English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Detecting anomalous activity in human mobility data has a number ofapplications including road hazard sensing, telematic based insurance, andfraud detection in taxi services and ride sharing. In this paper we address twochallenges that arise in the study of anomalous human trajectories: 1) a lackof ground truth data on what defines an anomaly and 2) the dependence ofexisting methods on significant pre-processing and feature engineering. Whilegenerative adversarial networks seem like a natural fit for addressing thesechallenges, we find that existing GAN based anomaly detection algorithmsperform poorly due to their inability to handle multimodal patterns. For thispurpose we introduce an infinite Gaussian mixture model coupled with(bi-directional) generative adversarial networks, IGMM-GAN, that is able togenerate synthetic, yet realistic, human mobility data and simultaneouslyfacilitates multimodal anomaly detection. Through estimation of a generativeprobability density on the space of human trajectories, we are able to generaterealistic synthetic datasets that can be used to benchmark existing anomalydetection methods. The estimated multimodal density also allows for a naturaldefinition of outlier that we use for detecting anomalous trajectories. Weillustrate our methodology and its improvement over existing GAN anomalydetection on several human mobility datasets, along with MNIST.

Coupled IGMM-GANs for deep multimodal anomaly detection in human mobility data