The LangChain AI support for graph data is incredibly exciting, though it is currently somewhat rudimentary. It provides us the ability to transform knowledge into semantic triples and use them for downstream LLM tasks. To unlock its full potential, I believe we still need the ability to integrate with external knowledge graphs.
This article proposes a workflow for information retrieval from a knowledge graph, combining ChatGPT and other tools, to improve the factuality of ChatGPT in domain-specific tasks. The key objective is to test ontology-guided generation/refinement of SPARQL queries by ChatGPT.
GPT Index is a Python library that facilitates the use of external knowledge for GPT prompting. In my preliminary exploration, it has proven to be effective in connecting GPT with external knowledge for Q&A and summarization tasks. I am intrigued to see how it may integrate with knowledge graphs for large-scale information retrieval for GPT.
Data is the backbone of R&D operations and data quality is undoubtedly important to R&D organizations. This article is a brief discussion on data quality in R&D, including why it is important, what are the key factors and how to measure data quality.
It's the time of the year again to set up your annual performance goals. Are you having trouble crafting your goal statements? Let's use ChatGPT to make 2023 goal setting faster and SMARTer.
