DKG Copilot
AI-powered copilot designed to retrieve, analyze, and manage information about scientific papers utilizing OriginTrail Decentrailized Knowledge Graph (DKG).
DKG Copilot is an AI-powered copilot used to retrieve, analyze, and manage information about scientific papers by utilizing OriginTrail Decentrailized Knowledge Graph. It is designed to handle various analytical tasks related to scientific papers in fields such as Computer Science, Mathematics, Medicine, and more. The tool integrates with SPARQL queries and plugins_prototype to assist users in accessing relevant data efficiently.
How to use
To leverage the DKG Copilot effectively, follow these steps:
- Access the tool with the provided welcome message to initiate queries.
- Pose questions related to scientific papers, such as retrieving paper details and performing similarity searches.
- Interact with the copilot using SPARQL queries for more in-depth analysis.
- Utilize the plugin_prototype feature for additional functionalities.
- Engage with the copilot to predict citation counts and analyze research fields.
Features
- AI-powered copilot for managing scientific paper information
- Utilizes OriginTrail Decentrailized Knowledge Graph for data retrieval
- Supports querying papers from various fields like Computer Science and Medicine
- Integrates with SPARQL queries and plugins_prototype for enhanced functionalities
Updates
2024/02/07
Language
English (English)
Welcome message
Hello! I'm SPARQL Ninja, ready to translate your questions into SPARQL queries. How can I assist you today?
Prompt starters
- How many papers exist in the @dkg corpus?
- Can you retrieve the top 10 papers (including abstracts) from Computer Science field ordered by publication date? Then, based on retrieved abstracts, perform a similarity search with KMeans algorithm (k=3). You can shorten abstracts to 128 characters each to avoid timeouts.
- What are the top 10 research fields from the corpus based on citation count?
- Retrieve citation count per year and research field, for the past 20 years for Mathematics, Medicine, and Computer Science. Then, predict citation count per year and research field for the next 3 years using linear regression.
Tools
- plugins_prototype
Tags
public
reportable
uses_function_calls