L1 norm based KPCA for novelty detection

In this work, a machine learning approach is proposed for the problem of leak detection in noisy industrial pipelines. The traditional multi-class or binary classification approaches depend on the fact that real training data are required. However, in real pipeline scenarios the data generation for the leak class relies on measurements of artificially generated leaks which are different in nature than actual ones. Also, some pipelines are not equipped with the components to generate these leaks and in some cases, it is difficult to acquire a large and balanced leak dataset. Thus, in this paper, a set of one class classification models are applied that do not require training with real leak data. In this study's case, four one class classification models are trained on a single class representing the pipeline's normal operating noise. Seven time and frequency domain features are extracted from the raw acoustic data acquired by a set of accelerometers. The trained models are then tested on a new dataset containing leak and noise measurements. This dataset is used to evaluate each model's ability to detect leaks as well as the effect that some introduced parameters have on their performance. Overall, high levels of accuracy are exhibited, and all models are able to distinguish between noise and leak data.

A novelty detection machine and its application to bank failure prediction

Biomedical relation extraction from scientific publications is a key task in biomedical natural language processing (NLP) and can facilitate the creation of large knowledge bases, enable more efficient knowledge discovery, and accelerate evidence synthesis. In this paper, building upon our previous effort in the BioCreative VIII BioRED Track, we propose an enhanced end-to-end pipeline approach for biomedical relation extraction (RE) and novelty detection (ND) that effectively leverages existing datasets and integrates state-of-the-art deep learning methods. Our pipeline consists of four tasks performed sequentially: named entity recognition (NER), entity linking (EL), RE, and ND. We trained models using the BioRED benchmark corpus that was the basis of the shared task. We explored several methods for each task and combinations thereof: for NER, we compared a BERT-based sequence labeling model that uses the BIO scheme with a span classification model. For EL, we trained a convolutional neural network model for diseases and chemicals and used an existing tool, PubTator 3.0, for mapping other entity types. For RE and ND, we adapted the BERT-based, sentence-bound PURE model to bidirectional and document-level extraction. We also performed extensive hyperparameter tuning to improve model performance. We obtained our best performance using BERT-based models for NER, RE, and ND, and the hybrid approach for EL. Our enhanced and optimized pipeline showed substantial improvement compared to our shared task submission, NER: 93.53 (+3.09), EL: 83.87 (+9.73), RE: 46.18 (+15.67), and ND: 38.86 (+14.9). While the performances of the NER and EL models are reasonably high, RE and ND tasks remain challenging at the document level. Further enhancements to the dataset could enable more accurate and useful models for practical use. We provide our models and code at https://github.com/janinaj/e2eBioMedRE/. Database URL: https://github.com/janinaj/e2eBioMedRE/.

This paper is concerned with the one class classification(OCC) problem. By introducing the vector-valued function with regularizations in Y-valued Reproducing Hilbert Kernel Space(RHKS), we build an unsupervised classifier and discover the outliers and inliers simultaneously. Manifold regularization is employed to preserve the local similarity of data in input space. Experimental results of the proposed and comparing methods on OCC data sets demonstrate the performance of the proposed algorithm.

Kernel dependence analysis and graph structure morphing for novelty detection with high-dimensional small size data set

Neuroimaging techniques combined with computational neuroanatomy have been playing a role in the investigation of healthy aging and Alzheimer's disease (AD). The definition of normative rules for brain features is a crucial step to establish typical and atypical aging trajectories. To introduce an unsupervised pattern recognition method; to define multivariate normative rules of neuroanatomical measures; and to propose a brain abnormality index. This study was based on a machine learning approach (one class classification or novelty detection) to neuroanatomical measures (brain regions, volume, and cortical thickness) extracted from the Alzheimer's Disease Neuroimaging Initiative (ADNI)'s database. We applied a ν-One-Class Support Vector Machine (ν-OC-SVM) trained with data from healthy subjects to build an abnormality index, which was compared with subjects diagnosed with mild cognitive impairment and AD. The method was able to classify AD subjects as outliers with an accuracy of 84.3% at a false alarm rate of 32.5%. The proposed brain abnormality index was found to be significantly associated with group diagnosis, clinical data, biomarkers, and future conversion to AD. These results suggest that one-class classification may be a promising approach to help in the detection of disease conditions. Our findings support a framework considering the continuum of brain abnormalities from healthy aging to AD, which is correlated with cognitive impairment and biomarkers measurements.

In condition monitoring lack of properly balanced data sets with faulty and healthy cases makes proper condition recognition very challenging. In many cases, one may have good condition data only as the machine is unique and there is no other example. This issue is addressed by proposing a support vector machine for novelty detection applied to health index data. In this scheme, the moving window approach has been utilized in which the simple statistical parameterization of the data is carried out. Then the model in the multidimensional (mD) space is constructed whose shape is defined by an estimated hypersphere border. If the data lies inside the border then it can be used to re-train the model. Whereas if it is outside the border then it cannot be recognized as a healthy case. The size of the mD hypersphere (for m = 2) describes the location of the good-condition data cloud as a potential feature. If the size of the data cloud is growing, it means more dispersion of the data. The efficiency of the method is tested on simulated and well-known real data sets having Gaussian and non-Gaussian disturbances.

Condition monitoring is an indispensable element related to the operation of rotating machinery. In this article, the monitoring system for the parallel gearbox was proposed. The novelty detection approach is used to develop the condition assessment support system, which requires data collection for a healthy structure. The measured signals were processed to extract quantitative indicators sensitive to the type of damage occurring in this type of structure. The indicator’s values were used for the development of four different novelty detection algorithms. Presented novelty detection models operate on three principles: feature space distance, probability distribution, and input reconstruction. One of the distance-based models is adaptive, adjusting to new data flowing in the form of a stream. The authors test the developed algorithms on experimental and simulation data with a similar distribution, using the training set consisting mainly of samples generated by the simulator. Presented in the article results demonstrate the effectiveness of the trained models on both data sets.

En este artículo se presenta un nuevo método para generar sistemas difusos interpretables, a partir de datos experimentalesde entrada y salida, para resolver problemas de clasificación. En la partición antecedente se emplean conjuntos triangulares con interpolación de 0.5 lo cual evita la presencia de solapamientos complejos que suceden en otros métodos. Los consecuentes, tipo Singleton, son generados por la proyección de los valores modales de cada función de membresía triangular en el espacio de salida y se emplea el método de mínimos cuadrados para el ajuste de los consecuentes. El método propuesto consigue una mayor precisión que la alcanzada con los métodos actuales existentes, empleando un número reducido de reglas y parámetros y sin sacrificar la interpretabilidad del modelo difuso. El enfoque propuesto es aplicado a dos problemas clásicos de clasificación: el Wisconsin Breast Cancer (WBC) y el Iris Data Classification Problem, para mostrar las ventajas del método y comparar los resultados con los alcanzados por otros investigadores.

Novelty detection is concerned with identifying abnormal system behaviours and abrupt changes from one regime to another. This paper proposes an on-line (causal) novelty detection method capable of detecting both outliers and regime change points in sequential time-series data. Our approach is based on a Kalman filter in order to model time-series data and extreme value theory is used to compute a novelty measure in a principled manner. The proposed approach is shown to be effective via experiments on several real-world data sets.

Previous research in novelty detection has focused on the task of finding novel material, given a set or stream of documents on a certain topic. This study investigates the more difficult two-part task defined by the TREC 2002 novelty track: given a topic and a group of documents relevant to that topic, 1) find the relevant sentences from the documents, and 2) find the novel sentences from the collection of relevant sentences. Our research shows that the former step appears to be the more difficult part of this task, and that the performance of novelty measures is very sensitive to the presence of non-relevant sentences.

In recent years, non-iterative learning approaches for kernel have received quite an attention by researchers and kernel ridge regression (KRR) approach is one of them. Recently, KRR-based Auto-Encoder is developed for the one-class classification (OCC) task and named as AEKOC. OCC is generally used for outlier or novelty detection. The brain can detect outlier just by learning from only normal samples. Similarly, OCC also uses only normal samples to train the model, and trained model can be used for outlier detection. In this paper, AEKOC is enabled to utilize privileged information, which is generally ignored by AEKOC or any traditional machine learning technique but usually present in human learning. For this purpose, we have combined learning using privileged information (LUPI) framework with AEKOC, and proposed a classifier, which is referred to as AEKOC+. Privileged information is only available during training but not during testing. Therefore, AEKOC is unable to utilize this information for building the model. However, AEKOC+ can efficiently handle the privileged information due to the inclusion of the LUPI framework with AEKOC. Experiments have been conducted on MNIST dataset and on various other datasets from UCI machine learning repository, which demonstrates the superiority of AEKOC+ over AEKOC. Our formulation shows that AEKOC does not utilize the privileged features in learning; however, formulation of AEKOC+ helps it in learning from the privileged features differently from other available features and improved generalization performance of AEKOC. Moreover, AEKOC+ also outperformed two LUPI framework–based one-class classifiers (i.e., OCSVM+ and SSVDD+).

Anomalies are samples that significantly deviate from the rest of the data and their detection plays a major role in building machine learning models that can be reliably used in applications such as data-driven design and novelty detection. The majority of existing anomaly detection methods either are exclusively developed for (semi) supervised settings, or provide poor performance in unsupervised applications where there are no training data with labeled anomalous samples. To bridge this research gap, we introduce a robust, efficient, and interpretable methodology based on nonlinear manifold learning to detect anomalies in unsupervised settings. The essence of our approach is to learn a low-dimensional and interpretable latent representation (aka manifold) for all the data points such that normal samples are automatically clustered together and hence can be easily and robustly identified. We learn this low-dimensional manifold by designing a learning algorithm that leverages either a latent map Gaussian process (LMGP) or a deep autoencoder (AE). Our LMGP-based approach, in particular, provides a probabilistic perspective on the learning task and is ideal for high-dimensional applications with scarce data. We demonstrate the superior performance of our approach over existing technologies via multiple analytic examples and real-world datasets.

Sparse unmixing is an ill-posed inverse problem to find the abundances of each mineral in a scene using a large spectral library. Since the homogeneous regions tend to have similar fractional abundances, including spatial–contextual information increases the performance of an unmixing algorithm. In this letter, we propose a double spatial graph Laplacian regularization for sparse unmixing (DSGLSU). The proposed method consists of two steps. In the first step, hyperspectral data are approximated using a suitable transformation to generate scaled (coarse) data to exploit the interpixel information. Then, a sparsity constrained optimization problem is solved in the approximation domain to produce a coarse abundance matrix. A low-resolution abundance map is obtained by applying an inverse transformation to a coarse abundance map. In the second step, a graph Laplacian and weighted sparsity regularized optimization problem is solved in the original domain to obtain an abundance map where the low-resolution abundance map obtained in the first step is used as a weight term in sparsity regularization. We perform experiments on two simulated data sets and a real data set. It has been shown that the proposed method outperforms the state-of-the-art spatially regularized sparse unmixing methods proposed in the literature.

Metabolite set enrichment analysis (MSEA) has gained increasing research interest for identification of perturbed metabolic pathways in metabolomics. The method incorporates predefined metabolic pathways information in the analysis where metabolite sets are typically assumed to be mutually exclusive to each other. However, metabolic pathways are known to contain common metabolites and intermediates. This situation, along with limitations in metabolite detection or coverage leads to overlapping, incomplete metabolite sets in pathway analysis. For overlapping metabolite sets, MSEA tends to result in high false positives due to improper weights allocated to the overlapping metabolites. Here, we proposed an extended partial least squares (PLS) model with a new sparse scheme for overlapping metabolite set enrichment analysis, named overlapping group PLS (ogPLS) analysis. The weight vector of the ogPLS model was decomposed into pathway-specific subvectors, and then a group lasso penalty was imposed on these subvectors to achieve a proper weight allocation for the overlapping metabolites. Two strategies were adopted in the proposed ogPLS model to identify the perturbed metabolic pathways. The first strategy involves debiasing regularization, which was used to reduce inequalities amongst the predefined metabolic pathways. The second strategy is stable selection, which was used to rank pathways while avoiding the nuisance problems of model parameter optimization. Both simulated and real-world metabolomic datasets were used to evaluate the proposed method and compare with two other MSEA methods including Global-test and the multiblock PLS (MB-PLS)-based pathway importance in projection (PIP) methods. Using a simulated dataset with known perturbed pathways, the average true discovery rate for the ogPLS method was found to be higher than the Global-test and the MB-PLS-based PIP methods. Analysis with a real-world metabolomics dataset also indicated that the developed method was less prone to select pathways with highly overlapped detected metabolite sets. Compared with the two other methods, the proposed method features higher accuracy, lower false-positive rate, and is more robust when applied to overlapping metabolite set analysis. The developed ogPLS method may serve as an alternative MSEA method to facilitate biological interpretation of metabolomics data for overlapping metabolite sets.