b8b_virtuoso

This repository is part of the Beat Byte Bot Ecosystem (B8B).

B8B Result Analysis: https://github.com/gilpanal/b8b_result_analysis
B8B Virtuoso: https://github.com/gilpanal/b8b_virtuoso
B8B Synapse: https://github.com/gilpanal/b8b_synapse

Requirements:

Node.js (v16)
Telegram Bot Token: https://core.telegram.org/bots#6-botfather
Virtuoso Universal Server (https://shop.openlinksw.com/onboarding/) or Virtuoso Docker: https://hub.docker.com/r/tenforce/virtuoso
B8B Synapse: https://github.com/gilpanal/b8b_synapse/#how-to-run-and-test-it-locally (Mandatory)
B8B Result Analysis: https://github.com/gilpanal/b8b_result_analysis#how-to-run-it-locally. (Optional: see Note 1)

How to run and test it locally:

git clone https://github.com/gilpanal/b8b_virtuoso.git
cd b8b_virtuoso
npm i
Rename config/env_vars_template.js to config/env_vars.js and include the Bot Token
Start the main server by typing npm start
In Telegram, create a new group
Add the bot you created (via BotFather) as an admin to the chat
Attach an audio file (MP3 or WAV) less than 20 MB size

NOTES:

1.- If you want to avoid running B8B Result Analysis to validate results just comment line await storeResultsExperiment1(message.audio,predictionInfo) at virtuoso_helper.js

2.- By default it's assumed that Virtuoso instance is running at http://127.0.0.1:8890/sparql. You can point a different host changing the address at bot.js

3.- In config/env_vars.js file there's a variable named MODE that is meant for deploying to different environments, PROD, STAGE or DEV. For local development is enough with the following configuration:

  const MODE = 'DEV' // DEV, STAGE, PROD
  const ENVIRONMENTS = {
      DEV : {
          TELEGRAM_TOKEN: '<BOT_TOKEN>',
          DATABASE_URL: ''
      },

4.- Two different ontologies are included in the folder ontology. One is generated via Protégé, entitled beat_byte_bot.owland the other, b8bontology.ttl, is meant to be used with the project Ontology Visualization: usc-isi-i2/ontology-visualization#18. More info: https://www.w3.org/2018/09/rdf-data-viz/. See picture below.

Some SPARQL queries:

Default Data Set Name (Graph IRI): http://beatbytebot.web.app/resource/

Query Text: PREFIX b8b: <http://beatbytebot.web.app/ontology/>

...

A. Get all items in the graph

SELECT * WHERE { ?s ?p ?o . }

B. Get the objects from all existing s-p-o triples

SELECT DISTINCT ?o WHERE { ?s ?p ?o . }

C. Get all classes being used


SELECT DISTINCT ?class WHERE {[] a ?class}

D. Get all AudioProjects

SELECT DISTINCT ?audioprojects WHERE  {?audioprojects rdf:type <http://beatbytebot.web.app/ontology/AudioProject>.}

E. Get all AudioTracks

SELECT DISTINCT ?audiotracks WHERE {?audiotracks rdf:type <http://beatbytebot.web.app/ontology/AudioTrack>.}

F. Get all props from a class

SELECT DISTINCT ?property ?object WHERE {<http://beatbytebot.web.app/ontology/##ID##> ?property ?object.}

gilpanal/b8b_virtuoso