SO_MAD: SensOr Mining for Anomaly Detection in Railway Data

Rabatel, Julien; Bringay, Sandra; Poncelet, Pascal

doi:10.1007/978-3-642-03067-3_16

Cited by 19 publications

(11 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the problems is that analysis is often of scarce, incomplete or, sometimes even has missing data [40]. The weakness in building a satisfying database arises mainly from building new lines, the new materials used in railway tech, and climate and traffic density changes over the years [41].…”

Section: Comparative Evaluations and Discussionmentioning

confidence: 99%

Identification of appropriate risk analysis techniques for railway turnout systems

Dindar

Kaewunruen

2016

Journal of Risk Research

View full text Add to dashboard Cite

Where a licence is displayed above, please note the terms and conditions of the licence govern your use of this document. When citing, please reference the published version. Take down policy While the University of Birmingham exercises care and attention in making items available there are rare occasions when an item has been uploaded in error or has been deemed to be commercially or otherwise sensitive.

show abstract

Section: Comparative Evaluations and Discussionmentioning

confidence: 99%

Identification of appropriate risk analysis techniques for railway turnout systems

Dindar

Kaewunruen

2016

Journal of Risk Research

View full text Add to dashboard Cite

show abstract

“…Values that can be predicted include the usage pattern of devices. An anomaly detection method [15] is used to improvise railway maintenance. They extracted the abnormal behavioral data that was based on environmental and structural changes in data.…”

Section: Sequential Pattern Miningmentioning

confidence: 99%

A Reasonable Exploration of Data Mining Techniques in Wireless Sensor Network

Agarwal¹,

Maitrey²,

Rana³

et al. 2016

IJCA

View full text Add to dashboard Cite

Data mining in WSN is the process of extracting model and pattern that are application oriented with possible accuracy from continuous, rapid flow of data. The whole huge amount of data cannot be stored and processed immediately. That is why the mining algorithm should be fast enough to process high speed arriving data. There are many conventional data mining techniques, but they are not able to handle dynamic amount of data. It is difficult to handle WSN data. There are several challenges that it has to face in WSN. The main aim of wireless sensor networks is to transmit data in such a manner that increased lifetime of the network and energy efficient routing can be done with significant accuracy. Data mining is the process of discovering interesting patterns (or knowledge) from large amounts of data. Knowledge discovery process attains several steps and can be interactive, iterative and userdriven. Data mining techniques of wireless sensor network are different from traditional techniques. Data mining techniques can be frequent pattern mining, sequential pattern mining, clustering and classification. All these techniques can use centralized or distributed approach, even after that the focus is decided that either you can focus on application or performance of wireless sensor network. Data mining techniques that work on sensor network-based application are still facing shortcomings in existing techniques. By seeing these shortcomings and special characteristics of WSNs, there is a need for data mining technique designed for WSNs. In this paper, we are finding the difference between traditional and sensor data processing. Also comparing the different data mining techniques used in wireless sensor network where all these methods have their own processing architecture and method of sensor distribution depending upon the attributes.

show abstract

“…Traditionally, knowledge discovery (data mining) in databases (KDD) [27] is a process to extract useful knowledge from a dataset or a database. Recently, KDD techniques have been extended to the WSNs field (called knowledge discovery in WSNs (KDW) [10]), and have attracted significant attention in many application areas [2], such as anomaly detection in railway data [28], relational temporal data mining [3], [29], [30] using Allen's temporal interval logic [31], object tracking [4], [11], [32], prediction of the location (target) of a missed reported event [33], and outlier (abnormal event) detection [34]. Until recently, on the whole there has been relatively little progress in the association rules applications to WSNs.…”

Section: B Knowledge Discovery In Wsnsmentioning

confidence: 99%

Mining Associated Ranking Patterns from Wireless Sensor Networks

2016

International Conference on Engineering Technologies and Big Data Analytics (ETBDA’2016) Jan. 21-22, 2016 Bangkok (Thailand)

View full text Add to dashboard Cite

Wireless Sensor Networks (WSNs) are complex networks consisting of many sensors which can detect and collect sensed data. Valuable information can be extracted from the sensed data of WSNs using data mining techniques, which provide unlimited possibilities of applications, and have been the object of an increasing number of studies in recent years. This study applies ranking sequence mining to WSNs. For example, if a sensor detects the following scenario during a time interval: A occurs three times; B occurs twice; C occurs four times, where A, B and C are events. The information, (C(4) > A(3) > B(2)), can be abstractly expressed as a ranking sequence (C ≻ A ≻ B), where "≻" indicates to the importance of the preceding item is greater than that of the succeeding item. To the best of our knowledge, there has been no research conducted on ranking pattern mining problems with WSNs. The objective of this study is to construct a novel model to discover interesting patterns from the sensed data of WSNs.

show abstract

SO_MAD: SensOr Mining for Anomaly Detection in Railway Data

Cited by 19 publications

References 9 publications

Identification of appropriate risk analysis techniques for railway turnout systems

Identification of appropriate risk analysis techniques for railway turnout systems

A Reasonable Exploration of Data Mining Techniques in Wireless Sensor Network

Mining Associated Ranking Patterns from Wireless Sensor Networks

Contact Info

Product

Resources

About