English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper presents an automated machine learning framework designed toassist hydrologists in detecting anomalies in time series data generated bysensors in a research watershed in the northeastern United States criticalzone. The framework specifically focuses on identifying peak-pattern anomalies,which may arise from sensor malfunctions or natural phenomena. However, the useof classification methods for anomaly detection poses challenges, such as therequirement for labeled data as ground truth and the selection of the mostsuitable deep learning model for the given task and dataset. To address thesechallenges, our framework generates labeled datasets by injecting syntheticpeak patterns into synthetically generated time series data and incorporates anautomated hyperparameter optimization mechanism. This mechanism generates anoptimized model instance with the best architectural and training parametersfrom a pool of five selected models, namely Temporal Convolutional Network(TCN), InceptionTime, MiniRocket, Residual Networks (ResNet), and LongShort-Term Memory (LSTM). The selection is based on the user's preferencesregarding anomaly detection accuracy and computational cost. The frameworkemploys Time-series Generative Adversarial Networks (TimeGAN) as the syntheticdataset generator. The generated model instances are evaluated using acombination of accuracy and computational cost metrics, including training timeand memory, during the anomaly detection process. Performance evaluation of theframework was conducted using a dataset from a watershed, demonstratingconsistent selection of the most fitting model instance that satisfies theuser's preferences.

An Automated Machine Learning Approach for Detecting Anomalous Peak  Patterns in Time Series Data from a Research Watershed in the Northeastern  United States Critical Zone