Harnessing Feature Clustering For Enhanced Anomaly Detection With Variational Autoencoder And Dynamic Threshold
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914869876359168 |
|---|---|
| author | Ale, Tolulope Schlegel, Nicole-Jeanne Janeja, Vandana P. |
| author_facet | Ale, Tolulope Schlegel, Nicole-Jeanne Janeja, Vandana P. |
| contents | We introduce an anomaly detection method for multivariate time series data with the aim of identifying critical periods and features influencing extreme climate events like snowmelt in the Arctic. This method leverages the Variational Autoencoder (VAE) integrated with dynamic thresholding and correlation-based feature clustering. This framework enhances the VAE's ability to identify localized dependencies and learn the temporal relationships in climate data, thereby improving the detection of anomalies as demonstrated by its higher F1-score on benchmark datasets. The study's main contributions include the development of a robust anomaly detection method, improving feature representation within VAEs through clustering, and creating a dynamic threshold algorithm for localized anomaly detection. This method offers explainability of climate anomalies across different regions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_10042 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Harnessing Feature Clustering For Enhanced Anomaly Detection With Variational Autoencoder And Dynamic Threshold Ale, Tolulope Schlegel, Nicole-Jeanne Janeja, Vandana P. Machine Learning Information Retrieval We introduce an anomaly detection method for multivariate time series data with the aim of identifying critical periods and features influencing extreme climate events like snowmelt in the Arctic. This method leverages the Variational Autoencoder (VAE) integrated with dynamic thresholding and correlation-based feature clustering. This framework enhances the VAE's ability to identify localized dependencies and learn the temporal relationships in climate data, thereby improving the detection of anomalies as demonstrated by its higher F1-score on benchmark datasets. The study's main contributions include the development of a robust anomaly detection method, improving feature representation within VAEs through clustering, and creating a dynamic threshold algorithm for localized anomaly detection. This method offers explainability of climate anomalies across different regions. |
| title | Harnessing Feature Clustering For Enhanced Anomaly Detection With Variational Autoencoder And Dynamic Threshold |
| topic | Machine Learning Information Retrieval |
| url | https://arxiv.org/abs/2407.10042 |