Machine failures decrease up-time and can lead to extra repair costs or even to human casualties and environmental pollution. Recent condition monitoring techniques use artificial intelligence in an effort to avoid time-consuming manual analysis and handcrafted feature extraction. Many of these only analyze a single machine and require a large historical data set. In practice, this can be difficult and expensive to collect. However, some industrial condition monitoring applications involve a fleet of similar operating machines. In most of these applications, it is safe to assume healthy conditions for the majority of machines. Deviating machine behavior is then an indicator for a machine fault. This work proposes an unsupervised, generic, anomaly detection framework for fleet-based condition monitoring. It uses generic building blocks and offers three key advantages. First, a historical data set is not required due to online fleet-based comparisons. Second, it allows incorporating domain expertise by user-defined comparison measures. Finally, contrary to most black-box artificial intelligence techniques, easy interpretability allows a domain expert to validate the predictions made by the framework. Two use-cases on an electrical machine fleet demonstrate the applicability of the framework to detect a voltage unbalance by means of electrical and vibration signatures.

In this paper we develop a data-driven smoothing technique for high-dimensional and non-linear panel data models. We allow for individual specific (non-linear) functions and estimation with econometric or machine learning methods by using weighted observations from other individuals. The weights are determined by a data-driven way and depend on the similarity between the corresponding functions and are measured based on initial estimates. The key feature of such a procedure is that it clusters individuals based on the distance / similarity between them, estimated in a first stage. Our estimation method can be combined with various statistical estimation procedures, in particular modern machine learning methods which are in particular fruitful in the high-dimensional case and with complex, heterogeneous data. The approach can be interpreted as a \textquotedblleft soft-clustering\textquotedblright\ in comparison to traditional\textquotedblleft\ hard clustering\textquotedblright that assigns each individual to exactly one group. We conduct a simulation study which shows that the prediction can be greatly improved by using our estimator. Finally, we analyze a big data set from didichuxing.com, a leading company in transportation industry, to analyze and predict the gap between supply and demand based on a large set of covariates. Our estimator clearly performs much better in out-of-sample prediction compared to existing linear panel data estimators.

Learning knowledge representation is an increasingly important technology applicable in many domain-specific machine learning problems. We discuss the effectiveness of traditional Link Prediction or Knowledge Graph Completion evaluation protocol when embedding knowledge representation for categorised multi-relational data in the clinical domain. Link prediction uses to split the data into training and evaluation subsets, leading to loss of information along training and harming the knowledge representation model accuracy. We propose a Clustering Evaluation Protocol as a replacement alternative to the traditionally used evaluation tasks. We used embedding models trained by a knowledge embedding approach which has been evaluated with clinical datasets. Experimental results with Pearson and Spearman correlations show strong evidence that the novel proposed evaluation protocol is pottentially able to replace link prediction.

The recent increase in the availability of system condition monitoring data has lead to increases in the use of data-driven approaches for fault diagnostics. The accuracy of the fault detection and classification using these approaches is generally good when abundant labelled data on healthy and faulty system conditions exists and the diagnosis problem is formulated as a supervised learning task, i.e. supervised fault diagnosis. It is, however, relatively common in real situations that only a small fraction of the system condition monitoring data are labeled as healthy and the rest is unlabeled due to the uncertainty of the number and type of faults that may occur. In this case, supervised fault diagnosis performs poorly. Fault diagnosis with an unknown number and nature of faults is an open set learning problem where the knowledge of the faulty system is incomplete during training and the number and extent of the faults, of different types, can evolve during testing. In this paper, we propose to formulate the open set diagnostics problem as a semi-supervised learning problem and we demonstrate how it can be solved using a knowledge-induced learning approach with adaptive sampling variational autoencoders (KIL-AdaVAE) in combination with a one-class classifier. The fault detection and segmentation capability of the proposed method is demonstrated on a simulated case study using the Advanced Geared Turbofan 30000 (AGTF30) dynamical model under real flight conditions and induced faults of 17 fault types. The performance of the method is compared to the different learning strategies (supervised learning, supervised learning with embedding and semi-supervised learning) and deep learning algorithms. The results demonstrate that the proposed method is able to significantly outperform all other tested methods in terms of fault detection and fault segmentation.

In machine learning, unsupervised learning is a machine learning model that infers the data pattern without any guidance or label. Many models are included in the unsupervised learning family, but one of my favorite models is Agglomerative Clustering. Agglomerative Clustering or bottom-up clustering essentially started from an individual cluster (each data point is considered as an individual cluster, also called leaf), then every cluster calculates their distance with each other. The two clusters with the shortest distance with each other would merge creating what we called node. Newly formed clusters once again calculating the member of their cluster distance with another cluster outside of their cluster.

The development of algorithms for unsupervised pattern recognition by nonlinear clustering is a notable problem in data science. Markov clustering (MCL) is a renowned algorithm that simulates stochastic flows on a network of sample similarities to detect the structural organization of clusters in the data, but it has never been generalized to deal with data nonlinearity. Minimum Curvilinearity (MC) is a principle that approximates nonlinear sample distances in the high-dimensional feature space by curvilinear distances, which are computed as transversal paths over their minimum spanning tree, and then stored in a kernel. Here we propose MC-MCL, which is the first nonlinear kernel extension of MCL and exploits Minimum Curvilinearity to enhance the performance of MCL in real and synthetic data with underlying nonlinear patterns. MC-MCL is compared with baseline clustering methods, including DBSCAN, K-means and affinity propagation. We find that Minimum Curvilinearity provides a valuable framework to estimate nonlinear distances also when its kernel is applied in combination with MCL. Indeed, MC-MCL overcomes classical MCL and even baseline clustering algorithms in different nonlinear datasets.

Many machine learning methods (classification, clustering, etc.) start with a known kernel that provides similarity or distance measure between two objects. Recent work has extended this to situations where the information about objects is limited to comparisons of distances between three objects (triplets). Humans find the comparison task much easier than the estimation of absolute similarities, so this kind of data can be easily obtained using crowd-sourcing. In this work, we give an efficient method of augmenting the triplets data, by utilizing additional implicit information inferred from the existing data. Triplets augmentation improves the quality of kernel-based and kernel-free data analytics tasks. Secondly, we also propose a novel set of algorithms for common supervised and unsupervised machine learning tasks based on triplets. These methods work directly with triplets, avoiding kernel evaluations. Experimental evaluation on real and synthetic datasets shows that our methods are more accurate than the current best-known techniques.

Clustering is one of the most challenging tasks in machine learning and pattern recognition, and perhaps, discovering the exact number of clusters of an unlabelled data set is the leading one. Many clustering methods find the clusters (or hidden classes) and the number of these clusters simultaneously (Frey and Dueck, 2007; Sajana et al., 2016). Although there exist methods for validating and comparing the quality of a partitioning of a data set, algorithms that provide the (estimated) number of clusters without any input parameter are still appealing. However, such methods or algorithms rely on other parameters viewed as the intensity, i.e. expected number of objects in a unit area. The value of the intensity parameter works as a threshold, and if the local intensity of the data set exceeds the threshold, it may indicate the existence of a possible cluster. However, the choice of such parameters is often a difficult task since different values of such parameters may drastically change the result of the algorithm. We use unsupervised adaptations of a family of vertex random digraphs, namely class cover catch digraphs (CCCDs), that showed relatively good performance in statistical pattern classification (Manukyan and Ceyhan, 2016; Priebe et al., 2003a). Unsupervised versions of CCCDs are called cluster catch digraphs (CCDs) (DeVinney, 2003; Marchette, 2004). Primarily, CCDs use statistics that require an intensity parameter to be specified or estimated.

In this work, learning schemes for measure-valued data are proposed, i.e. data that their structure can be more efficiently represented as probability measures instead of points on $\R^d$, employing the concept of probability barycenters as defined with respect to the Wasserstein metric. Such type of learning approaches are highly appreciated in many fields where the observational/experimental error is significant (e.g. astronomy, biology, remote sensing, etc.) or the data nature is more complex and the traditional learning algorithms are not applicable or effective to treat them (e.g. network data, interval data, high frequency records, matrix data, etc.). Under this perspective, each observation is identified by an appropriate probability measure and the proposed statistical learning schemes rely on discrimination criteria that utilize the geometric structure of the space of probability measures through core techniques from the optimal transport theory. The discussed approaches are implemented in two real world applications: (a) clustering eurozone countries according to their observed government bond yield curves and (b) classifying the areas of a satellite image to certain land uses categories which is a standard task in remote sensing. In both case studies the results are particularly interesting and meaningful while the accuracy obtained is high.

We are often interested in clustering objects that evolve over time and identifying solutions to the clustering problem for every time step. Evolutionary clustering provides insight into cluster evolution and temporal changes in cluster memberships while enabling performance superior to that achieved by independently clustering data collected at different time points. In this paper we introduce evolutionary affinity propagation (EAP), an evolutionary clustering algorithm that groups data points by exchanging messages on a factor graph. EAP promotes temporal smoothness of the solution to clustering time-evolving data by linking the nodes of the factor graph that are associated with adjacent data snapshots, and introduces consensus nodes to enable cluster tracking and identification of cluster births and deaths. Unlike existing evolutionary clustering methods that require additional processing to approximate the number of clusters or match them across time, EAP determines the number of clusters and tracks them automatically. A comparison with existing methods on simulated and experimental data demonstrates effectiveness of the proposed EAP algorithm.