No abstract

Anomalies in online social networks can signify irregular, and often illegal behaviour. Detection of such anomalies has been used to identify malicious individuals, including spammers, sexual predators, and online fraudsters. In this paper we survey existing computational techniques for detecting anomalies in online social networks. We characterise anomalies as being either static or dynamic, and as being labelled or unlabelled, and survey methods for detecting these different types of anomalies. We suggest that the detection of anomalies in online social networks is composed of two sub-processes; the selection and calculation of network features, and the classification of observations from this feature space. In addition, this paper provides an overview of the types of problems that anomaly detection can address and identifies key areas for future research.

No abstract

Introduction Dizziness is a common complaint presented in the emergency department (ED). A subset of these patients will present with acute vestibular syndrome (AVS). AVS is a clinical syndrome defined by the presence of vertigo, nystagmus, head motion intolerance, ataxia, and nausea/vomiting. These symptoms are most often due to benign vestibular neuritis; however, they can be a sign of a dangerous central cause, i.e., vertebrobasilar stroke. The Head Impulse test, Nystagmus, Test of Skew (HINTS) examination has been proposed as a bedside test for frontline clinicians to rule out stroke in those presenting with AVS. Our objective was to assess the diagnostic accuracy of the HINTS examination to rule out a central cause of vertigo in an adult population presenting to the ED with AVS. Our aim was to assess the diagnostic accuracy when performed by emergency physicians versus neurologists. Methods We searched PubMed, Medline, Embase, the Cochrane database, and relevant conference abstracts from 2009 to September 2019 and performed hand searches. No restrictions for language or study type were imposed. Prospective studies with patients presenting with AVS using criterion standard of computed tomography and/or magnetic resonance imaging were selected for review. Two independent reviewers extracted data from relevant studies. Studies were combined if low clinical and statistical heterogeneity was present. Study quality was assessed using the QUADAS‐2 tool. Random effects meta‐analysis was performed using RevMan 5 and SAS 9.3. Results A total of five studies with 617 participants met the inclusion criteria. The mean (±SD) study length was 5.3 (±3.3) years. Prevalence of vertebrobasilar stroke ranged 9.3% to 44% (mean ± SD = 39.1% ± 17.1%). The most common diagnoses were vertebrobasilar stroke (mean ± SD = 34.8% ± 17.1%), peripheral cause (mean ± SD = 30.9% ± 16%), and intracerebral hemorrhage (mean ± SD = 2.2% ± 0.5%). The HINTS examination, when performed by neurologists, had a sensitivity of 96.7% (95% CI = 93.1% to 98.5%, I2 = 0%) and specificity of 94.8% (95% CI = 91% to 97.1%, I2 = 0%). When performed by a cohort of physicians including both emergency physicians (board certified) and neurologists (fellowship trained in neurootology or vascular neurology) the sensitivity was 83% (95% CI = 63% to 95%) and specificity was 44% (95% CI = 36% to 51%). Conclusions The HINTS examination, when used in isolation by emergency physicians, has not been shown to be sufficiently accurate to rule out a stroke in those presenting with AVS.