Research Graph schema is an accessible meta-model for connecting research objects. This schema is designed to provide a practical approach to construct large scale graphs from a distributed network of scholarly works, by following these design principles:
This schema enables the rapid development of local, national, or domain-specific research graphs with a trade-off between practicality and completeness.
The National Graph project is a collaborative approach to building a national-level graph of persistent identifiers. This capability provides insights into the collaborations between Australian research institutions, industry, and international partners.
Research Graph Augment API transforms disconnected research information to a connected graph, and augment this graph with the data from the global network of scholarly works.
Exploring the collaboration network of researchers by building an interoperable graph of research collaboration networks in Australia and internationally. The aim is to visualise how researchers collaborate in different domains across universities.
Exploring the collaboration network of coronavirus research at the global level, and knowledge mining from coronavirus literature.
GESIS – Leibniz-Institute for the Social Sciences is the largest infrastructure institution for the Social Sciences in Germany. With more than 300 employees in two locations (Mannheim and Cologne) GESIS render substantial, nationally and internationally relevant research-based infrastructure services.
The National Computational Infrastructure (NCI) hosts 10+ Pbytes research data collections on it high-performance file systems, co-located with high-performance computing resources. NCI has developed a connected graph that shows the connections between research datasets, publications, researcher profiles and grants across research repositories and infrastructures such as DataCite and ORCID.
RAG: Supercharging AI with Smart Data Retrieval https://aigraph.researchgraph.org/2023/11/13/rag-supercharging-ai-with-smart-data-retrieval/
Platform to assess applications to access sensitive data. Safe person? Correct data for correct project?
https://youtu.be/BbT_3qyeDM0?si=wub7CVtHUaR_3I45
Great to see Reproducible Outputs as one of the new functions of GPT-4 #gpt #artificalintelligence
15 New Initiatives in the New White House Order on AI
https://aigraph.researchgraph.org/2023/11/06/15-initiatives-from-the-white-houses-executive-order-on-artificial-intelligence/
While they are effective in understanding and generating natural language, they have limitations such as producing incorrect responses, known as the hallucination effect. This can lead to potential risks in clinical settings, especially in areas like imaging appropriateness
Large language models (LLMs) like ChatGPT, developed by OpenAI, have shown success in various tasks due to their advanced architecture and training mechanisms.
#aigraph
Mistral AI is 187x cheaper compared to GPT-4
https://aigraph.researchgraph.org/2023/10/27/mistral-ai-is-187x-cheaper-than-gpt4/
Feature engineering refers to the process of creating new input variables from the available data. #aigraph
data preparation can be treated as another hyperparameter to tune as part of the modeling pipeline.
Transformers, introduced in 2017, are a pivotal architecture for LLMs, designed around the concept of attention, which helps process longer text sequences efficiently
#LLM #NLP