Regression Function and Noise Variance Tracking Methods for Data Streams with Concept Drift

Most of the current data sources generate large amounts of data over time. Renewable energy generation is one example of such data sources. Machine learning is often applied to forecast time series. Since data flows are usually large, trends in data may change and learned patterns might not be optimal in the most recent data. In this paper, we analyse wind energy generation data extracted from the Sistema de Información del Operador del Sistema (ESIOS) of the Spanish power grid. We perform a study to evaluate detecting concept drifts to retrain models and thus improve the quality of forecasting. To this end, we compare the performance of a linear regression model when it is retrained randomly and when a concept drift is detected, respectively. Our experiments show that a concept drift approach improves forecasting between a 7.88% and a 33.97% depending on the concept drift technique applied.

show abstract

“…Especially in situations where data flows are continuous. In this context, it is necessary to retrain the models to improve their results [5].…”

Section: Introductionmentioning

confidence: 99%

Concept Drift Detection to Improve Time Series Forecasting of Wind Energy Generation

Cabello-López

Cañizares-Juan

Carranza-García

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…The integration of such areas is made in order to increase the levels of automation and efficiency in manufacturing and further requires the creation of a CPS, which is one of the main tasks of I4.0 (Roldán et al, 2019). Moreover, the significant increase * Corresponding author in data amounts needs to be processed in different activity fields of each manufacturing company (Jaworski, 2018). The analysis of the literature leads to the conclusion that, to be competitive, every modern and innovative manufacturing enterprise should implement the concept of Industry 4.0.…”

Section: Introductionmentioning

confidence: 99%

Modelling information for the burnishing process in a cyber-physical production system

Patalas-Maliszewska,

Posdzich,

Skrzypek

2022

International Journal of Applied Mathematics and Computer Scien

View full text Add to dashboard Cite

Currently, the manufacturing management board applies technologies in line with the concept of Industry 4.0. Cyberphysical production systems (CPSs) mean integrating computational processes with the corresponding physical ones, i.e., allowing work at the operational level and at the strategic level to run side by side. This paper proposes a framework to collect data and information from a production process, namely, the burnishing one, in order to monitor real-time deviations from the correct course of the process and thus reduce the number of defective products within the manufacturing process. The proposed new solutions consist of (i) the data and information of the production process, acquired from sensors, (ii) a predictive model, based on the Hellwig method for errors in the production process, relying on indications of a machine status, and (iii) an information layer system, integrating the process data acquired in real time with the model for predicting errors within the production process in an enterprise resource planning (ERP) system, that is, the business intelligence module. The possibilities of using the results of research in managerial practice are demonstrated through the application of an actual burnishing process. This new framework can be treated as a solution which will help managers to monitor the production flow and respond, in real time, to interruptions.

show abstract

“…In other words, the matrix K needs to be positive semidefinite. In many cases, like Laplacians of graphs (Wierzchoń and Kłopotek, 2018) or density-based regression (Jaworski, 2018), one knows in advance that they can be deemed as kernels embedded into an Euclidean space, so that there are no obstacles to apply kernel-k-means clustering. However, the kernel matrix may not be a Mercer kernel matrix.…”

Section: Introductionmentioning

confidence: 99%

A feasible k-means kernel trick under non-Euclidean feature space

Kłopotek¹,

Kłopotek²,

Wierzchoń³

2020

View full text Add to dashboard Cite

This paper poses the question of whether or not the usage of the kernel trick is justified. We investigate it for the special case of its usage in the kernel k-means algorithm. Kernel-k-means is a clustering algorithm, allowing clustering data in a similar way to k-means when an embedding of data points into Euclidean space is not provided and instead a matrix of "distances" (dissimilarities) or similarities is available. The kernel trick allows us to by-pass the need of finding an embedding into Euclidean space. We show that the algorithm returns wrong results if the embedding actually does not exist. This means that the embedding must be found prior to the usage of the algorithm. If it is found, then the kernel trick is pointless. If it is not found, the distance matrix needs to be repaired. But the reparation methods require the construction of an embedding, which first makes the kernel trick pointless, because it is not needed, and second, the kernel-k-means may return different clusterings prior to repairing and after repairing so that the value of the clustering is questioned. In the paper, we identify a distance repairing method that produces the same clustering prior to its application and afterwards and does not need to be performed explicitly, so that the embedding does not need to be constructed explicitly. This renders the kernel trick applicable for kernel-k-means.

show abstract

Regression Function and Noise Variance Tracking Methods for Data Streams with Concept Drift

Cited by 11 publications

References 44 publications

Concept Drift Detection to Improve Time Series Forecasting of Wind Energy Generation

Concept Drift Detection to Improve Time Series Forecasting of Wind Energy Generation

Modelling information for the burnishing process in a cyber-physical production system

A feasible k-means kernel trick under non-Euclidean feature space

Contact Info

Product

Resources

About