Combining Unsupervised Anomaly Detection and Neural Networks for Driver Identification

Tanprasert, Thitaree; Saiprasert, Chalermpol; Thajchayapong, Suttipong

doi:10.1155/2017/6057830

Cited by 30 publications

(19 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Model tiga lapis telah digunakan untuk melatih dan mempelajari berbagai atribut neuron tersembunyi dalam jaringan. Penelitian [11] mengusulkan algoritma untuk identifikasi driver real-time menggunakan kombinasi deteksi anomali tanpa pengawasan dan jaringan saraf. Algoritma yang diusulkan menggunakan sinyal nonfisiologis sebagai input, yaitu sinyal perilaku mengemudi dari sensor inersia (mis.…”

Section: Pendahuluanunclassified

Pengembangan Model Jaringan Syaraf Tiruan untuk Mendeteksi Anomali Satelit LAPAN-TUBSAT

Herawan¹

2019

JEPIN

View full text Add to dashboard Cite

Kasus satelit mengalami anomali seringkali di temukan pada satelit-satelit yang beroperasi pada orbit polar. Namun permasalahan yang muncul adalah kondisi satelit sering berubah-ubah sehingga operator belum bisa mengantisipasi kondisi tersebut. Oleh sebab itu, model deteksi kondisi satelit dapat berperan sebagai early warning operator satelit untuk mempersiapkan strategi yang berkaitan dengan kebijakan preventif terkait pencegahan ketika satelit mengalami kondisi tidak normal. Tujuan penelitian ini adalah menerapkan Jaringan Syaraf Tiruan (JST) bakcpropagation dalam mendeteksi kondisi anomali pada satelit LAPAN-TUBSAT, serta mengetahui tingkat akurasi dari proses deteksi tersebut sehingga diperoleh parameter dan arsitektur jaringan JST terbaik. Proses pembelajaran dan pengujian JST menggunakan data kejadian anomali tahun 2009 sampai 2014. Arsitektur JST yang digunakan adalah jumlah node input 4, dua hidden layer, jumlah node lapisan tersembunyi (hidden neuron) divariasikan pada nilai 5, 10, 15 dan 20. Parameter yang diberikan pada proses pembelajaran antara lain adalah fungsi aktivasi, toleransi galat, jumlah epoch maksimal dan variasi nilai laju pembelajaran (learning rate). Empat parameter input yang digunakan yakni elektron (mep0e1), proton (mep0p1), indeks Kp serta indeks Dst. Hasil penelitian menunjukkan bahwa arsitektur jaringan syaraf terbaik dihasilkan oleh jaringan dengan jumlah input node empat, hidden neuron 20 dan 10, nilai learning rate sebesar 0.05 dengan 306 epoch, menghasilan rata-rata akurasi sebesar 98.13%, serta nilai precision dan recall sebesar 98.21% dan 94.81%.

show abstract

Section: Pendahuluanunclassified

Pengembangan Model Jaringan Syaraf Tiruan untuk Mendeteksi Anomali Satelit LAPAN-TUBSAT

Herawan¹

2019

JEPIN

View full text Add to dashboard Cite

show abstract

“…Since the labelled data, which is required for supervised anomaly detection, is often not available or in other words, collecting sufficient anomolous samples is infeasible in most of the cases, many researchers have tried to detect anomalies without labelled data. However, they were forced to take advantage of supervised learning along with unsupervised for other steps of their algorithms such as pattern recognition to achieve better diagnosis [29], [30].…”

Section: Introductionmentioning

confidence: 99%

Developing a Data-Driven Unsupervised Pattern Recognition Approach for Sensor Signal Anomaly Detection

Iranmehr¹,

Ferreira²,

Böhnert³

et al. 2021

Preprint

View full text Add to dashboard Cite

Coming up with a system for early detection of machine damages and failures is one of the important challenges in the industrial maintenance procedure to avoid additional costs and downtimes. To approach this goal, this paper uses the signal gathered by a sensing system which employed a spintropic sensor to measure the magnetic field around the machine which somehow shows the machine's behaviour. Using this signal and focusing on analysing and processing the signal, this paper develops a data-driven method to recognize signal patterns and subsequently detects anomalies. A challenging task that we succeeded to overcome in this paper is recognizing relevant signal patterns without having any prior knowledge. An algorithm designed for this task is therefore completely unsupervised which makes it consistent and suitable to apply it for the signals gathered for other types of machines. Using both frequency and time domain information, the proposed algorithm, which utilizes signal processing and machine learning techniques, is able to efficiently identify relevant signal patterns. Clustering results on the real data gathered by the aforementioned sensor have shown the high accuracy of 99.38% in recognizing patterns. Furthermore, an anomaly score measure is used and according to its distribution, anomalies are detected appropriately. <br>

show abstract

“…Typically, different from the MNL model, machine learning methods, particularly deep learning models using a data-oriented approach, are becoming increasingly prominent in numerous research fields, including transportation [2][3][4]. Deep neural networks (DNNs) are mathematical tools that are loosely inspired by the functional aspects of biological neural systems.…”

Section: Introductionmentioning

confidence: 99%

A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information

Zhang

Wang

et al. 2020

Journal of Advanced Transportation

View full text Add to dashboard Cite

Discrete choice modeling of travel modes is an essential part of traffic planning and management. Thus far, this field has been dominated by multinomial logit (MNL) models with a linear utility specification. However, deep neural networks (DNNs), owing to their powerful capacity of nonlinear fitting, are now rapidly replacing these models. This is because, by using DNNs, mode choice can be assimilated with the classification problems within the machine learning community. This article proposes a newly designed DNN framework for traffic mode choice in the style of two hidden layers. First, a local-connected layer automatically extracts an effective utility specification from the available data, and then, a fully connected layer augments the feature representation. Validated by a practical city-wide multimodal traffic dataset in Beijing, our model significantly outperforms the random utility models and simple fully connected neural network in terms of the prediction accuracy. Besides the comparison of the predictive power, we also present the interpretability of the proposed model.

show abstract

Combining Unsupervised Anomaly Detection and Neural Networks for Driver Identification

Cited by 30 publications

References 21 publications

Pengembangan Model Jaringan Syaraf Tiruan untuk Mendeteksi Anomali Satelit LAPAN-TUBSAT

Pengembangan Model Jaringan Syaraf Tiruan untuk Mendeteksi Anomali Satelit LAPAN-TUBSAT

Developing a Data-Driven Unsupervised Pattern Recognition Approach for Sensor Signal Anomaly Detection

A Customized Deep Neural Network Approach to Investigate Travel Mode Choice with Interpretable Utility Information

Contact Info

Product

Resources

About