An agent that lets you access the semantic web through the social web, acting as a bridge between the two.
Connected to the openlink-virtuoso-endpoint based on dbpedia and SPARQL:
- Link to dbpedia: https://dbpedia.org
- Link to endpoint: https://dbpedia.org/sparql
- Link to openlink-virtuoso-documentation for SPARQL-endpoints: https://docs.openlinksw.com/virtuoso/rdfsparqlprotocolendpoint/
-
Clone the Repository:
-
Install Dependencies:
From now on, in the /app-folder, run:
npm install- Configure Your Host:
Create a .env file to define your environment variables.
Schema of configuration (can also be found at app/.env.template:
MASTODON_API_KEY=your-api-key
ACCOUNT_NAME=your-account-name
URL=url-to-my-instance
USER_ANSWERING_SCHEDULE="*/3 * * * *" #schedule to answer requests from users
SE_HASHTAG="semanticweb"
SPARQL_ENDPOINT="http://localhost:3030/my-knowledge-base"
SPARQL_QUERY_SUFFIX="/query"
SPARQL_BODY_FORMAT="json"
SPARQL_QUERY_ACCEPT_HEADER="application/trig"Depending on the triple store, the .env-setup might be different, for example:
- SPARQL_ENDPOINT="https://dbpedia.org"
- SPARQL_QUERY_SUFFIX="/sparql"
- SPARQL_BODY_FORMAT="application/rdf+xml"
- SPARQL_QUERY_ACCEPT_HEADER="application/rdf+xml"
- SPARQL_ENDPOINT="http://localhost:3030/my-knowledge-base"
- SPARQL_QUERY_SUFFIX="/query"
- SPARQL_BODY_FORMAT="json"
- SPARQL_QUERY_ACCEPT_HEADER="application/trig"
That's it! Your server should be up and running!
Setup virtual environment:
bash python -m venv venv
Install requirements:
bash python -m pip install -r requirements.txt
Build with Meta Lama 3 and ollama.