Methods
1. Expert Interviews and Literature Analysis
At the beginning of the project, we placed great emphasis on expert interviews and thorough analysis of the literature. Our goal was to deeply understand the current situation and identify best practices. During the expert interviews, we engaged the most outstanding professionals to shape the direction of the project based on their experience and knowledge. The literary analysis enabled us to work based on the latest research and innovations, thus ensuring the scientific foundation of the project.
2. Collection of Best Practices
It was important for us to incorporate the best practices from other similar projects into our own development processes. To this end, we collected and analysed data to identify and apply the most effective methods. This practice allowed us to avoid previously encountered mistakes and use the most efficient solutions. In the future, we will regularly review and update the collected best practices to ensure that we always have up-to-date knowledge.
3. Pilot Developments
To test and fine-tune the developments, we launched pilot projects. These experimental projects allowed us to try out our innovations in real-world environments and integrate the experiences gained into the final solutions. During the pilot projects, we established close cooperation with the involved institutions and users to receive direct feedback. This helped ensure that our developments truly met user needs and expectations.
The goal of the Jók.ai lab is to enrich and make Hungarian culture accessible in the digital world with the help of artificial intelligence. To achieve this, we applied a comprehensive and well-organized methodology that ensured our developments were accurate, relevant and easy to use. Learn about our methodology, which is built on the following key elements:
4. Questionnaire Surveys
In order to understand user needs and expectations, we conducted questionnaire surveys. The data obtained were used to optimize our services to truly meet the audience's expectations. When designing the questionnaires, it was important to collect the most accurate and relevant information possible. The results were analysed using statistical methods, and the data obtained were integrated into our development processes.
5. Development and Testing of the Text2SQL Module
We paid particular attention to the development and testing of the Text2SQL module. The goal of this module is to convert natural language questions into SQL queries, thereby facilitating access to cultural databases. The development and testing of the module were carried out through manual prompt creation and human verification. During the development process, we continuously fine-tuned the algorithms to ensure they provide the most accurate and efficient responses to user queries.
6. Database Integration and API Development
To ensure the seamless flow of data, we implemented integration and API development processes. The primary goal was to facilitate smooth and efficient communication between various databases and systems. During API development, we paid special attention to data security and system reliability. The integrations enabled the linking of different data sources, resulting in a richer and more comprehensive database.
7. Continuous Development and Feedback
The methodology of the Jók.ai lab does not stop at initial developments. We continuously monitor user feedback and new technological advancements to provide the most up-to-date solutions. Feedback is regularly evaluated and incorporated into further developments, ensuring the project's continuous evolution and up-to-dateness.
Our methodology is based on detailed planning, continuous testing, and consideration of user feedback. This ensures that the developments of the Jók.ai lab are outstanding not only from a technological perspective but also in terms of user experience. We are proud that our work contributes to the digital preservation and wide accessibility of Hungarian cultural heritage.