جستجوی مقالات مرتبط با کلیدواژه "فرایند گوسی" در نشریات گروه "جغرافیا"

Background and goal: 

Optimizing and reducing energy consumption is a noble goal in the direction of reducing environmental risks and sustainable urban development. Therefore, in this research, the role of urban regeneration in reducing energy consumption in the residential use of District 10 of Tehran Municipality has been discussed.

The present research is developmental-applicative in terms of its purpose and descriptive-analytical in terms of its nature and method. This research consists of a compilation of learning algorithms that has been implemented in a quantitative way. The statistical population is all residential buildings in District 10 of Tehran Municipality. The evaluation indicators include the amount of water, electricity and gas consumption of households in addition to 35 other indicators which are placed in four economic, social, cultural and physical dimensions. To analyze the data, Gaussian process with random forest kernel and multi-layer Presteron method have been used. Also, Arc GIS software has been used for spatial analysis of building units and Excel software for calculations and data classification.

The findings indicate that according to the correlation coefficient of 0.42 and the root mean square error of 0.24, the average absolute error of 0.19, the test section for the water output parameter has the best performance in the Gaussian process model and In the electricity parameter test section, the correlation coefficient is 0.99, the root mean square error is 0.001, and the average absolute error is 0.001 in the above model, it has the best performance among the modeling, and also in the gas consumption test section, the correlation coefficient is 0.47 , root mean square error of 0.39 and absolute error of 0.33 among the models, the above model has the best performance. The results of this research showed that the values of this percentage index are acceptable for determining the best model due to the higher correlation coefficient and the root mean square error and the lower mean absolute error for the sensitivity analysis of independent parameters.