Machine learning techniques for diagrid building design: Architectural–Structural correlations with feature selection and data augmentation

Kazemi, Pooyan; Entezami, Alireza; Ghisi, Aldo

doi:10.1016/j.jobe.2024.108766

Cited by 3 publications

(9 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Data Synthesis and Augmentation Algorithmsmentioning

confidence: 64%

“…From this rationale, five responses, such as "displacement of the top story", "maximum utilization in compression", "EDW", "displacement/EDW", and "EDW/AR", are pinpointed. Compared to the results shown in our previous work [43], the same five responses are obtained from the procedure using Spearman's correlation (current work) instead of Pearson's correlation coefficient (previous work). This result is relatively unsurprising due to the similarity between the two indices (Spearman is a kind of Pearson on ranked values in a column of data).…”

Section: Response Selectionmentioning

confidence: 68%

“…The combination of innovative architectural design concepts with limited data poses significant challenges. When ML tools are employed with small datasets, in fact, inconsistencies in results have been observed, as in our previous work [43], often attributed to To sum up, this method can pinpoint common features across different responses. When examining the designated features for each distinct response, it is clear that the "bottom plan side count" is consistently selected across most responses.…”

Section: Data Synthesis and Augmentation Algorithmsmentioning

confidence: 80%

“…By generating novel synthetic data points, DA is established as a potent tool in tackling challenges arising from limited training data, bolstering the resilience and efficacy of machine learning models [44,45]. While in our previous work [43], we only showed the results for the Gaussian copula, in the present paper several alternative DA algorithms are instead comparatively evaluated.…”

Section: Introductionmentioning

confidence: 93%

“…Both methods aim to boost model precision and clarity while efficiently using computational resources, and they are discussed in this study in the specific context of the early design phase of tall buildings with outer diagrids. In our previous work [43], we already provided the results related to the filter method with a specific metric, the Pearson correlation coefficient, while here we show the alternative Spearman's rank.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Artificial Intelligence-Powered Computational Strategies in Selecting and Augmenting Data for Early Design of Tall Buildings with Outer Diagrids

Kazemi,

Ghisi,

Entezami

2024

Buildings

Self Cite

View full text Add to dashboard Cite

In the realm of architectural computing, this study explores the integration of parametric design with machine learning algorithms to advance the early design phase of tall buildings with outer diagrid systems. The success of such an endeavor relies heavily on a data-driven and artificial intelligence-enhanced workflow aimed at identifying key architectural and structural variables through a feature/response selection process within a supervised machine learning framework. By augmenting an initial dataset, which was notably limited, through four distinct techniques—namely Gaussian copula, conditional generative adversarial networks, Gaussian copula generative adversarial network, and variational autoencoder—this study demonstrates a methodical approach to data enhancement in architectural design. The results indicate a slight preference for the Gaussian copula method, attributed to its less complex hyperparameter tuning process. Evaluation through a random forest regressor revealed stable performance across various cross-validation techniques on synthetic data, although with an acceptable decrease in the coefficient of determination, from an original average score of 0.925 to an augmented score of 0.764. This investigation underscores the potential of artificial intelligence-powered computational tools to guide design decisions by pinpointing the variables with the most significant impact on relevant outputs, quantitatively assessing their influence through the accuracy of the employed machine learning methods.

show abstract

Section: Data Synthesis and Augmentation Algorithmsmentioning

confidence: 64%

Section: Response Selectionmentioning

confidence: 68%

Section: Data Synthesis and Augmentation Algorithmsmentioning

confidence: 80%