Anomaly detection using Autoencoders and Deep Convolution Generative Adversarial Networks

Kopcan, Jaroslav; Škvarek, Ondrej; Klimo, Martin

doi:10.1016/j.trpro.2021.07.113

Cited by 8 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Autoencoder accepts high-dimensional input data, compress it down to the latent-space representation in the bottleneck hidden layer; the Decoder takes the latent representation of the data as an input to reconstruct the original input data. Therefore, Autoencoders have been used for Anomaly Detection tasks (for instance [64,65]), by comparing the output from the Decoder and the input to the Network and using a threshold, either manually set or learnt from the data itself. If the loss value exceeds the threshold, then the instance is categorised or classified as an anomaly.…”

Section: Antwerpmentioning

confidence: 99%

An AutoML-based approach for automatic traffic incident detection in smart cities

Gkioka,

Dominguez,

Mentzas

2024

IDT

View full text Add to dashboard Cite

In the realm of modern urban mobility, automatic incident detection is a critical element of intelligent transportation systems (ITS), since the ability to promptly identify unexpected events allows for quick implementation of preventive measures and efficient response to the situations as they arise. With the growing availability of traffic data, Machine Learning (ML) has become a vital tool for enhancing traditional incident detection methods. Automated machine-learning (AutoML) techniques present a promising solution by streamlining the machine-learning process; however the application of AutoML for incident detection has not been widely explored in scientific research In this paper, we propose and apply an AutoML-based methodology for traffic incident detection and compare it with state-ofthe-art ML approaches. Our approach integrates data preprocessing with AutoML, and uses Tree-based Pipeline Optimization Tool (TPOT) to refine the process from raw data to prediction. We have tested the efficiency of our approach in two major European cities, Athens and Antwerp. Finally, we present the limitations of our work and outline recommendations for application of AutoML in the incident detection task and potentially in other domains.

show abstract

Section: Antwerpmentioning

confidence: 99%

An AutoML-based approach for automatic traffic incident detection in smart cities

Gkioka,

Dominguez,

Mentzas

2024

IDT

View full text Add to dashboard Cite

show abstract

“…A three-layer neural network and backpropagation algorithm [27] were proposed to classify cracks, joint displacement, cross-sectional area reduction and fall off on the surface of underground sewage pipes, and the recognition accuracy reached 98.2%. A comprehensive crack imaging system based on artificial neural network was proposed [28], and the crack type classification accuracy reached 95.2% on the artificially made pavement crack image dataset. A backpropagation neural network is proposed to identify cracks on the surface of concrete structures [29].…”

Section: Related Workmentioning

confidence: 99%

Research on Damage Detection of Civil Structures Based on Machine Learning of Multiple Vegetation Index Time Series

2024

Teh. vjesn.

View full text Add to dashboard Cite

On the basis of analyzing the natural frequency of the structure, the identification quantity of each process is constructed with modal parameters and input into the machine learning as characteristic parameters to realize the damage identification. By extracting the median curve of vegetation index time series after 5G filtering in the damaged area of typical civil structures, and comparing it with the actual growth curve of crops in the area, the vegetation index time series monitoring model was constructed, and 10 was selected as the best threshold. The accuracy of the result is verified, and the iteration time is 0.18 hours. A damage detection method based on machine learning is proposed. Good prediction results are obtained for three common surface damage of concrete cracks, spalling and exposed steel bars, which verify the ability of this method to accurately identify and detect structural surface damage at pixel level.

show abstract

“…The authors in [26] used GAN networks for detection. The combination of Autoencoders and Deep Convolution GAN Networks for determining anomalies in IP flow is discussed in [27]. The GAN Network with two Discriminators is used in [26,28].…”

Section: Related Workmentioning

confidence: 99%

Machine Recognition of DDoS Attacks Using Statistical Parameters

Smiesko,

Segec,

Kontsek

2023

Mathematics

View full text Add to dashboard Cite

As part of the research in the recently ended project SANET II, we were trying to create a new machine-learning system without a teacher. This system was designed to recognize DDoS attacks in real time, based on adaptation to real-time arbitrary traffic and with the ability to be embedded into the hardware implementation of network probes. The reason for considering this goal was our hands-on experience with the high-speed SANET network, which interconnects Slovak universities and high schools and also provides a connection to the Internet. Similar to any other public-facing infrastructure, it is often the target of DDoS attacks. In this article, we are extending our previous research, mainly by dealing with the use of various statistical parameters for DDoS attack detection. We tested the coefficients of Variation, Kurtosis, Skewness, Autoregression, Correlation, Hurst exponent, and Kullback–Leibler Divergence estimates on traffic captures of different types of DDoS attacks. For early machine recognition of the attack, we have proposed several detection functions that use the response of the investigated statistical parameters to the start of a DDoS attack. The proposed detection methods are easily implementable for monitoring actual IP traffic.

show abstract

Anomaly detection using Autoencoders and Deep Convolution Generative Adversarial Networks

Cited by 8 publications

References 11 publications

An AutoML-based approach for automatic traffic incident detection in smart cities

An AutoML-based approach for automatic traffic incident detection in smart cities

Research on Damage Detection of Civil Structures Based on Machine Learning of Multiple Vegetation Index Time Series

Machine Recognition of DDoS Attacks Using Statistical Parameters

Contact Info

Product

Resources

About