Analysis of traditional urban texture using data mining method and a model proposal based on artificial intelligence for new building desings

Underneath their seemingly complex and irregular structures, cities have geographical, climatic, socio-cultural, economic, etc. They are systems that operate and live in accordance with the rules created by various factors. This structure of the city is revealed by analyzing the relationships between the parts that make it up and finding a system of rules. In creating this rule system, tangible and intangible cultural heritage elements are used. However, over time, cultural heritage is lost and these items are often in danger of being lost. It is important in the context of urban research to be able to make more conscious and informed decisions in urban intervention processes by increasing our current knowledge of the complex structures of cities. Large-scale computing systems and new generation data collection and analysis techniques offer new opportunities in the field of urban research. Nowadays, with the development of technology, more data can be obtained every day, and methods and techniques based on data mining and artificial intelligence are used in interdisciplinary studies to discover useful information that can emerge from these data. If we think of cities as a database consisting of raw data, these raw data can be processed with the help of data mining and transformed into usable information that can be useful for cities in the context of solving urban problems. Thus, important developments in the fields of computational science and engineering will be able to help in finding solutions in urban data analysis studies. In this way, with the help of the emerging algorithms, seemingly complex city structures and rule systems will be analyzed and simplified. The main purpose of the study is to produce a computational model and software that can support decision makers and designers for new building designs in traditional urban textures that have been preserved in a holistic manner until today. Traditional urban textures need to be protected, along with their cultural values, lifestyles and all other features of the place that reflect the traces of the past. Three basic approaches stand out in order to ensure sustainability in terms of cultural heritage in these cities that have cultural heritage value. The first of these is the determination of the characteristics of the place chosen for construction, the second is the problem of how to re-evaluate the existing buildings in that place, and the third and lastly is the design of new buildings to be built within the historical protected area. These three basic approaches should not be considered independently of each other. Building designs to be realized in line with new needs require good analysis of the texture. If adequate analysis is not performed on traditional tissues, wrong decisions may be made. It is not possible for the designers who play a role in designing these structures to carry out this analysis alone. For this reason, for new building designs, it is necessary to determine the attributes of the city and to quickly analyze the relationship between the attribute values. Here, design tools that can be beneficial in terms of both time and cost are needed. With the development of today's technologies and the increase in data size, tools such as data mining and artificial intelligence save time for individuals working on these issues. Sille District of Konya province, which is an important example of a traditional city with its historical and architectural texture that has hosted different civilizations, was chosen as the application area of the study. In the field study, in accordance with the architectural conservation discipline, the attribute values of a total of 263 civil architectural buildings, which are the 2nd group cultural assets on 62 islands and 541 parcels within the historical texture, were collected in a database by preparing building information forms. Data analysis of the buildings was carried out using techniques based on data mining and artificial intelligence, and association rules that could help in decision-making for new building designs were obtained with an application prepared in Matlab. In addition, based on these association rules, alternatives were presented for the mass clearance for new building design in 34 empty parcels within the urban renewal boundaries. Thus, by using site-specific data about the city, an attempt was made to benefit from the traces of the past in the production of new buildings for the future and to contribute to the sustainability of urban memory.