Novel statistical methods for detecting anomalies in data streams.

用于检测数据流异常的新颖统计方法。

• 批准号: 1967547
• 金额: --
• 依托单位: Lancaster University
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2016
• 资助国家: 英国
• 起止时间: 2016 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-1967547
• 关键词: Novel; statistical; methods; detecting; anomalies

The low cost of sensors means that the performance of many mechanical devices, from plane engines to routers, is now monitored continuously. This is done in order to detect problems with the underlying device in order to allow for action to be taken. However, the amount of data gathered has become so large that manual inspection is no longer possible. This makes automated methods to monitor performance data indispensable. My PhD focusses on developing novel statistical methods to detect anomalies, or untypical behaviour, in such data streams. More effective methods would allow to detect a wider range of anomalies, which in turn would allow to detect problems earlier, thus reducing their impact. Anomaly detection methods are also used for a range of other applications ranging from fraud prevention to cyber security. Specific research questions include (i) how to differentiate between different types of anomaly; (ii) developing statistical algorithms that can scale to high-dimensional and high-frequency data streams; (iii) understanding the theoretical properties of the new statistical methods.In partnership with BT.This project lies with the area of Statistics and Applied Probability (Computational Statistics, Statistical Methodology, Time Series).

传感器的低成本意味着许多机械设备的性能，从飞机发动机到路由器，现在都可以持续监控。这样做是为了检测底层设备的问题，以便采取行动。然而，收集的数据量已经变得如此之大，以至于人工检查不再可能。这使得监控性能数据的自动化方法不可或缺。我的博士学位专注于开发新的统计方法，以检测此类数据流中的异常或非典型行为。更有效的方法将有助于发现更广泛的异常情况，从而能够更早地发现问题，从而减少其影响。异常检测方法还用于从欺诈预防到网络安全的一系列其他应用。具体研究问题包括（i）如何区分不同类型的异常;（ii）开发可以扩展到高维和高频数据流的统计算法;（iii）理解新统计方法的理论属性。与BT合作。该项目属于统计和应用概率领域（计算统计，统计方法，时间序列）。

项目成果

A linear time method for the detection of point and collective anomalies

用于检测点和集体异常的线性时间方法

• DOI: 10.48550/arxiv.1806.01947
• 发表时间: 2018
• 期刊: arXiv e-prints
• 作者: Fisch Alexander T. M.

Subset Multivariate Collective And Point Anomaly Detection

子集多元集体和点异常检测

• DOI: 10.48550/arxiv.1909.01691
• 发表时间: 2019
• 期刊: arXiv e-prints
• 作者: Fisch Alexander T M