEEE Access

Urban pollution is usually monitored via fixed stations that provide detailed and reliable information, thanks to equipment quality and effective measuring protocols, but these sampled data are gathered from very limited areas and through discontinuous monitoring campaigns. Currently, the spread of mobile devices has fostered the development of new approaches, like Mobile Crowd Sensing (MCS), increasing the chances of using smartphones as suitable sensors in the urban monitoring scenario, because it potentially contributes massive ubiquitous data at relatively low cost. However, MCS is useless (or even counter-productive), if contributed data are not trustworthy, due to wrong data-collection procedures by non-expert practitioners. Contextualizing monitored data with those coming from phone-embedded sensors and from time/space proximity can improve data trustworthiness. This work focuses on the development of an algorithm that exploits context awareness to improve the reliability of MCS collected data. It has been validated against some real use cases for noise pollution and promises to improve the trustworthiness of end users generated data.

View PDFchevron_right
Improving the Performance of Mobile Phone Crowdsourcing Applications
Gita Sukthankar

Adaptive Agents and Multi-Agents Systems, 2015

Mobile phone crowdsourcing is a powerful tool for many types of distributed sensing problems. However, a central issue with this type of system is that it relies on user contributed data, which may be sparse or erroneous. This paper describes our experiences developing a mobile phone crowdsourcing app, Kpark, for monitoring parking availability on a university campus. Our system combines multiple trustbased data fusion techniques to improve the quality of user submitted parking reports and is currently being used by over 1500 students.

View PDFchevron_right

References (43)

  1. Amazon mechanical turk. https://www.mturk.com/mturk/ welcome.
  2. Field agent. http://www.fieldagent.net.
  3. Gigwalk. http://gigwalk.com.
  4. Location obfuscation. http://en.wikipedia.org/wiki/Location obfuscation.
  5. Taskrabbit. http://www.taskrabbit.com.
  6. R. Amici et al. Performance assessment of an epidemic protocol in vanet using real traces. Procedia Computer Science, 40:92-99, 2014.
  7. A. R. Beresford and F. Stajano. Location privacy in pervasive computing. Pervasive Computing, IEEE, 2(1):46-55, 2003.
  8. C. M. Bishop and N. M. Nasrabadi. Pattern recognition and machine learning. Springer New York, 2006.
  9. A. Das et al. Debiasing social wisdom. In KDD, pages 500-508. ACM, 2013.
  10. A. S. Das et al. Google news personalization: scalable online collaborative filtering. In WWW, pages 271-280. ACM, 2007.
  11. A. P. Dawid and A. M. Skene. Maximum likelihood estimation of observer error-rates using the em algorithm. Applied Statistics, pages 20-28, 1979.
  12. J. Dean and S. Ghemawat. Mapreduce: simplified data processing on large clusters. Communications of the ACM, 51(1):107-113, 2008.
  13. X. L. Dong et al. Integrating conflicting data: the role of source dependence. VLDB Endowment, 2(1):550-561, 2009.
  14. X. L. Dong et al. Truth discovery and copying detection in a dynamic world. VLDB Endowment, 2(1):562-573, 2009.
  15. M. C. Gonzalez et al. Understanding individual human mobility patterns. Nature, 453(7196):779-782, 2008.
  16. S. Isaacman et al. Human mobility modeling at metropolitan scales. In MobiSys, pages 239-252. ACM, 2012.
  17. S. Kisilevich et al. Spatio-temporal clustering. Springer, 2010.
  18. J. M. Kleinberg. Authoritative sources in a hyperlinked environ- ment. Journal of the ACM, 46(5):604-632, 1999.
  19. K. Lin et al. Energy-accuracy trade-off for continuous mobile device location. In MobiSys, pages 285-298. ACM, 2010.
  20. J. S. Liu. The collapsed gibbs sampler in bayesian computations with applications to a gene regulation problem. Journal of the American Statistical Association, 89(427):958-966, 1994.
  21. J. Lorenz et al. How social influence can undermine the wisdom of crowd effect. Proceedings of the National Academy of Sciences, 108(22):9020-9025, 2011.
  22. Y. Lou et al. Map-matching for low-sampling-rate gps trajectories. In GIS, pages 352-361. ACM, 2009.
  23. P. Mohan et al. Nericell: rich monitoring of road and traffic conditions using mobile smartphones. In SenSys, pages 323-336. ACM, 2008.
  24. M. Musthag and D. Ganesan. Labor dynamics in a mobile micro- task market. In CHI, pages 641-650. ACM, 2013.
  25. R. W. Ouyang et al. Energy efficient assisted gps measurement and path reconstruction for people tracking. In GLOBECOM, pages 1-5. IEEE, 2010.
  26. R. W. Ouyang et al. If you see something, swipe towards it: crowdsourced event localization using smartphones. In UbiComp, pages 23-32. ACM, 2013.
  27. R. W. Ouyang et al. Truth discovery in crowdsourced detection of spatial events. In CIKM, pages 461-470. ACM, 2014.
  28. J. Paek et al. Energy-efficient rate-adaptive gps-based positioning for smartphones. In MobiSys, pages 299-314. ACM, 2010.
  29. J. Pasternack and D. Roth. Knowing what to believe (when you already know something). In COLING, pages 877-885. Association for Computational Linguistics, 2010.
  30. M. Pi órkowski et al. A parsimonious model of mobile partitioned networks with clustering. In COMSNETS, pages 1-10. IEEE, 2009.
  31. G.-J. Qi et al. Mining collective intelligence in diverse groups. In WWW, pages 1041-1052. ACM, 2013.
  32. V. C. Raykar et al. Learning from crowds. The Journal of Machine Learning Research, 99:1297-1322, 2010.
  33. S. Reddy et al. Recruitment framework for participatory sensing data collections. In Pervasive Computing, pages 138-155. Springer, 2010.
  34. D. Wang et al. On truth discovery in social sensing: a maximum likelihood estimation approach. In IPSN, pages 233-244. ACM, 2012.
  35. D. Wang et al. On credibility estimation tradeoffs in assured social sensing. IEEE JSAC, 31(6):1026-1037, 2013.
  36. T. Wang et al. Quantifying herding effects in crowd wisdom. In KDD, pages 1087-1096. ACM, 2014.
  37. P. Welinder et al. The multidimensional wisdom of crowds. In NIPS, pages 2424-2432, 2010.
  38. J. Whitehill et al. Whose vote should count more: Optimal integration of labels from labelers of unknown expertise. In NIPS, pages 2035-2043, 2009.
  39. M. Ye et al. Exploiting geographical influence for collaborative point-of-interest recommendation. In SIGIR, pages 325-334. ACM, 2011.
  40. X. Yin, J. Han, and P. S. Yu. Truth discovery with multiple conflict- ing information providers on the web. IEEE TKDE, 20(6):796-808, 2008.
  41. X. Yin and W. Tan. Semi-supervised truth discovery. In WWW, pages 217-226. ACM, 2011.
  42. B. Zhao et al. A bayesian approach to discovering truth from con- flicting sources for data integration. VLDB Endowment, 5(6):550- 561, 2012.
  43. Z. Zhuang et al. Improving energy efficiency of location sensing on smartphones. In MobiSys, pages 315-330. ACM, 2010.

Related papers

Improving Location Reliability in Crowd Sensed Data with Minimal Efforts
裕初 左

—People-centric sensing with smart phones can be used for large scale sensing of the physical world by leveraging the cameras, microphones, GPSs, accelerometers, and other sensors on the phones. Ranging from manual photo tasks to automated sensing tasks for activity monitoring, any task can be crowd sourced to smart phones to sense data from different locations at reduced cost. However, the sensed data submitted by participants is not always reliable as they can submit false data to earn money without executing the actual task. Therefore, it is important to validate the sensed data. Validating the context of every sensed data point of each participant is not a scalable solution. One alternative is to first validate the location associated with the sensed data points in order to achieve a certain degree of reliability about the sensed data. However, location validation without support from the wireless carriers is difficult. To address this problem, we propose ILR, a scheme in which we Improve the Location Reliability of mobile crowd sensed data with minimal human efforts. In this scheme, we bootstrap the trust in the system by first manually or automatically using image processing techniques validating a small number of photos submitted by participants. Based on these validations, the location of these photos is assumed to be trusted. Second, we extend this location trust to co-located sensed data points found in the Bluetooth range of the devices that provided the validated photos. This transitive trust is extended until all the co-located tasks are trusted or no new data points are found. In addition, the scheme also helps to detect false location claims associated with sensed data. We applied ILR on data collected from our McSense prototype deployed on Android phones used by students on our campus and detected a significant percentage of the malicious users. Simulation results demonstrate that ILR works well at various densities and helps detect the false location claims based on a minimal number of validations.

View PDFchevron_right
On truth discovery in social sensing
Hieu Le

Proceedings of the 11th international conference on Information Processing in Sensor Networks - IPSN '12, 2012

This paper addresses the challenge of truth discovery from noisy social sensing data. The work is motivated by the emergence of social sensing as a data collection paradigm of growing interest, where humans perform sensory data collection tasks. A challenge in social sensing applications lies in the noisy nature of data. Unlike the case with wellcalibrated and well-tested infrastructure sensors, humans are less reliable, and the likelihood that participants' measurements are correct is often unknown a priori. Given a set of human participants of unknown reliability together with their sensory measurements, this paper poses the question of whether one can use this information alone to determine, in an analytically founded manner, the probability that a given measurement is true. The paper focuses on binary measurements. While some previous work approached the answer in a heuristic manner, we offer the first optimal solution to the above truth discovery problem. Optimality, in the sense of maximum likelihood estimation, is attained by solving an expectation maximization problem that returns the best guess regarding the correctness of each measurement. The approach is shown to outperform the state of the art fact-finding heuristics, as well as simple baselines such as majority voting.

View PDFchevron_right
Scalable and privacy-respectful interactive discovery of place semantics from human mobility traces
Piotr Jankowski

Information Visualization, 2015

Mobility diaries of a large number of people are needed for assessing transportation infrastructure and spatial development planning. Acquisition of personal mobility diaries through population surveys is a costly and error-prone endeavour. We examine an alternative approach to obtaining similar information from episodic digital traces of people’s presence in various locations, which appear when people use their mobile devices for making phone calls, accessing the Internet or posting georeferenced contents (texts, photos or videos) in social media. Having episodic traces of a person over a long time period, it is possible to detect significant (repeatedly visited) personal places and identify them as home, work or place of social activities based on temporal patterns of a person’s presence in these places. Such analysis, however, can lead to compromising personal privacy. We have investigated the feasibility of deriving place meanings and reconstructing personal mobility diaries whi...

View PDFchevron_right
Location estimation using crowdsourced geospatial narratives
G. Skoumas

The "crowd" has become a very important geospatial data provider. Subsumed under the term Volunteered Geographic Information (VGI), non-expert users have been providing a wealth of quantitative geospatial data online. With spatial reasoning being a basic form of human cognition, narratives expressing geospatial experiences, e.g., travel blogs, would provide an even bigger source of geospatial data. Textual narratives typically contain qualitative data in the form of objects and spatial relationships. The scope of this work is (i) to extract these relationships from user-generated texts, (ii) to quantify them and (iii) to reason about object locations based only on this qualitative data. We use information extraction methods to identify toponyms and spatial relationships and to formulate a quantitative approach based on distance and orientation features to represent the latter. Positional probability distributions for spatial relationships are determined by means of a greedy Expectation Maximization-based (EM) algorithm. These estimates are then used to "triangulate" the positions of unknown object locations. Experiments using a text corpus harvested from travel blog sites establish the considerable location estimation accuracy of the proposed approach.

View PDFchevron_right
Reshaping mobile crowd sensing using cross validation to improve data credibility
Tony T. Luo

IEEE Globecom, 2017

Data credibility is a crucial issue in mobile crowd sensing (MCS) and, more generally, people-centric Internet of Things (IoT). Prior work takes approaches such as incentive mechanism design and data mining to address this issue, while overlooking the power of crowds itself, which we exploit in this paper. In particular, we propose a cross validation approach which seeks a validating crowd to verify the data credibility of the original sensing crowd, and uses the verification result to reshape the original sensing dataset into a more credible posterior belief of the ground truth. Following this approach, we design a specific cross validation mechanism, which integrates four sampling techniques with a privacy-aware competency-adaptive push (PACAP) algorithm and is applicable to time-sensitive and quality-critical MCS applications. It does not require redesigning a new MCS system but rather functions as a lightweight "plug-in", making it easier for practical adoption. Our results demonstrate that the proposed mechanism substantially improves data credibility in terms of both reinforcing obscure truths and scavenging hidden truths.

View PDFchevron_right

Related topics

  • Graphical Models
    • 580 California St., Suite 400
    San Francisco, CA, 94104
    © 2025 Academia. All rights reserved