This paper attempts to identify $$k_1$$ -most demanding products using K-Means clustering algorithm. A comparison of proposed algorithm with existing algorithms has been made. The experiments performed on synthetic and real datasets showed the effectiveness of our proposed algorithm.

Predictive maintenance is regarded by many as a key factor in Industrial Internet of Things (IIoT) and the development of “smart” factories. With the growing use of sensors and embedded computing systems, the term predictive maintenance is most often understood as a strategy that relies on collecting streaming sensor data and performing condition monitoring. Thus, the majority of academic papers base their research work solely on sensorial sources coming from the shop floor machinery, neglecting the knowledge already existing in legacy systems and maintenance historical logs. The UPTIME project aims to develop a unified predictive maintenance framework that incorporates information from heterogeneous data sources, both from sensor devices and legacy/operational systems. In this contribution, we share our first insights on legacy data analytics in the predictive maintenance context, and outline the tools and approaches we developed in the course of the project. Experimental work has been conducted using real world datasets deriving from an actual manufacturing facility in the White Goods/Home Appliances sector. The results provide significant knowledge about the manufacturing processes and show the potential of the proposed methodology.

Fault detection in wind turbine which is identified with complete system monitoring under multi-fault scenario is proposed. When a fault is detected, its types and location are recognized for easy maintenance. Fault in wind turbines is caused due to the high speed of gearbox, generator bearing and the failures occurred in various parts. In wind farm, wind turbine condition monitoring is used to reduce the maintenance cost and also improves the accuracy. Generally, in wind turbine gearbox condition monitoring using sensor is a gainful method to monitor wind turbine performance and fault. This paper nominates a method to decide the parameters for support vector machine (SVM) in wind turbine called Cuckoo search optimization (CSO). The combination of optimization technique with classification technique is evaluated. MATLAB platform was used to evaluate the various faults under fixed value and gain factor conditions. Comparing the accuracy with SVM, particle swarm optimized SVM and k-nearest neighbor, the proposed fault detection and fault isolation technique (CSO-SVM) is improved by 2.5%, 3.5% and 6.5%, respectively. The result shows the CSO model based on SVM algorithm accomplishes the most accurate fault detection than the past models.

Existing data mining techniques, more particularly iterative learning algorithms, become overwhelmed with big data. While parallelism is an obvious and, usually, necessary strategy, we observe that both (1) continually revisiting data and (2) visiting all data are two of the most prominent problems especially for iterative, unsupervised algorithms like expectation maximization algorithm for clustering (EM-T). Our strategy is to embed EM-T into a nonlinear hierarchical data structure (heap) that allows us to (1) separate data that needs to be revisited from data that does not and (2) narrow the iteration toward the data that is more difficult to cluster. We call this extended EM-T, EM*. We show our EM* algorithm outperform EM-T algorithm over large real-world and synthetic data sets. We lastly conclude with some theoretical underpinnings that explain why EM* is successful.

Efficient management and analysis of large volumes of data is a demanding task of increasing scientific and industrial importance, as the ubiquitous generation of information governs more and more aspects of human life. In this article, we introduce FML-kNN, a novel distributed processing framework for Big Data that performs probabilistic classification and regression, implemented in Apache Flink. The framework’s core is consisted of a k-nearest neighbor joins algorithm which, contrary to similar approaches, is executed in a single distributed session and is able to operate on very large volumes of data of variable granularity and dimensionality. We assess FML-kNN’s performance and scalability in a detailed experimental evaluation, in which it is compared to similar methods implemented in Apache Hadoop, Spark, and Flink distributed processing engines. The results indicate an overall superiority of our framework in all the performed comparisons. Further, we apply FML-kNN in two motivating uses cases for water demand management, against real-world domestic water consumption data. In particular, we focus on forecasting water consumption using 1-h smart meter data, and extracting consumer characteristics from water use data in the shower. We further discuss on the obtained results, demonstrating the framework’s potential in useful knowledge extraction.

Scientific retrieval systems need to be given domain search terms for searching publications, however, as natural language, search terms provided by users are often fuzzy and limited and some relevant terms are always overlooked in searching. Meanwhile, users always desire to be given domain related keywords to enlighten themselves what other terms can be used for their searching. This paper presents a concepts recommendation model in scientific paper retrieval, in which concepts are extracted from keyword in scientific papers, and some data mining algorithms are used to calculate the similarity between search terms and concepts and do recommendation for users. This model is simple and can be used with small dataset, in which all training data used is from meta data of papers that is easy to acquired. Experimental result hold good precision, which shows that this research not only simplifies searching step and improves the searching quality for users, but also lays the foundation for semantic search.

The article describes extension of λ-calculation for creation of parallel data mining algorithms. The proposed approach uses presentation of the algorithm as a consequence of pure functions with unified interfaces. For parallel execution we use special function that allows to change a structure of the algorithm and to implement various strategies for processing of data set and model.

In this study we analyze the dynamics of the contact list evolution of millions of users of the Skype communication network. We find that egocentric networks evolve heterogeneously in time as events of edge additions and deletions of individuals are grouped in long bursty clusters, which are separated by long inactive periods. We classify users by their link creation dynamics and show that bursty peaks of contact additions are likely to appear shortly after user account creation. We also study possible relations between bursty contact addition activity and other user-initiated actions like free and paid service adoption events. We show that bursts of contact additions are associated with increases in activity and adoption—an observation that can inform the design of targeted marketing tactics.

Random Forests (RF) models have been proven to perform well in both classification and regression. However, with the randomizing mechanism in both bagging samples and feature selection, the performance of RF can deteriorate when applied to high-dimensional data. In this paper, we propose a new approach for feature sampling for RF to deal with high-dimensional data. We first apply $$p$$ -value to assess the feature importance on finding a cut-off between informative and less informative features. The set of informative features is then further partitioned into two groups, highly informative and informative features, using some statistical measures. When sampling the feature subspace for learning RFs, features from the three groups are taken into account. The new subspace sampling method maintains the diversity and the randomness of the forest and enables one to generate trees with a lower prediction error. In addition, quantile regression is employed to obtain predictions in the regression problem for a robustness towards outliers. The experimental results demonstrated that the proposed approach for learning random forests significantly reduced prediction errors and outperformed most existing random forests when dealing with high-dimensional data.