The research focused on "neuroarchitecture," which relies on neural networks in the design process, allowing for the exploration of multiple possibilities to generate new designs. Through the analysis of previous studies, a knowledge gap was identified regarding the need to organize information on how to apply neural networks in architectural design, particularly at the local level. The goal of the research was to provide clear and organized knowledge on how to use these tools to facilitate their application by architectural designers.
The research followed two approaches: the first was descriptive, analyzing previous studies to identify the main aspects of neuroarchitecture; the second was experimental, where artificial neural networks and AI tools were applied to a local dataset that included the designs and texts of architect Rifat Al-Jaderji. The results showed that each artificial neural network and tool had a different approach to handling data, and the capabilities of these tools greatly depend on the quality and quantity of the data used.
Regarding design generation from texts, the research found that the nature of the input texts significantly affects the outputs. Abstract texts generate creative designs, while clear and directed texts help produce more realistic designs. Based on these results, recommendations and future visions were presented to expand the use of artificial intelligence in architecture, enhancing innovation and sustainability in the local architectural sector.
The discussion committee consisted of esteemed professors:
- Asst. Prof. Dr. Zainab Hussein Raouf … Chair
- Asst. Prof. Dr. Mustafa Walid Hashim … Member
- Dr. Nagham Ahmed Jassem … Member
- Asst. Prof. Dr. Anwar Sobhi Ramadan … Member and Supervisor
- Asst. Prof. Dr. Zuhair Abdul-Azeem Nassar … Member and Supervisor
We wish the researcher and all our graduate students success and prosperity.
