Time series grouping algorithm for load pattern recognition

Manojlović, Igor; Svenda, Goran; Erdeljan, Aleksandar; Gavric, Milan

doi:10.1016/j.compind.2019.07.009

Cited by 10 publications

(2 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other side, this approach requires defining some time series grouping criteria, which can lead to suboptimal groupings [15]. Among pooling strategies, we can find model-based clustering [14,16,17], random clustering [8], grouping based on similarity measures [18][19][20], or expert judgement [21]. Recent research has explored creating global models considering all available time series, regardless of their heterogeneity, obtaining promising results [3,8,22].…”

Section: Forecasting Time Series: Local Vs Global Approachmentioning

confidence: 99%

Explaining Bad Forecasts in Global Time Series Models

et al. 2021

View full text Add to dashboard Cite

While increasing empirical evidence suggests that global time series forecasting models can achieve better forecasting performance than local ones, there is a research void regarding when and why the global models fail to provide a good forecast. This paper uses anomaly detection algorithms and explainable artificial intelligence (XAI) to answer when and why a forecast should not be trusted. To address this issue, a dashboard was built to inform the user regarding (i) the relevance of the features for that particular forecast, (ii) which training samples most likely influenced the forecast outcome, (iii) why the forecast is considered an outlier, and (iv) provide a range of counterfactual examples to understand value changes, in the feature vector or the predicted value, can lead to a different outcome. Moreover, a modular architecture and a methodology were developed to iteratively remove noisy data instances from the train set, to enhance the overall global time series forecasting model performance. Finally, to test the effectiveness of the proposed approach, it was validated on two publicly available real-world datasets.

show abstract

Section: Forecasting Time Series: Local Vs Global Approachmentioning

confidence: 99%

Explaining Bad Forecasts in Global Time Series Models

et al. 2021

View full text Add to dashboard Cite

show abstract

“…On the other side, this approach requires defining some time series grouping criteria, which can lead to suboptimal groupings [15]. Among pooling strategies we can find model-based clustering [14,16,17], random clustering [8], grouping based on similarity measures [18][19][20], or expert judgement [21]. Recent research has explored creating global models considering all available time series, regardless of their heterogeneity, obtaining promising results [3,8,22].…”

Section: Review Of Related Scientific Work 21 Forecasting Time Series: Local Vs Global Approachmentioning

confidence: 99%

Explaining Bad Forecasts in Global Time Series Models

Rožanec¹,

Trajkova²,

Kenda³

et al. 2021

Preprint

View full text Add to dashboard Cite

While increasing empirical evidence suggests that global time series forecasting models can achieve better forecasting performance than local ones, there is a research void regarding when and why the global models fail to provide a good forecast. This paper uses anomaly detection algorithms and Explainable Artificial Intelligence (XAI) to answer when and why a forecast should not be trusted. To address this issue, a dashboard was built to inform the user regarding (i) the relevance of the features for that particular forecast, (ii) which training samples most likely influenced the forecast outcome, (iii) why the forecast is considered an outlier, and (iv) provide a range of counterfactual examples to understand value changes, in the feature vector or the predicted value, can lead to a different outcome. Moreover, a modular architecture and a methodology were developed to iteratively remove noisy data instances from the train set, to enhance the overall global time series forecasting model performance. Finally, to test the effectiveness of the proposed approach, it was validated on two publicly available real-world datasets.

show abstract

Theoretical Fundamentals of the Three-Phase Distribution State Estimator

Švenda

2024

Power Electronics and Power Systems

View full text Add to dashboard Cite

Time series grouping algorithm for load pattern recognition

Cited by 10 publications

References 34 publications

Explaining Bad Forecasts in Global Time Series Models

Explaining Bad Forecasts in Global Time Series Models

Explaining Bad Forecasts in Global Time Series Models

Theoretical Fundamentals of the Three-Phase Distribution State Estimator

Contact Info

Product

Resources

About