Frequently Asked Questions

Large Language Models (LLMs), ChatGPT, and Virtuoso

query
qtext
answer
atext
https://data.openlinksw.com/oplweb/faq/virtuoso-llms/Q1#this
What is Virtuoso?
https://data.openlinksw.com/oplweb/faq/virtuoso-llms/Q1#A1
Virtuoso is a combined multi-model DBMS and HTTP-based Application Server platform that supports critical open standards for providing applications with loosely-coupled access to disparate data sources. It is cross-platform, high-performance, scalable, and secure. It also includes native reasoning and inference for identity reconciliation and data wrangling across disparate data sources.
https://data.openlinksw.com/oplweb/faq/virtuoso-llms/Q2#this
What is ChatGPT?
https://data.openlinksw.com/oplweb/faq/virtuoso-llms/Q2#A2
ChatGPT is an advanced multi-lingual natural language processing technology, facilitated by a large language model (LLM), with an uncanny understanding of sentence syntax and semantics. It can produce responses, in conversational mode, based on user inputs.
https://data.openlinksw.com/oplweb/faq/virtuoso-llms/Q3#this
How does Virtuoso integrate with Large Language Models (LLMs)?
https://data.openlinksw.com/oplweb/faq/virtuoso-llms/Q3#A3
Virtuoso uses LLMs as a natural language based conversational UI/UX into its modern multi-model data layer comprising structured data represented as tables (n-tuples) or RDF-based entity relationship graphs (3-tuples) which are also generally referred to as Knowledge Graphs. Integration occurs in loosely-coupled fashion, courtesy of Linked Data Principles where hyperlinks functioning as powerful pointers usable for creating data source names from SQL and/or SPARQL query templates for LLM fine-tuning. This capability extends to the following functionality realms: - Reasoning & Inference: Built-in and/or custom reasoning and inference that scales massively for conceptual data analysis, integration, and management - Data Virtualization: Powerful virtual database core that extends its data reach to HTTP-, ODBC-, and JDBC-accessible data source across private and/or public networks. Once attached, this data is harmonized semantically using a variety of mapping techniques based on existing open standards e.g., the RDF-based RDF to Relational Mapping Language commonly referred to as R2RML. - Data Crawling: Unique native ability to crawl data from across a public or private network as part of its query solution production pipeline in very fine-grained form i.e., to the level of query language variables and constants.
https://data.openlinksw.com/oplweb/faq/virtuoso-llms/Q4#this
What are the benefits of using Virtuoso and ChatGPT together?
https://data.openlinksw.com/oplweb/faq/virtuoso-llms/Q4#A4
Some of the benefits of using Virtuoso and ChatGPT together are: - You can access and integrate data from any source, including web pages, databases, APIs, files, etc. - You can query and explore data using natural language in a manner that's transparently mapped to declarative languages such as SQL and/or SPARQL - You can generate natural language responses that are informative, relevant, and engaging - You can create content such as summaries, reports, articles, etc. based on data - You can crawl a web of documents and/or a semantic web of data en route to discovering new data sources and entity relationship types - You can learn from ChatGPT's suggestions and insights to improve your productivity and knowledge, at levels unrivaled to date.