A storyteller that uses deep learning as pre and post processing and different programming paradigms to think.
Reference code for work: "This paper describes an intelligent conversational agent model that uses different programming paradigms combined for processing over a semantic representation graph model of the natural language named Abstract Meaning Representation (AMR) performing various Natural Language Understanding (NLU) tasks. Knowledge graphs (KG) are also used to increase the agent's knowledge of semantically related entities and allow for the expansion of semantic relationships. An AMR graph manipulation layer was developed using algorithms and methods in the imperative paradigm, in addition to several predicates in the logical paradigm that serves to extract specific information from such graphs. Several parameterizable AMR templates were also defined that serve to assemble information and use it to provide feedback to the user. Both this layer and the templates make use of the KG. Finally, above this AMR graph manipulation layer and these templates, a conversational agent is built using an agent model named Belief–Desire–Intention (BDI), structured in a programming paradigm named Notification Oriented Paradigm (NOP), in a combination of both named BDI-NOP. In the NOP paradigm, several collaborative entities interact with each other through notifications. This paradigm with the BDI-NOP combination proves to be interesting for the development of the model."
Install the library for python called amrlib and its dependencies via pip.
Download the "graph to sentence" trained model and place it inside the "gtos"
folder; download the "sentence to graph" trained model and place it in the
"stog" folder. Run the python server python3 server.py and the application
deno run --allow-all --unstable app.ts.
Author: Henrique Emanoel Viana, a Brazilian computer scientist, enthusiast of web technologies, cel: +55 (41) 99999-4664. URL: https://sites.google.com/view/henriqueviana
Improvements and suggestions are welcome!