Artificial intelligence is already influencing the formation of a new culture. The more you keep data secret and do not share it, the more you limit the development opportunities. All major companies and countries have already begun to adopt this open and sharing culture. Otherwise, failure is just around the corner.
The fact that the results of research conducted in the world of science are simply published in scientific journals, and that data is not reused and cannot be accessed is an issue that has drawn my attention and even worried me for a long time. Valuable scientific research, on which billions of dollars and thousands of hours have been spent, is simply lost as inaccessible data. It really saddens me that such valuable knowledge remains unused, and concrete steps must be taken to change this.
Scientist Sean Hill says that if we really want science to develop further, we need to prepare the research data in a form that is easy to find and can be used by artificial intelligence. He states that the main obstacle is precisely the difficulty of finding, accessing and reusing data.
To solve this problem, Sean Hill has launched a new initiative called Senscience. The initiative uses Frontiers’ new FAIR² Data Management service and aims to make scientific data AI-ready. The service structures data in a way that AI can easily use and makes it openly accessible. As a result, this data becomes more useful for both scientists and companies operating in areas such as drug production and the development of new materials.
According to Hill, progress in science will be very difficult as long as scientists cannot use the data they already have in future research. Therefore, it is crucial that data is AI-friendly, easily accessible and open to use.
Improving the Quality of Science
“Senscience” is an artificial intelligence product created by Frontiers, a scientific publisher based in Switzerland. Its goal is to facilitate collaboration between different scientists and academia and improve the quality of scientific research by supporting open science. The other founder, Henry Mark, is a renowned scientist who is trying to understand how the brain works. He is the founder of the Blue Brain Project, which is famous for creating digital models of the brain, and the Human Brain Project, which is supported by the European Union. He has more than 450 scientific articles and these articles have been cited about 55 thousand times.
The main challenge is to organize various scientific data in a form that can be easily understood and used by both humans and artificial intelligence. According to him, scientific data is more complex than sharing text and numbers because all the details are important.
Hill tried various database solutions for about a decade, but the result was not successful. Therefore, he led the creation of the Blue Brain Nexus platform, which can manage data located in different places and record even the smallest details of each scientific data. The platform is based on knowledge graph technology. Knowledge graphs were not widely used before, but are now widely used to combine data from different sources (structured or unstructured) and different formats.
FAIR² Data Sharing
For many years, the FAIR principles (discoverable, accessible, adaptable, and reusable) have been the guiding principles for sharing scientific data. However, as machine learning and artificial intelligence become more widely used in scientific research, the need for data to be compatible with these technologies has become more important.
Senscience has developed a new specification called FAIR², which takes the FAIR principles and further develops them. This specification transforms data into a format that can be used by artificial intelligence. It automates the discovery, retrieval, matching, and reuse of data. The FAIR² methodology automates the organization of data, simplifies its use, and helps manage it through an AI-powered “data steward.”
The FAIR² data format (fair2.ai) is compatible with MLCommons Croissant. This format combines the description, structure, and metadata of data to make it machine learning-ready.
Senscience's solution enables scientists to transform their research data into FAIR² Data Packages, create interactive data portals, and publish articles in Frontiers for scientific data sharing. This increases the visibility and usability of scientists' data, creating new opportunities for collaboration.
With the Senscience system, scientists can easily transform their research into a usable form. To do this, they simply
manuscripts, spreadsheets and other documents to the platform. The system creates an interactive information portal from this data and answers users' questions about the data through artificial intelligence.
Our country also has favorable conditions and potential for implementing such innovative approaches. For example, it is possible to combine various scientific data collected in universities and state institutions into a single open data platform (open data portal). The fact that this data is open and compatible with artificial intelligence will facilitate the work of researchers and increase the scientific potential of our country.
According to Hill, progress in science will be very difficult as long as scientists cannot use the information already obtained in future research. Therefore, it is of great importance that the data is compatible with artificial intelligence, easily accessible and open to use. The development of such an existing open data portal in our country will significantly boost the development of our science and the creation of innovative cooperation opportunities in various fields.
