KFU. Card publication. How to quickly find anomalies in number series using the Hampel method

KAZAN
FEDERAL UNIVERSITY

- RUS
- 中文
- ES

HOW TO QUICKLY FIND ANOMALIES IN NUMBER SERIES USING THE HAMPEL METHOD

Form of presentation

Other electronic educational resources

Year of publication

2023

Язык

английский

Authors, employees of KFU

Gilmullin Mansur Fayzrakhmanovich, author

Other authors

Gilmullin Timur Mansurovich, author

Bibliographic description in the original language

T. Gilmullin, M. Gilmullin. How to quickly find anomalies in number series using the Hampel method. - https://forworktests.blogspot.com/2023/01/how-to-quickly-find-anomalies-in-number.html

Annotation

In practice, there are problems for the solution of which it is required to find anomalies in the numerical series. Such tasks are found in various areas: in data science, machine learning, cybersecurity, algorithmic trading, etc. The article shows examples of how to quickly and efficiently find anomalies in numerical series using the modified Hampel method (Hampel F.R.) using sliding windows. To filter a number series for the presence of anomalies in it, it is proposed to use the Python implementation of the HampelFilter() function. The use of the created functions is possible, among other things, using the example of the problem of searching for outliers in stock data.

Keywords

anomaly, cybersecurity, data science, machine learning, python, outliers, series, filtering, Hampel, Median Absolute Deviation, sliding windows

URL

https://forworktests.blogspot.com/2023/01/how-to-quickly-find-anomalies-in-number.html

Please use this ID to quote from or refer to the card

https://repository.kpfu.ru/eng/?p_id=275143&p_lang=2

Full metadata record

Field DC	Value	Language
dc.contributor.author	Gilmullin Mansur Fayzrakhmanovich	ru_RU
dc.contributor.author	Gilmullin Timur Mansurovich	ru_RU
dc.date.accessioned	2023-01-01T00:00:00Z	ru_RU
dc.date.available	2023-01-01T00:00:00Z	ru_RU
dc.date.issued	2023	ru_RU
dc.identifier.citation	T. Gilmullin, M. Gilmullin. How to quickly find anomalies in number series using the Hampel method. - https://forworktests.blogspot.com/2023/01/how-to-quickly-find-anomalies-in-number.html	ru_RU
dc.identifier.uri	https://repository.kpfu.ru/eng/?p_id=275143&p_lang=2	ru_RU
dc.description.abstract	In practice, there are problems for the solution of which it is required to find anomalies in the numerical series. Such tasks are found in various areas: in data science, machine learning, cybersecurity, algorithmic trading, etc. The article shows examples of how to quickly and efficiently find anomalies in numerical series using the modified Hampel method (Hampel F.R.) using sliding windows. To filter a number series for the presence of anomalies in it, it is proposed to use the Python implementation of the HampelFilter() function. The use of the created functions is possible, among other things, using the example of the problem of searching for outliers in stock data.	ru_RU
dc.language.iso	ru	ru_RU
dc.subject	anomaly	ru_RU
dc.subject	cybersecurity	ru_RU
dc.subject	data science	ru_RU
dc.subject	machine learning	ru_RU
dc.subject	python	ru_RU
dc.subject	outliers	ru_RU
dc.subject	series	ru_RU
dc.subject	filtering	ru_RU
dc.subject	Hampel	ru_RU
dc.subject	Median Absolute Deviation	ru_RU
dc.subject	sliding windows	ru_RU
dc.subject	anomaly	ru_RU
dc.subject	cybersecurity	ru_RU
dc.subject	data science	ru_RU
dc.subject	machine learning	ru_RU
dc.subject	python	ru_RU
dc.subject	outliers	ru_RU
dc.subject	series	ru_RU
dc.subject	filtering	ru_RU
dc.subject	Hampel	ru_RU
dc.subject	Median Absolute Deviation	ru_RU
dc.subject	sliding windows	ru_RU
dc.title	How to quickly find anomalies in number series using the Hampel method	ru_RU
dc.type	Other electronic educational resources	ru_RU