An approach to spacecraft anomaly detection problem using kernel feature space

Fujimaki, Ryohei; Yairi, Takehisa; Machida, Kazuo

doi:10.1145/1081870.1081917

Cited by 181 publications

(83 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some of them decompose space to normal, anomaly and noise subspaces. The anomalies can be then detected in anomaly subspace [13]. -Classification approaches: In this case the problem is posed as the identification of which categories an observation belongs to.…”

Section: Anomaly Detectionmentioning

confidence: 99%

Combining Support Vector Machines and Segmentation Algorithms for Efficient Anomaly Detection: A Petroleum Industry Application

Martí

Sánchez-Pi²,

Molina

et al. 2014

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Abstract. Anomaly detection is the problem of finding patterns in data that do not conform to expected behavior. Similarly, when patterns are numerically distant from the rest of sample, anomalies are indicated as outliers. Anomaly detection had recently attracted the attention of the research community for real-world applications. The petroleum industry is one of the application contexts where these problems are present. The correct detection of such types of unusual information empowers the decision maker with the capacity to act on the system in order to correctly avoid, correct, or react to the situations associated with them. In that sense, heavy extraction machines for pumping and generation operations like turbomachines are intensively monitored by hundreds of sensors each that send measurements with a high frequency for damage prevention. For dealing with this and with the lack of labeled data, in this paper we propose a combination of a fast and high quality segmentation algorithm with a one-class support vector machine approach for efficient anomaly detection in turbomachines. As result we perform empirical studies comparing our approach to other methods applied to benchmark problems and a real-life application related to oil platform turbomachinery anomaly detection.

show abstract

Section: Anomaly Detectionmentioning

confidence: 99%

Combining Support Vector Machines and Segmentation Algorithms for Efficient Anomaly Detection: A Petroleum Industry Application

Martí

Sánchez-Pi²,

Molina

et al. 2014

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

show abstract

“…Fujimaki et al [13] proposed the use of vMF distribution for spacecraft outliers detection. Both of these two papers use a single vMF distribution and compute the angular difference of two vectors to determine outliers.…”

Section: Directional Distributionmentioning

confidence: 99%

Mining Outlier Participants: Insights Using Directional Distributions in Latent Models

Surian

Chawla²

2013

Advanced Information Systems Engineering

View full text Add to dashboard Cite

Abstract. In this paper we will propose a new probabilistic topic model to score the expertise of participants on the projects that they contribute to based on their previous experience. Based on each participant's score, we rank participants and define those who have the lowest scores as outlier participants. Since the focus of our study is on outliers, we name the model as Mining Outlier Participants from Projects (MOPP) model. MOPP is a topic model that is based on directional distributions which are particularly suitable for outlier detection in high-dimensional spaces. Extensive experiments on both synthetic and real data sets have shown that MOPP gives better results on both topic modeling and outlier detection tasks.

show abstract

“…Outlier detection has been extensively studied these past years, since it has a wide range of applications, such as fraud detection for credit card [1], cyber security [4] or safety of critical systems [8]. Those fields of application rely on methods to find patterns which deviate significantly from a well-defined notion of normality.…”

Section: Related Workmentioning

confidence: 99%

“…Actually, outlier detection aims to find records that deviate significantly from a well-defined notion of normality. It has a wide range of applications, such as fraud detection for credit card [1], health care, cyber security [4] or safety of critical systems [8]. However, the main drawback of detecting intrusions by means of anomaly (outliers) detection is the high rate of false alarms since an alarm can be triggered because of a new kind of usages that has never been seen before (and is thus considered as abnormal).…”

Section: Introductionmentioning

confidence: 99%

Data Mining for Intrusion Detection: From Outliers to True Intrusions

Singh

Masséglia

Fiot

et al. 2009

Advances in Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Abstract. Data mining for intrusion detection can be divided into several sub-topics, among which unsupervised clustering has controversial properties. Unsupervised clustering for intrusion detection aims to i) group behaviors together depending on their similarity and ii) detect groups containing only one (or very few) behaviour. Such isolated behaviours are then considered as deviating from a model of normality and are therefore considered as malicious. Obviously, all atypical behaviours are not attacks or intrusion attempts. Hence, this is the limits of unsupervised clustering for intrusion detection. In this paper, we consider to add a new feature to such isolated behaviours before they can be considered as malicious. This feature is based on their possible repetition from one information system to another. We propose a new outlier mining principle and validate it through a set of experiments.

show abstract

An approach to spacecraft anomaly detection problem using kernel feature space

Cited by 181 publications

References 7 publications

Combining Support Vector Machines and Segmentation Algorithms for Efficient Anomaly Detection: A Petroleum Industry Application

Combining Support Vector Machines and Segmentation Algorithms for Efficient Anomaly Detection: A Petroleum Industry Application

Mining Outlier Participants: Insights Using Directional Distributions in Latent Models

Data Mining for Intrusion Detection: From Outliers to True Intrusions

Contact Info

Product

Resources

About