A Question Answering pipeline to provide users with richer responses.
This repository will provide the implementation of the project proposed that is funded by the Google Summer of Code program.
The final Dbpedia Chatbot [Demo].
The project Supervision and support is done by:
Two backend services of this chatbot consists of webhook and rdf_visualization. To build and run the docker images of these two services, please follow the below steps:
bash ./Webhook/build.sh
in your main directory (dbpedia-chatbot-backend).
Open your browser to the URL given on your terminal once the server is up!bash ./rdf_visualization/build.sh
in your main directory (dbpedia-chatbot-backend).
Open your browser to the URL given on your terminal once the server is up!Use cases of the future QA system:
Visualization of the current QA system:
graph TD
subgraph End-User
res(Default Response)
exp(Input Expression)
cres(Custom Response)
superres(Response) --> | | cres(Custom Response)
superres(Response) --> | | res(Default Response)
end
subgraph Fullfillment
wh(Webhook)
wh(Webhook) --> |contains| H(components.js);
wh(Webhook) --> |contains| I(app.js);
wh(Webhook) --> |contains| J(intent.js);
end
subgraph Dialogflow
A(Dialogflow-Agent) -->|contains| B(Entities);
A(Dialogflow-Agent) -->|contains| C(Intents);
A(Dialogflow-Agent) --> |check fulfillment| fc(Fulfillment Checker)
fc(Fulfillment Checker) --> |enabled| wh(Webhook)
fc(Fulfillment Checker) --> |disabled| res(Default Response)
end
subgraph RDF-Visualization
F(app.py) --> |uses| E(index.html);
I(app.js) --> |graph request| F(app.py);
end
exp(Input Expression) --> |intent matching| A(Dialogflow-Agent)
I(app.js) --> |uses| H(components.js);
I(app.js) --> |uses| J(intent.js);
I(app.js) --> |give response| cres(Custom Response)