Explainable K-Means Clustering for Occupancy Estimation

Prabhakaran, Kumar; Dridi, Jawher; Amayri, Manar; Bouguila, Nizar

doi:10.1016/j.procs.2022.07.041

Cited by 18 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To increase the interpretability of the approach, the authors also implemented a CART algorithm, showing how it is possible to perform an early FDD analysis based on anomalous loads profiles. Prabhakaran et al [146] proposed a small binary decision tree to increase the interpretability of a Kmeans model for indoor occupancy estimation. Galli et al [147] proposed a multi-step methodology based on a clustering algorithm and a LIME to investigate the energy per-formance classes of buildings using a large set of energy performance certificates (EPCs) as input.…”

Section: Clustering and Feature Extractionmentioning

confidence: 99%

A Future Direction of Machine Learning for Building Energy Management: Interpretable Models

Gugliermetti,

Cumo,

Agostinelli

2024

Energies

View full text Add to dashboard Cite

Machine learning (ML) algorithms are now part of everyday life, as many technological devices use these algorithms. The spectrum of uses is wide, but it is evident that ML represents a revolution that may change almost every human activity. However, as for all innovations, it comes with challenges. One of the most critical of these challenges is providing users with an understanding of how models’ output is related to input data. This is called “interpretability”, and it is focused on explaining what feature influences a model’s output. Some algorithms have a simple and easy-to-understand relationship between input and output, while other models are “black boxes” that return an output without giving the user information as to what influenced it. The lack of this knowledge creates a truthfulness issue when the output is inspected by a human, especially when the operator is not a data scientist. The Building and Construction sector is starting to face this innovation, and its scientific community is working to define best practices and models. This work is intended for developing a deep analysis to determine how interpretable ML models could be among the most promising future technologies for the energy management in built environments.

show abstract

Section: Clustering and Feature Extractionmentioning

confidence: 99%

A Future Direction of Machine Learning for Building Energy Management: Interpretable Models

Gugliermetti,

Cumo,

Agostinelli

2024

Energies

View full text Add to dashboard Cite

show abstract

“…Moreover, users do not need to provide additional information, making the algorithm highly applicable. Rule-based clustering explanation [16][17][18] sometimes generates unnecessary rules, which diminishes the explanation of the algorithm.…”

Section: Rule-based Interpretation Of Clusteringmentioning

confidence: 99%

Clustering Explanation based on Multi-Hyperrectangle

Zeng,

Zhong,

Pan

2024

Preprint

View full text Add to dashboard Cite

Clustering plays a crucial role in data mining and pattern recognition, but the interpretation of clustering results is often challenging. Existing interpretation methods usually lack an intuitive and accurate description of irregular shapes and high-dimensional data. This paper proposes a clustering explanation method based on a Multi-HyperRectangle (MHR), for extracting post hoc explanations of clustering results. MHR first generates initial hyperrectangles to cover a cluster, and then gradually merges these hyperrectangles to obtain the optimal shape that fits the cluster. Furthermore, we propose a refinement method to improve the tightness of the hyperrectangles to make the interpretation results more precise and easier to understand. Experimental results show that MHR has significant advantages in terms of tightness and accuracy of cluster interpretation.

show abstract