In an era of rapid advancements in Artificial Intelligence (AI), scientists are now able to publish nano-publications linked to their scholarly articles through a set of pilot projects initiated by Knowledge Pixels. These projects involve collaborations with IOS Press, specifically their Data Science journal, and Pensoft, publisher of Research Ideas and Outcomes. The aim is to make precise scientific statements easily understandable by both human users and computer algorithms, allowing information to be delivered instantly in the right format to the intended recipient.
As AI technology continues to evolve, it has become crucial to enable computer algorithms to interpret complex scientific literature and understand the underlying data, methodology, and evidence. Until AI systems can independently evaluate scholarly publications, it is our responsibility to ensure that existing knowledge is optimally accessible to these algorithms. In turn, these algorithms can assist researchers by delivering the right information at the right time in a readily usable format.
Knowledge Pixels, a startup focused on developing software and services, has designed a framework that allows scientific findings to be published in a manner that is both human-readable and machine-actionable. To achieve this, Knowledge Pixels partnered with forward-thinking scholarly publishers Pensoft and IOS Press.
Both Knowledge Pixels and the collaborating publishers share a common vision and mission grounded in the FAIR Principles, which aim to ensure that scientific outputs are Findable, Accessible, Interoperable, and Reusable. This means that FAIR publications can be easily found on the internet, accessible to both humans and algorithms, and integrated seamlessly into various outputs.
The concept of sharing scientific findings in a FAIR manner has been discussed extensively within the academic community, but practical implementation has been challenging. Philipp von Essen and Tobias Kuhn, the founders of Knowledge Pixels, highlight that despite recent advancements in AI tools like ChatGPT, current narrative-based scientific publications are imprecise and not directly interpretable by machines. This creates a need for a more structured approach.
Through pilot projects launched with IOS Press's Data Science journal and Pensoft's Research Ideas and Outcomes (RIO Journal), scientists now have the opportunity to publish nano-publications alongside their regular research articles. Unlike traditional articles, nano-publications are concise snippets of scientific findings (e.g., medication X treats disease Y) that are stored on decentralized server networks. Algorithms can fully understand and use them to deliver precise information to the appropriate users, instantly. Nano-publications remain linked to the original research paper, allowing for thorough scientific scrutiny.
Data Science and RIO Journal have already incorporated a workflow where authors can create and browse nano-publications using the dedicated Nanodash tool. The process is user-friendly, requiring minimal effort and no programming skills. Authors will soon be able to include citations for their nano-publications on their academic CVs, with this feature expected to be released later this year.
The introduction of these pilot projects represents a significant step forward in scientific communication, facilitating the integration of AI technologies into scholarly publishing. By enabling the publication of nano-publications alongside traditional research articles, scientists will be able to communicate their findings in a format that is easily understood by both humans and machines. This bridging of the gap between human and machine comprehension will have far-reaching implications for the accessibility, discoverability, and usability of scientific knowledge.
As the pilot projects progress, it is expected that more publishers and researchers will embrace this innovative approach, paving the way for a future where scientific knowledge is effectively harnessed to accelerate discoveries and advancements in various fields.
Click here to read the original press release.
