Continuous monitoring of distance-based outliers over data streams

Kontaki, Maria; Gounaris, Anastasios; Papadopoulos, Apostolos N.; Tsichlas, Kostas; Manolopoulos, Yannis

doi:10.1109/icde.2011.5767923

Cited by 137 publications

(140 citation statements)

References 15 publications

Supporting

Mentioning

138

Contrasting

Order By: Relevance

“…This effectively guides the analysts to determine what outlier type and parameter settings are proper for the given context. 4) demonstrate of our runtime performance monitoring view by contrasting our LEAP strategy-based distance-threshold outlier algorithm against the alternative methods in [10]. This demonstration confirms the 2 to 3 orders of magnitude performance gain of LEAP strategy in response time (Section 3).…”

Section: Vsoutlier Demonstrationsupporting

confidence: 62%

“…That is, to identify whether a point pi is an outlier in a dataset D, one may not need the distance between pi to every other point in D. Therefore the state-of-the-art techniques [1,16,10] which rely on complete neighborhood searches to determine the status (outlier or inlier) of each data point are not efficient.…”

Section: Minimal Probing Optimizationmentioning

confidence: 99%

“…Although in [7] several distance-threshold outlier detection algorithms [1,10] have been demonstrated, they cannot handle big data streams in realtime due to the expense of the exhaustive neighborhood searches that they rely on to locate outliers. Yet such big data streams are increasingly common in modern streaming applications.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Interactive outlier exploration in big data streams

2014

View full text Add to dashboard Cite

We demonstrate our VSOutlier system for supporting interactive exploration of outliers in big data streams. VSOutlier not only supports a rich variety of outlier types supported by innovative and efficient outlier detection strategies, but also provides a rich set of interactive interfaces to explore outliers in real time. Using the stock transactions dataset from the US stock market and the moving objects dataset from MITRE, we demonstrate that the VSOutlier system enables analysts to more efficiently identify, understand, and respond to phenomena of interest in near real-time even when applied to high volume streams.

show abstract

Section: Vsoutlier Demonstrationsupporting

confidence: 62%

Section: Minimal Probing Optimizationmentioning

confidence: 99%

See 1 more Smart Citation

Interactive outlier exploration in big data streams

2014

View full text Add to dashboard Cite

show abstract

“…The problem of outlier detection has also been addressed by the database and data mining communities [16], aiming at solving the problem of scalability. The Banking Industries have not been spared from the need to keep up with their constantly changing industry to stay viable and competitive [17].…”

Section: Related Workmentioning

confidence: 99%

“…Outlier detection has been studied broadly by the statistics community [16], where the objects are modelled as a distribution, and objects are marked as outliers depending on their deviation from this distribution. The problem of outlier detection has also been addressed by the database and data mining communities [16], aiming at solving the problem of scalability.…”

Section: Related Workmentioning

confidence: 99%

Fraud Detection on Bulk Tax Data Using Business Intelligence Data Mining Tool: A Case of Zambia Revenue Authority

Mwanza¹

2016

International Journal of Advanced Research in Computer and Comm

View full text Add to dashboard Cite

Zambia Revenue Authority (ZRA) generates large volumes of data that need complex mechanisms in order to extract useful tax information. The purpose of the study was to develop a data mining model for detection of fraud on tax and taxpayer data for ZRA. This study focused on two areas. These were (1) the baseline study that helped to establish the extent of the challenges in fraud detection for the tax payers and (2) the automation and development of the fraud detection tool using the results from the baseline study. Our baseline study showed that the current methodologies, processes, architectures, and technologies that were being used to transform raw data into meaningful and useful information were tedious and time consuming. In order to detect fraud they depended on random audits, informants and under-cover operations. A model which implements outlier algorithms for fraud detection, Continuous Monitoring of Distance Based and Distance Based Outlier Queries was then developed. We used both algorithms to analyse the domestic tax payments to detect underpayments and overpayments according to business rules. Underpayments and overpayments are marked as outliers. Results generated by our tool showed improved accuracy and takes less time in order to detect under and over payments as outliers when compared to the older methods.

show abstract

A high‐performance IoT solution to reduce frost damages in stone fruits

Guillén‐Navarro

Martínez‐España

Lopez

et al. 2019

Concurrency and Computation

View full text Add to dashboard Cite

Agriculture is one of the key sectors where technology is opening new opportunities to break up the market. The Internet of Things (IoT) could reduce the production costs and increase the product quality by providing intelligence services via IoT analytics. However, the hard weather conditions and the lack of connectivity in this field limit the successful deployment of such services as they require both, ie, fully connected infrastructures and highly computational resources. Edge computing has emerged as a solution to bring computing power in close proximity to the sensors, providing energy savings, highly responsive web services, and the ability to mask transient cloud outages. In this paper, we propose an IoT monitoring system to activate anti-frost techniques to avoid crop loss, by defining two intelligent services to detect outliers caused by the sensor errors. The former is a nearest neighbor technique and the latter is the k-means algorithm, which provides better quality results but it increases the computational cost. Cloud versus edge computing approaches are analyzed by targeting two different low-power GPUs. Our experimental results show that cloud-based approaches provides highest performance in general but edge computing is a compelling alternative to mask transient cloud outages and provide highly responsive data analytic services in technologically hostile environments.

show abstract

Continuous monitoring of distance-based outliers over data streams

Cited by 137 publications

References 15 publications

Interactive outlier exploration in big data streams

Interactive outlier exploration in big data streams

Fraud Detection on Bulk Tax Data Using Business Intelligence Data Mining Tool: A Case of Zambia Revenue Authority

A high‐performance IoT solution to reduce frost damages in stone fruits

Contact Info

Product

Resources

About