A simple method for unsupervised anomaly detection: An application to Web time series data

We propose a simple anomaly detection method that is applicable to unlabeled time series data and is sufficiently tractable, even for non-technical entities, by using the density ratio estimation based on the state space model. Our detection rule is based on the ratio of log-likelihoods estimated by...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	PloS one Ročník 17; číslo 1; s. e0262463
Hlavní autoři:	Yoshihara, Keisuke, Takahashi, Kei
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	United States Public Library of Science 11.01.2022 Public Library of Science (PLoS)
Témata:	Advertising agencies Advertising campaigns Algorithms Anomalies Benchmarks Computer and Information Sciences Datasets Datasets as Topic - statistics & numerical data Density ratio Electronic commerce Electronic data processing Engineering and Technology Humans Insurance Internet Marketing Methods Neural Networks, Computer Normal distribution Online advertising Performance evaluation Physical Sciences Research and Analysis Methods Science Policy Search Engine Social Sciences State space models Statistics as Topic - methods Time Factors Time series Time-series analysis Web sites Japan
ISSN:	1932-6203, 1932-6203
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	We propose a simple anomaly detection method that is applicable to unlabeled time series data and is sufficiently tractable, even for non-technical entities, by using the density ratio estimation based on the state space model. Our detection rule is based on the ratio of log-likelihoods estimated by the dynamic linear model, i.e. the ratio of log-likelihood in our model to that in an over-dispersed model that we will call the NULL model. Using the Yahoo S5 data set and the Numenta Anomaly Benchmark data set, publicly available and commonly used benchmark data sets, we find that our method achieves better or comparable performance compared to the existing methods. The result implies that it is essential in time series anomaly detection to incorporate the specific information on time series data into the model. In addition, we apply the proposed method to unlabeled Web time series data, specifically, daily page view and average session duration data on an electronic commerce site that deals in insurance goods to show the applicability of our method to unlabeled real-world data. We find that the increase in page view caused by e-mail newsletter deliveries is less likely to contribute to completing an insurance contract. The result also suggests the importance of the simultaneous monitoring of more than one time series.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 Competing Interests: The authors have declared that no competing interests exist.
ISSN:	1932-6203 1932-6203
DOI:	10.1371/journal.pone.0262463