PANTER: The future of transparency in science
How AI and IT are revolutionizing the open access market
Libraries and research institutions today face a huge challenge: keeping track of the constantly growing publication landscape is anything but easy. Prices, funding models and open access options seem confusing and difficult to compare. What if this complexity could be penetrated and restructured using state-of-the-art IT and artificial intelligence (AI)? This is exactly where the Publishing Analytics Tracker (PANTER) comes in. A solution that takes the open access market to a new level. On behalf of the Saxony Consortium, a cooperation of Saxon university libraries, the virtual7 has implemented a digitization project that makes it possible to analyze price developments and financing models of open access journals in a transparent and structured manner. PANTER is a publicly accessible price monitor that enables libraries and research institutions to make informed decisions and at the same time make scientific work more transparent and accessible.
Making knowledge accessible to all
One of the biggest challenges in the scientific publication market is the diversity of data sources – from HTML pages to PDFs and Excel files. But how can this diverse data be efficiently captured, transformed and harmonized? virtual7 solves this problem with an innovative combination of a Large Language Model (LLM) from OpenAI and a specially developed, highly flexible ETL pipeline (Extract, Transform, Load). In this way, even complex, unstructured data volumes are transformed into usable, easily understandable insights. The core of the solution is a data-driven architecture that focuses on maximum flexibility and adaptability. Thanks to the JSON-configurable pipeline, new data sources can be easily integrated or existing workflows modified without in-depth programming knowledge. This enables rapid adaptation to constantly changing requirements in the open access market.
AI meets data management
PANTER impressively demonstrates how modern IT technologies work together to overcome challenges that would be almost impossible to solve without these tools. The use of Docker-Compose ensures efficient containerization of processes, MongoDB Atlas enables scalable data storage, and tools such as Selenium and Tesseract OCR guarantee that even complex data sources such as tables from PDFs are processed reliably. An important element of PANTER is the use of AI: the Large Language Model (LLM) not only analyses the data, but also harmonizes and standardizes it. It extracts relevant information from a variety of data sources, some of which are very unstructured, and structures them in such a way that they can be used for scientific evaluations. This makes even complex information tangible and comparable – all at a speed that would not be possible manually.
More transparency for science
PANTER is more than just a technical tool – it is a decisive step towards a data-driven future in science. For libraries, the platform offers the opportunity to make informed decisions about license agreements and to use their budgets in a targeted manner. Research institutions in turn receive a solid basis for fair negotiations with publishers. But researchers also benefit, as they have better and more transparent publication options at their disposal. By making the results available as an open access database, PANTER creates transparency that extends far beyond the boundaries of libraries and research institutions. It is a flagship project that shows how IT and AI not only create efficiency, but also make a valuable contribution to society: promoting free access to knowledge and transparent science.
Success in record time
What has been achieved in just over two months is more than remarkable: from the initial concept to successful implementation, virtual7 has created a solution for the Saxony consortium that could revolutionize the scientific community. PANTER is not just a digitalization project, but also an example of how modern IT and AI can work hand in hand to create a more transparent, fairer scientific landscape. PANTER tells the story of data becoming insight, of AI bringing order to chaos and of a vision that shows how IT and science can work together to create a more transparent future. It is another example of how the use of the latest technology is paving the way for greater transparency and comparability in the academic publishing market – to the benefit of libraries, research institutions and society as a whole.
