Pattern Classification: A Unified View of Statistical and Neural Approaches.

McLachlan, Geoff; Schürmann, Jürgen

doi:10.2307/2534033

Cited by 52 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Euclidean distance, Mahalanobis distance, Kullback-Leibler distance, and Bayesian distance are all common distance approaches. Artes et al [30] and Schurmann [31] investigated the use of cluster analysis methodologies. Goumas et al [32] and Lou and Loparo [33] are two papers that investigate the usage of distance measures for fault diagnosis.…”

Section: Diagnosticsmentioning

confidence: 99%

The Discriminants of Long and Short Duration Failures in Fulfillment Sortation Equipment: A Machine Learning Approach

Mutemi

Bação

2023

Journal of Engineering

View full text Add to dashboard Cite

Due to the difficulties inherent in diagnostics and prognostics, maintaining machine health remains a substantial issue in industrial production. Current approaches rely substantially on human engagement, making them costly and unsustainable, especially in high-volume industrial complexes like fulfillment centers. The length of time that fulfillment center equipment failures last is particularly important because it affects operational costs dramatically. A machine learning approach for identifying long and short equipment failures is presented using historical equipment failure and fault data. Under a variety of hyperparameter configurations, we test and compare the outcomes of eight different machine learning classification algorithms, seven individual classifiers, and a stacked ensemble. The gradient boosting classifier (GBC) produces state-of-the-art results in this setting, with precision of 0.76, recall of 0.82, and false positive rate (FPR) of 0.002. This model has since been applied successfully to automate the detection of long- and short-term defects, which has improved equipment maintenance schedules and personnel allocation towards fulfillment operations. Since its launch, this system has contributed to saving over $500 million in fulfillment expenses. It has also resulted in a better understanding of the flaws that cause long-term failures, which is now being used to build more sophisticated failure prediction and risk-mitigation systems for fulfillment equipment.

show abstract

Section: Diagnosticsmentioning

confidence: 99%

The Discriminants of Long and Short Duration Failures in Fulfillment Sortation Equipment: A Machine Learning Approach

Mutemi

Bação

2023

Journal of Engineering

View full text Add to dashboard Cite

show abstract

“…Therefore, to get better results, kernel methods should be implemented. Schurmann (1996) explained the concepts behind the designing of kernels and its properties. Ghosh (2000) elaborated that the Bayesian approach has been recognized as an promising technique for tackling clinical decision making problems and has the ability to represent uncertain knowledge.…”

Section: Literature Reviewmentioning

confidence: 99%

Bayesian Kernel Methods

Saxena

Kumar

2021

International Journal of Big Data and Analytics in Healthcare

View full text Add to dashboard Cite

In the healthcare industry, sources look after different customers with diverse diseases and complications. Thus, at the source, a great amount of data in all aspects like status of the patients, behaviour of the diseases, etc. are collected, and now it becomes the job of the practitioner at source to use the available data for diagnosing the diseases accurately and then prescribe the relevant treatment. Machine learning techniques are useful to deal with large datasets, with an aim to produce meaningful information from the raw information for the purpose of decision making. The inharmonious behavior of the data is the motivation behind the development of new tools and demonstrates the available information to some meaningful information for decision making. As per the literature, healthcare of patients can be analyzed through machine learning tools, and henceforth, in the article, a Bayesian kernel method for medical decision-making problems has been discussed, which suits the purpose of researchers in the enhancement of their research in the domain of medical decision making.

show abstract

“…This does not necessarily result in the highest achievable acc, but guarantees a balance between sens and spec. To quantify the added value of merging information from multiple sources, we trained second order polynomial classifiers (PCs) (Schürmann 1996) with combinations of two to six features and assessed reclassification sens, spec and acc. The optimization criterion for the PC was the classification acc.…”

Section: Classification and Performance Assessmentmentioning

confidence: 99%

Steps toward subject-specific classification in ECG-based detection of sleep apnea

2011

View full text Add to dashboard Cite

This study deals with ECG-based recognition of sleep apnea in epochs of 1 min duration using spectral- and correlation-based features extracted from the modulation of QRS amplitude, respiratory myogram interference and RR intervals. On a database comprising 140 simultaneous recordings of polysomnograms (PSGs) and 8-lead Holter-ECGs, it is shown that a single-parameter ROC threshold classification can achieve high detection rates up to 81.0% sensitivity and 85.6% specificity. Still, individual accuracy may be low, and the improvement employing feature combination by means of second order polynomial classifiers is only marginal. We speculate that individual differences, like co-morbidities, and even intra-individual confounding factors, like nocturnal changes in body position (BP), are major reasons for the difficulties to significantly raise the detection rate using multivariate techniques, which is evident in virtually all papers on that subject. Using the BP information in the PSG, we show a potential benefit for individualized single-feature classifiers by comparing the maximally achievable individual and global accuracy when either one optimal global threshold for the total dataset, individual threshold values for each subject or individual thresholds for each BP are applied. We developed an ECG-based BP segmentation algorithm and finally suggest a potential strategy to derive individually optimized subject-specific threshold values.

show abstract

Pattern Classification: A Unified View of Statistical and Neural Approaches.

Cited by 52 publications

References 0 publications

The Discriminants of Long and Short Duration Failures in Fulfillment Sortation Equipment: A Machine Learning Approach

The Discriminants of Long and Short Duration Failures in Fulfillment Sortation Equipment: A Machine Learning Approach

Bayesian Kernel Methods

Steps toward subject-specific classification in ECG-based detection of sleep apnea

Contact Info

Product

Resources

About