Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jan 19, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.3K

Unsupervised Anomaly Detection With LSTM Neural Networks.

Tolga Ergen, Suleyman Serdar Kozat

    IEEE Transactions on Neural Networks and Learning Systems
    |September 20, 2019
    PubMed
    Summary

    This study introduces novel anomaly detection algorithms using long short-term memory (LSTM) networks and one-class support vector machines (OC-SVM) or support vector data description (SVDD). The approach achieves high performance on variable-length time series data.

    Related Concept Videos

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Spatiotemporal Sequence Prediction With Point Processes and Self-Organizing Decision Trees.

    IEEE transactions on neural networks and learning systems·2021
    Same author

    Markovian RNN: An Adaptive Time Series Prediction Network With HMM-Based Switching for Nonstationary Environments.

    IEEE transactions on neural networks and learning systems·2021
    Same author

    Multi-Label Sentiment Analysis on 100 Languages With Dynamic Weighting for Label Imbalance.

    IEEE transactions on neural networks and learning systems·2021
    Same author

    Achieving Online Regression Performance of LSTMs With Simple RNNs.

    IEEE transactions on neural networks and learning systems·2021
    Same author

    Online Anomaly Detection With Bandwidth Optimized Hierarchical Kernel Density Estimators.

    IEEE transactions on neural networks and learning systems·2020
    Same author

    Energy-Efficient LSTM Networks for Online Learning.

    IEEE transactions on neural networks and learning systems·2019

    Area of Science:

    • Machine Learning
    • Artificial Intelligence
    • Data Science

    Background:

    • Anomaly detection is crucial for identifying unusual patterns in data.
    • Unsupervised learning frameworks are often preferred for anomaly detection due to data labeling challenges.
    • Handling variable-length data sequences remains a significant challenge in anomaly detection.

    Purpose of the Study:

    • To develop advanced anomaly detection algorithms capable of processing variable-length data sequences.
    • To integrate long short-term memory (LSTM) neural networks with one-class support vector machines (OC-SVM) and support vector data description (SVDD).
    • To jointly train and optimize the parameters of LSTM architectures and OC-SVM/SVDD algorithms.

    Main Methods:

    • Utilized LSTM networks to convert variable-length sequences into fixed-length representations.

    Related Experiment Videos

    Last Updated: Jan 19, 2026

    Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
    06:37

    Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

    Published on: December 15, 2023

    5.3K
  • Employed OC-SVM and SVDD for anomaly detection decision functions.
  • Developed novel gradient-based and quadratic programming-based methods for joint training and optimization.
  • Modified objective criteria for OC-SVM and SVDD to enable gradient-based training, proving convergence.
  • Extended the framework to semisupervised and fully supervised settings.
  • Demonstrated the generic nature of the approach by applying it to gated recurrent unit (GRU) architectures.

    • Main Results:

    • Achieved high performance in anomaly detection, particularly for time series data.
    • Demonstrated significant performance gains compared to conventional anomaly detection methods.
    • Successfully processed variable-length data sequences effectively.
    • Validated the approach's effectiveness and adaptability across different recurrent neural network architectures (LSTM and GRU).

    Conclusions:

    • The proposed joint training framework offers a powerful and flexible solution for anomaly detection in variable-length sequences.
    • The integration of deep learning (LSTM/GRU) with established machine learning algorithms (OC-SVM/SVDD) yields superior performance.
    • The method is robust, adaptable, and shows promise for various real-world applications requiring anomaly detection.