TU Darmstadt / ULB / TUbiblio

Bayesian Heatmaps: Probabilistic Classification with Multiple Unreliable Information Sources

Simpson, Edwin and Reece, Steven and Roberts, Stephen J. (2017):
Bayesian Heatmaps: Probabilistic Classification with Multiple Unreliable Information Sources.
In: Lecture Notes in Computer Science, Lecture Notes in Computer, In: Machine Learning and Knowledge Discovery in Databases (ECML PKDD 2017), Springer, Skopje, Macedonia, pp. 109-125, [Online-Edition: https://link.springer.com/chapter/10.1007/978-3-319-71246-8_...],
[Conference or Workshop Item]

Abstract

Unstructured data from diverse sources, such as social media and aerial imagery, can provide valuable up-to-date information for intelligent situation assessment. Mining these different information sources could bring major benefits to applications such as situation awareness in disaster zones and mapping the spread of diseases. Such applications depend on classifying the situation across a region of interest, which can be depicted as a spatial 'heatmap'. Annotating unstructured data using crowdsourcing or automated classifiers produces individual classifications at sparse locations that typically contain many errors. We propose a novel Bayesian approach that models the relevance, error rates and bias of each information source, enabling us to learn a spatial Gaussian Process classifier by aggregating data from multiple sources with varying reliability and relevance. Our method does not require gold-labelled data and can make predictions at any location in an area of interest given only sparse observations. We show empirically that our approach can handle noisy and biased data sources, and that simultaneously inferring reliability and transferring information between neighbouring reports leads to more accurate predictions. We demonstrate our method on two real-world problems from disaster response, showing how our approach reduces the amount of crowdsourced data required and can be used to generate valuable heatmap visualisations from SMS messages and satellite images.

Item Type: Conference or Workshop Item
Erschienen: 2017
Creators: Simpson, Edwin and Reece, Steven and Roberts, Stephen J.
Title: Bayesian Heatmaps: Probabilistic Classification with Multiple Unreliable Information Sources
Language: English
Abstract:

Unstructured data from diverse sources, such as social media and aerial imagery, can provide valuable up-to-date information for intelligent situation assessment. Mining these different information sources could bring major benefits to applications such as situation awareness in disaster zones and mapping the spread of diseases. Such applications depend on classifying the situation across a region of interest, which can be depicted as a spatial 'heatmap'. Annotating unstructured data using crowdsourcing or automated classifiers produces individual classifications at sparse locations that typically contain many errors. We propose a novel Bayesian approach that models the relevance, error rates and bias of each information source, enabling us to learn a spatial Gaussian Process classifier by aggregating data from multiple sources with varying reliability and relevance. Our method does not require gold-labelled data and can make predictions at any location in an area of interest given only sparse observations. We show empirically that our approach can handle noisy and biased data sources, and that simultaneously inferring reliability and transferring information between neighbouring reports leads to more accurate predictions. We demonstrate our method on two real-world problems from disaster response, showing how our approach reduces the amount of crowdsourced data required and can be used to generate valuable heatmap visualisations from SMS messages and satellite images.

Title of Book: Machine Learning and Knowledge Discovery in Databases (ECML PKDD 2017)
Series Name: Lecture Notes in Computer Science
Volume: Lecture Notes in Computer
Number: 10535
Publisher: Springer
Uncontrolled Keywords: Bayesian machine learning, gaussian process, crowdsourcing, variational
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Ubiquitous Knowledge Processing
Event Location: Skopje, Macedonia
Date Deposited: 12 Sep 2017 19:22
Official URL: https://link.springer.com/chapter/10.1007/978-3-319-71246-8_...
Identification Number: TUD-CS-2017-0245
Related URLs:
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)

View Item View Item