Non-formal data mining and contemporization of housing plan by artificial neural network; case studies: Tehran selected buildings during 1950s to 1970s

The aim of the research is to use and apply the artificial intelligence network and data mining of the non-form pattern in the ten valuable landmark buildings of Tehran (1330s to 1350s) in the direction of modernization.

In the present study, the research method used in terms of purpose is applied-developmental and the method of study is descriptive-survey in terms of method and nature. In this research, the MLP (Multilayer perceptron) artificial intelligence network and clustering have been used to validate the non-form analysis of residential building plans in the period 1330-1350. The data were randomly divided into three sets, 70% of the data were used for training, 15% for validation, and 15% for testing.

According to the analysis and matching with non-formal analysis, the results show that plans have 15, 14, 13 and 11 components in terms of non-form. which exactly corresponds to the plan's amorphous analytical tables. Therefore, the results of the non-form analysis of the plans have been validated by artificial intelligence.

Modernization of buildings and preservation of historical buildings are important for the majority of people and the results of this research showed that by using modern technology such as creating an artificial intelligence network, it is possible to find the invisible and hidden components in the plans of the mentioned period and use them in today's residential plans. The use of modern technologies such as artificial intelligence in order to cluster and identify the hidden relationships of plans can be very helpful.