An Accurate and Efficient Approach to Knowledge Extraction from Scientific Publications Using Structured Ontology Models, Graph Neural Networks, and Large Language Models
Timofey V. Ivanisenko, Pavel S. Demenkov, Vladimir A. Ivanisenko · 2024
Read An Accurate and Efficient Approach to Knowledge Extraction from Scientific Publications Using Structured Ontology Models, Graph Neural Networks, and Large Language Models by Timofey V. Ivanisenko, Pavel S. Demenkov, Vladimir A. Ivanisenko (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11546091.txt.
Description
Timofey Ivanisenko, Pavel Demenkov, and Vladimir Ivanisenko combined three AI tools to mine new knowledge from biomedical papers, in a 2024 study. The literature grows too fast for anyone to read, and older mining tools miss context or invent facts. Their hybrid joins text mining with graph networks and large language models.
The graph network proposes new links between proteins, and a tuned language model checks each proposal against the papers and explains it. On a set of proven protein pairs, the method scored a strong 0.772 on a balanced accuracy measure. Turned to insomnia, it found 25 links among 32 human proteins that no database held. The named finds touch mood chemistry, nerve disease, and the body clock. The authors say the method can speed the hunt for drug targets.
Catalog record
| CREATOR | Timofey V. Ivanisenko, Pavel S. Demenkov, Vladimir A. Ivanisenko |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |