Anomaly detection in multi-class time series

Abstract For modern operation and maintenance systems, they are usually required to monitor multiple types and large quantities of machine’s key performance indicators (KPIs) at the same time with limited resources. In this paper, to tackle these problems, we propose a highly compatible time series anomaly detection model based on K-means clustering algorithm with a new Wavelet Feature Distance (WFD). Our work is inspired by some ideas from image processing and signal processing domain. Our model detects abnormalities in the time series datasets which are first clustered by K-means to boost the accuracy. Our experiments show significant accuracy improvements compared with traditional algorithms, and excellent compatibilities and operating efficiencies compared with algorithms based on deep learning.

Download Full-text

Hidden Markov Model Based Time-Series Images Clustering Algorithm and its Application in Sports Image Processing

Recent Advances in Electrical & Electronic Engineering (Formerly Recent Patents on Electrical & Electronic Engineering) ◽

10.2174/235209650901160419003807 ◽

2016 ◽

Vol 9 (1) ◽

pp. 44-52

Author(s):

Yazhi Ding ◽

Rong Ma

Keyword(s):

Image Processing ◽

Time Series ◽

Markov Model ◽

Hidden Markov Model ◽

Clustering Algorithm ◽

Hidden Markov ◽

Model Based ◽

Time Series Images

Download Full-text

Semi-supervised Time Series Anomaly Detection Model Based on LSTM Autoencoder

Communications in Computer and Information Science - Big Data and Security ◽

10.1007/978-981-16-3150-4_4 ◽

2021 ◽

pp. 41-53

Author(s):

Hui Xiao ◽

Donghai Guan ◽

Rui Zhao ◽

Weiwei Yuan ◽

Yaofeng Tu ◽

...

Keyword(s):

Time Series ◽

Anomaly Detection ◽

Detection Model ◽

Model Based

Download Full-text

Unsupervised event classification with graphs on classical and photonic quantum computers

Journal of High Energy Physics ◽

10.1007/jhep08(2021)170 ◽

2021 ◽

Vol 2021 (8) ◽

Author(s):

Andrew Blance ◽

Michael Spannowsky

Keyword(s):

Anomaly Detection ◽

Clustering Algorithm ◽

High Energy ◽

New Physics ◽

Continuous Variable ◽

Quantum Computers ◽

Detection Model ◽

Novel Method ◽

Formula Complexity ◽

Boson Sampling

Abstract Photonic Quantum Computers provide several benefits over the discrete qubit-based paradigm of quantum computing. By using the power of continuous-variable computing we build an anomaly detection model to use on searches for New Physics. Our model uses Gaussian Boson Sampling, a #P-hard problem and thus not efficiently accessible to classical devices. This is used to create feature vectors from graph data, a natural format for representing data of high-energy collision events. A simple K-means clustering algorithm is used to provide a baseline method of classification. We then present a novel method of anomaly detection, combining the use of Gaussian Boson Sampling and a quantum extension to K-means known as Q-means. This is found to give equivalent results compared to the classical clustering version while also reducing the $$ \mathcal{O} $$ O complexity, with respect to the sample’s feature-vector length, from $$ \mathcal{O}(N) $$ O N to $$ \mathcal{O}\left(\log (N)\right) $$ O log N .

Download Full-text

Ensemble Neuroevolution-Based Approach for Multivariate Time Series Anomaly Detection

Entropy ◽

10.3390/e23111466 ◽

2021 ◽

Vol 23 (11) ◽

pp. 1466

Author(s):

Kamil Faber ◽

Marcin Pietron ◽

Dominik Zurek

Keyword(s):

Time Series ◽

Deep Learning ◽

Anomaly Detection ◽

Multivariate Time Series ◽

Window Size ◽

Learning Models ◽

Failure Prevention ◽

Detection Model ◽

Repair Costs ◽

Fully Automatic

Multivariate time series anomaly detection is a widespread problem in the field of failure prevention. Fast prevention means lower repair costs and losses. The amount of sensors in novel industry systems makes the anomaly detection process quite difficult for humans. Algorithms that automate the process of detecting anomalies are crucial in modern failure prevention systems. Therefore, many machine learning models have been designed to address this problem. Mostly, they are autoencoder-based architectures with some generative adversarial elements. This work shows a framework that incorporates neuroevolution methods to boost the anomaly detection scores of new and already known models. The presented approach adapts evolution strategies for evolving an ensemble model, in which every single model works on a subgroup of data sensors. The next goal of neuroevolution is to optimize the architecture and hyperparameters such as the window size, the number of layers, and the layer depths. The proposed framework shows that it is possible to boost most anomaly detection deep learning models in a reasonable time and a fully automated mode. We ran tests on the SWAT and WADI datasets. To the best of our knowledge, this is the first approach in which an ensemble deep learning anomaly detection model is built in a fully automatic way using a neuroevolution strategy.

Download Full-text

A Modified DBSCAN Algorithm for Anomaly Detection in Time-series Data with Seasonality

The International Arab Journal of Information Technology ◽

10.34028/iajit/19/1/3 ◽

2022 ◽

Vol 19 (1) ◽

Author(s):

Praphula Jain ◽

Mani Shankar Bajpai ◽

Rajendra Pamula

Keyword(s):

Time Series ◽

Anomaly Detection ◽

Clustering Algorithm ◽

Time Series Data ◽

Spatial Clustering ◽

Series Data ◽

Practical Applications ◽

Dbscan Algorithm ◽

Local Anomalies ◽

Seasonal Data

Anomaly detection concerns identifying anomalous observations or patterns that are a deviation from the dataset's expected behaviour. The detection of anomalies has significant and practical applications in several industrial domains such as public health, finance, Information Technology (IT), security, medical, energy, and climate studies. Density-Based Spatial Clustering of Applications with Noise (DBSCAN) Algorithm is a density-based clustering algorithm with the capability of identifying anomalous data. In this paper, a modified DBSCAN algorithm is proposed for anomaly detection in time-series data with seasonality. For experimental evaluation, a monthly temperature dataset was employed and the analysis set forth the advantages of the modified DBSCAN over the standard DBSCAN algorithm for the seasonal datasets. From the result analysis, we may conclude that DBSCAN is used for finding the anomalies in a dataset but fails to find local anomalies in seasonal data. The proposed Modified DBSCAN approach helps to find both the global and local anomalies from the seasonal data. Using normal DBSCAN we are able to get 19 (2.16%) anomaly points. While using the modified approach for DBSCAN we are able to get 42 (4.79%) anomaly points. In comparison we can say that we are able to get 2.11% more anomalies using the modified DBSCAN approach. Hence, the proposed Modified DBSCAN algorithm outperforms in comparison with the DBSCAN algorithm to find local anomalies.

Download Full-text

Anomaly Detection Based on Mining Six Local Data Features and BP Neural Network

Symmetry ◽

10.3390/sym11040571 ◽

2019 ◽

Vol 11 (4) ◽

pp. 571 ◽

Cited By ~ 1

Author(s):

Zhang ◽

Zhu ◽

Li ◽

Wang ◽

Guo

Keyword(s):

Neural Network ◽

Time Series ◽

Anomaly Detection ◽

Bp Neural Network ◽

Back Propagation ◽

Local Data ◽

Detection Model ◽

Proposed Model ◽

Comparison Results ◽

Local Monotonicity

Key performance indicators (KPIs) are time series with the format of (timestamp, value). The accuracy of KPIs anomaly detection is far beyond our initial expectations sometimes. The reasons include the unbalanced distribution between the normal data and the anomalies as well as the existence of many different types of the KPIs data curves. In this paper, we propose a new anomaly detection model based on mining six local data features as the input of back-propagation (BP) neural network. By means of vectorization description on a normalized dataset innovatively, the local geometric characteristics of one time series curve could be well described in a precise mathematical way. Differing from some traditional statistics data characteristics describing the entire variation situation of one sequence, the six mined local data features give a subtle insight of local dynamics by describing the local monotonicity, the local convexity/concavity, the local inflection property and peaks distribution of one KPI time series. In order to demonstrate the validity of the proposed model, we applied our method on 14 classical KPIs time series datasets. Numerical results show that the new given scheme achieves an average F1-score over 90%. Comparison results show that the proposed model detects the anomaly more precisely.

Download Full-text