Outlier detection using neighborhood rank difference

Bhattacharya, Gautam; Ghosh, Koushik; Chowdhury, Ananda S.

doi:10.1016/j.patrec.2015.04.004

Cited by 46 publications

(17 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A better quantification of the actual distribution of the scores via, e.g., GMM, Kernel Density Estimation (KDE) [27], or Neighborhood Rank Difference (NRD) [8] rather than assuming multi-variate normality could improve monitoring performance. However, these techniques typically assume that a high number of data points-in this case: normal batches-is available.…”

Section: Discussionmentioning

confidence: 99%

An extensive reference dataset for fault detection and identification in batch processes

Impe¹,

Gins²

2015

Chemometrics and Intelligent Laboratory Systems

View full text Add to dashboard Cite

Close process monitoring (i.e., detection and identification of disturbances) is important to achieve high process efficiency and safety. The Tennessee Eastman process is an extensive benchmark dataset for fault detection and identification, but it is only representative for continuous processes because it does not contain the inherent nonstationarity that complicates monitoring of batch processes. Nevertheless, batch processes also play an important role in many types of industry. This paper therefore presents an extensive reference dataset for benchmarking data-driven methodologies for fault detection and identification in batch processes. The original Pensim model [10] is expanded with sensor noise. By changing the properties of the initial conditions and/or model parameters, four subsets of different complexity are generated, each containing 400 batches with normal operation. To correctly assess the fault detection and identification in batch processes, 15 faults are simulated with various amplitudes and onset times for a total of 22,200 faulty batches for each subset, or 90,400 batches in total. Analysis of the data indicates that the presented types of process faults and their various amplitudes in each of the four subsets present a suitable benchmark for fault detection and identification in batch processes. The dataset is freely available at http://cit.kuleuven.be/biotec/batchbenchmark.

show abstract

Section: Discussionmentioning

confidence: 99%

An extensive reference dataset for fault detection and identification in batch processes

Impe¹,

Gins²

2015

Chemometrics and Intelligent Laboratory Systems

View full text Add to dashboard Cite

show abstract

“…To address the various issues over the choice of appropriate number of nearest neighbors with improvement in the classification accuracy in machine learning and pattern recognition domain was presented in [15][16][17]21]. While issues related to the size of training sample set and its impact on classification performance using nearest neighbor rule was presented in [18].…”

Section: Related Workmentioning

confidence: 99%

Retracted: A Novel Voting Mathematical Rule Classification for Image Recognition

Abbasi

Shahriari

Nemati

2016

Computational Science and Its Applications – ICCSA 2016

View full text Add to dashboard Cite

Abstract. In machine learning, the accuracy of the system depends upon classification result. Classification accuracy plays an imperative role in various domains. Non-parametric classifier like K-Nearest Neighbor (KNN) is the most widely used classifier for pattern analysis. Besides its easiness, simplicity and effectiveness characteristics, the main problem associated with KNN classifier is the selection of number of nearest neighbors i.e. 'k' for computation. At present it is hard to find the optimal value of 'k' using any statistical algorithm, which gives perfect accuracy in terms of low misclassification error rate.Motivated by prescribed problem, a new sample space reduction weighted voting mathematical rule (AVNM) is proposed for classification in machine learning. Proposed AVNM rule is also non-parametric in nature like KNN. AVNM uses the weighted voting mechanism with sample space reduction to learn and examine the predicted class label for unidentified sample. AVNM is free from any initial selection of predefined variable and neighbor selection as found in KNN algorithm. Proposed classifier also reduces the effect of outliers.To verify the performance of the proposed AVNM classifier, experiments are made on 10 standard dataset taken from UCI database and one manually created dataset. Experimental result shows that the proposed AVNM rule outperforms KNN classifier and its variants. Experimentation results based on confusion matrix accuracy parameter proves higher accuracy value with AVNM rule.Proposed AVNM rule is based on sample space reduction mechanism for identification of optimal number of nearest neighbor selections. AVNM results in better classification accuracy and minimum error rate as compared with state-of-art algorithm, KNN and its variants. Proposed rule automates the selection of nearest neighbor selection and improves classification rate for UCI dataset and manual created dataset.

show abstract

“…A novel method has been proposed using neighbourhood rank difference for the detection of outliers. The results of experiments are recorded for real and synthetic datasets with different dimensions [7]. This paper proposes a framework for detecting outliers in evolving data streams.…”

Section: Literature Reviewmentioning

confidence: 99%

Outlier Detection using Clustering Techniques

Srividya¹,

Mohanavalli²,

Sripriya³

et al. 2018

IJET

View full text Add to dashboard Cite

An outlier is nothing but a pattern that is different compared to the other existing patterns in a particular dataset. In some applications it is very important to understand and identify outliers. Detecting outlier is of major importance in many of the fields like cybersecurity, machine learning, finance, healthcare, etc., A clustering based method is proposed to detect outliers using different algorithms like k means, PAM, Clara, DBScan and LOF on different data sets like breast cancer, heart diseases, multi shaped datasets. This work aims to identify the best suitable method to detect the outliners accurately.

show abstract

Outlier detection using neighborhood rank difference

Cited by 46 publications

References 13 publications

An extensive reference dataset for fault detection and identification in batch processes

An extensive reference dataset for fault detection and identification in batch processes

Retracted: A Novel Voting Mathematical Rule Classification for Image Recognition

Outlier Detection using Clustering Techniques

Contact Info

Product

Resources

About