Data

Tools

Indexes

Applications

Experiments

Open Source Science

simona-anomalydetection-gnn

Non-stationary signal analysis: detrending and anomaly detection. Source codes repository of framework on detrending sequential sensor data and anomaly detection.

https://github.com/sigpro25/simona-anomalydetection-gnn

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (6.7%) to scientific vocabulary

Keywords

anomaly-detection detrending graph-neural-networks neural-networks nonstationary-time-series
Last synced: 9 months ago · JSON representation ·

Repository

Non-stationary signal analysis: detrending and anomaly detection. Source codes repository of framework on detrending sequential sensor data and anomaly detection.

Basic Info
  • Host: GitHub
  • Owner: sigpro25
  • Language: Python
  • Default Branch: main
  • Homepage:
  • Size: 189 KB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
anomaly-detection detrending graph-neural-networks neural-networks nonstationary-time-series
Created over 1 year ago · Last pushed 11 months ago
Metadata Files
Readme Citation

README.md

simona-anomalydetection-gnn

Non-stationary signal analysis: detrending and anomaly detection

This project provides a framework for removing trend and detecting anomalies in time-series signals using Graph Neural Network (GNN). It includes functionality for generating synthetic signal datasets, training models, and evaluating performance using outlier detection methods such as Local Outlier Factor (LOF).

Overview

The core objective of this project is to identify anomalies and detect trends in time-series signals. The project includes:

  • Signal Generation: Generator synthetic nonstationary signals with varying trends, noise, and anomalies (peaks).
  • Trend Estimation: Trend estiamtor of nonstationary signals.
  • Anomaly Detection: Utilization of anomaly detection on detrened signal data.
  • Model Evaluation: An evaluator the detrending and anomaly detection framework.

Files and Directories

  • main.py: Contains functions to manage training and evaluation of algorithms.
  • data/generate_dataset.py: Contains a class for generating synthetic signals with various trends, noise, and anomalies (e.g., peaks).
  • src/model/GNNModel.py: Defines the architecture of a Graph Neural Network (GNN) model.
  • src/evaluator.py, src/results_visualiser.py: Contains the classes for evaluating models and visualizing results.

Owner

  • Login: sigpro25
  • Kind: user

Citation (CITATION.cff) 

cff-version: 1.3.0
message: "If you use this software, please cite it as below."
authors:
- family-names: "Drahý"
  given-names: "Vojtěch"
  orcid: "https://orcid.org/0009-0002-5450-8814"
- family-names: "Mařík"
  given-names: "Radek"
  orcid: "https://orcid.org/0000-0003-2000-6541"
- family-names: "Kälviäinen"
  given-names: "Heikki"
  orcid: "https://orcid.org/0000-0002-0790-6847"
title: "simona-anomalydetection-gnn"
version: 1.0.0
doi: "https://doi.org/10.1007/978-3-031-95911-0_4"
date-released: 2025-01-24
url: "https://github.com/sigpro25/simona-anomalydetection-gnn"

GitHub Events

Total
  • Push event: 19
Last Year
  • Push event: 19

Dependencies

requirements.txt pypi
  • matplotlib ==3.10.0
  • numpy ==2.2.2
  • plotly ==5.24.1
  • pybaselines ==1.1.0
  • scikit_learn ==1.6.1
  • torch ==2.3.0
  • torch_geometric ==2.6.1
  • tqdm ==4.66.4