eXplainable AI (XAI)-Based Input Variable Selection Methodology for Forecasting Energy Consumption

Sim, Taeyong; Choi, Seon-Bin; Kim, Yun-Jae; Youn, Su Hyun; Jang, Dong‐Jin; Chun, Chang-Jae

doi:10.3390/electronics11182947

Cited by 12 publications

(5 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Shapley value, a tool originating from game theory that assesses the average contribution of players to a game, finds application in various fields, from economics to signal processing [56][57][58]. In this study, we employed Shapley values to gauge the contribution of features to classifier performance.…”

Section: Feature Sets and Classification Methodsmentioning

confidence: 99%

A New Shapley-Based Feature Selection Method in a Clinical Decision Support System for the Identification of Lung Diseases

Kababulut,

Gürkan Kuntalp,

Düzyel

et al. 2023

Diagnostics

View full text Add to dashboard Cite

The aim of this study is to propose a new feature selection method based on the class-based contribution of Shapley values. For this purpose, a clinical decision support system was developed to assist doctors in their diagnosis of lung diseases from lung sounds. The developed systems, which are based on the Decision Tree Algorithm (DTA), create a classification for five different cases: healthy and disease (URTI, COPD, Pneumonia, and Bronchiolitis) states. The most important reason for using a Decision Tree Classifier instead of other high-performance classifiers such as CNN and RNN is that the class contributions of Shapley values can be seen with this classifier. The systems developed consist of either a single DTA classifier or five parallel DTA classifiers each of which is optimized to make a binary classification such as healthy vs. others, COPD vs. Others, etc. Feature sets based on Power Spectral Density (PSD), Mel Frequency Cepstral Coefficients (MFCC), and statistical characteristics extracted from lung sound recordings were used in these classifications. The results indicate that employing features selected based on the class-based contribution of Shapley values, along with utilizing an ensemble (parallel) system, leads to improved classification performance compared to performances using either raw features alone or traditional use of Shapley values.

show abstract

Section: Feature Sets and Classification Methodsmentioning

confidence: 99%

A New Shapley-Based Feature Selection Method in a Clinical Decision Support System for the Identification of Lung Diseases

Kababulut,

Gürkan Kuntalp,

Düzyel

et al. 2023

Diagnostics

View full text Add to dashboard Cite

show abstract

“…Sim et al [11] presented a unique approach for variable selection in energy forecasting models. The authors utilized Support Vector Regressor (SVR), Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), long-LSTM, and the SHAP framework to identify relevant input factors using energy data from a university building in Seoul.…”

Section: B Studies On Xai Techniques For Forecasting Modelsmentioning

confidence: 99%

“…We chose residential energy usage since most of the previously conducted studies [11], [12], [13], [14] in this area focus primarily on energy usage in the industrial and commercial sectors. While these sectors contribute significantly to overall energy consumption, there is growing awareness that homes play a critical role in overall energy usage patterns.…”

Section: Introductionmentioning

confidence: 99%

Enhancing Household Energy Consumption Predictions Through Explainable AI Frameworks

Bhandary,

Dobariya,

Yenduri

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Effective energy management is crucial for sustainability, carbon reduction, resource conservation, and cost savings. However, conventional energy forecasting methods often lack accuracy, suggesting the need for advanced approaches. Artificial intelligence (AI) has emerged as a powerful tool for energy forecasting, but its lack of transparency and interpretability poses challenges for understanding its predictions. In response, Explainable AI (XAI) frameworks have been developed to enhance the transparency and interpretability of black-box AI models. Accordingly, this paper focuses on achieving accurate household energy consumption predictions by comparing prediction models based on several evaluation metrics, namely the Coefficient of Determination (R 2 ), Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and Mean Absolute Error (MAE). The best model is identified by comparison after making predictions on unseen data, after which the predictions are explained by leveraging two XAI frameworks: Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP). These explanations help identify crucial characteristics contributing to energy consumption predictions, including insights into feature importance. Our findings underscore the significance of current consumption patterns and lagged energy consumption values in estimating energy usage. This paper further demonstrates the role of XAI in developing consistent and reliable predictive models.

show abstract

“…Specifically, LIME and SHAP have been investigated for generating local explanations. Examples include Srinivasan's SHAP-based XAI-FDD [61], Sim's SHAP-based analysis of input variables for energy consumption forecasting [63], and Wastensteiner's LIME and SHAP-based visualizations for personalized feedback on electricity consumption time-series data [62]. Srinivasan et al [61] proposed XAI-FDD (explainable artificial intelligence-fault detection and diagnosis), which uses explanations generated for each data instance to detect incipient faults.…”

Section: Energy and Building Managementmentioning

confidence: 99%

“…The XAI-FDD may examine air-handling systems, renewable energy sources, and other building energy components. Sim et al [63] used SHAP-based XAI to examine how input variables affect energy use forecasting. Their research divided the input variables into three categories-strong, ambiguous, and weak-providing insights into which variables had the most significant impact.…”

Section: Energy and Building Managementmentioning

confidence: 99%

Exploring Local Explanation of Practical Industrial AI Applications: A Systematic Literature Review

Prihatno

Oktian

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

In recent years, numerous explainable artificial intelligence (XAI) use cases have been developed, to solve numerous real problems in industrial applications while maintaining the explainability level of the used artificial intelligence (AI) models to judge their quality and potentially hold the models accountable if they become corrupted. Therefore, understanding the state-of-the-art methods, pointing out recent issues, and deriving future directions are important to drive XAI research efficiently. This paper presents a systematic literature review of local explanation techniques and their practical applications in various industrial sectors. We first establish the need for XAI in response to opaque AI models and survey different local explanation methods for industrial AI applications. The number of studies is then examined with several factors, including industry sectors, AI models, data types, and XAI-based usage and purpose. We also look at the advantages and disadvantages of local explanation methods and how well they work in practical settings. The difficulties of using local explanation techniques are also covered, including computing complexity and the trade-off between precision and interpretability. Our findings demonstrate that local explanation techniques can boost industrial AI models’ transparency and interpretability and give insightful information about them. The efficiency of these procedures must be improved, and ethical concerns about their application must be resolved. This paper contributes to the increasing knowledge of local explanation strategies and offers guidance to academics and industry professionals who want to use these methods in practical settings.

show abstract

eXplainable AI (XAI)-Based Input Variable Selection Methodology for Forecasting Energy Consumption

Cited by 12 publications

References 45 publications

A New Shapley-Based Feature Selection Method in a Clinical Decision Support System for the Identification of Lung Diseases

A New Shapley-Based Feature Selection Method in a Clinical Decision Support System for the Identification of Lung Diseases

Enhancing Household Energy Consumption Predictions Through Explainable AI Frameworks

Exploring Local Explanation of Practical Industrial AI Applications: A Systematic Literature Review

Contact Info

Product

Resources

About