Redirigiendo al acceso original de articulo en 15 segundos...
Inicio  /  Buildings  /  Vol: 13 Par: 7 (2023)  /  Artículo
ARTÍCULO
TITULO

Investigating the Effects of Parameter Tuning on Machine Learning for Occupant Behavior Analysis in Japanese Residential Buildings

Kaito Furuhashi and Takashi Nakaya    

Resumen

Global warming is currently progressing worldwide, and it is important to control greenhouse gas emissions from the perspective of adaptation and mitigation. Occupant behavior is highly individualized and must be analyzed to accurately determine a building?s energy consumption. However, most of the resident behavior models in existing studies are based on statistical methods, and their accuracy in parameter tuning has not been examined. The accuracy of heating behavior prediction has been studied using three different methods: logistic regression, support vector machine (SVM), and deep neural network (DNN). The generalization ability of the support vector machine and the deep neural network was improved by parameter tuning. The parameter tuning of the SVM showed that the values of C and gamma affected the prediction accuracy. The prediction accuracy improved by approximately 11.9%, confirming the effectiveness of parameter tuning on the SVM. The parameter tuning of the DNN showed that the values of the layer and neuron affected prediction accuracy. Although parameter tuning also improved the prediction accuracy of the DNN, the rate of increase was lower than that of the SVM.

 Artículos similares

       
 
Mohamed Galal Eltarabily, Ismail Abd-Elaty, Ahmed Elbeltagi, Martina Zelenáková and Ismail Fathy    
Climate change (CC) directly affects crops? growth stages or level of maturity, solar radiation, humidity, temperature, and wind speed, and thus crop evapotranspiration (ETc). Increased crop ETc shifts the fraction of discharge from groundwater aquifers,... ver más
Revista: Water

 
Alessandra Martines, Giulia Furfaro, Michele Solca, Maurizio Muzzi, Andrea Di Giulio and Sergio Rossi    
Microplastic pollution constitutes a serious environmental problem that requires more effective scientific research to describe its potential impacts on marine fauna. The interaction between microplastics and marine biota can have significant negative ef... ver más
Revista: Water

 
Noaman Mohammed Ridha Dabbagh, Ahmed W. Al Zand, Mohammed Chyad Liejy, Mohammad Ansari, Wadhah M. Tawfeeq, Wan Hamidon Wan Badaruzzaman, A. B. M. A. Kaish and Zaher Mundher Yaseen    
Recently, the adoption of recycled concrete instead of normal concrete as infill material in tubular stainless steel members has received great attention from researchers regarding environmental improvement. However, the flexural behavior of recycled con... ver más
Revista: Buildings

 
Lan Luo, Yue Yang, Guangdong Wu, Junwei Zheng and Defa Liu    
Organizational leadership is a key factor affecting the management performance of complex construction projects, but seldom have studies attempted to explore the effect mechanisms of organizational leadership on the project management performance, especi... ver más
Revista: Buildings

 
Ana Luiza Rodrigues, Caio Falcão and R. Christopher Williams    
Crosslinking agents, notably sulfur, are used in asphalt binder modification to facilitate chemical bonding between polymer chains and the asphalt binder. Despite some prior research indicating the benefits of sulfur crosslinking in enhancing polymer-mod... ver más
Revista: Infrastructures