RASA (NLP AI Chatbot)

Rasa is an open source machine learning framework for automated text and voice-based conversations. Understand messages, hold conversations, and connect to messaging channels and APIs.

RASA : 머신러닝 기반 챗봇. rasa.com/docs/rasa/

사용툴 : Pycharm

사용 방법:

1 goto Pycharm Terminal

2 Pip(Package Installer for Python) 로 rasa 설치

pip install rasa
exception- pip install --U pip

              pip install --U rasa

3 rasa 초기화 (hello world 로드)

rasa init
4 학습모델로 훈련.

rasa train
5 shell로 대화하기

rasa shell
기본 모듈이 로딩됐으면 glossary를 만들어서 각 모듈 분석/이해.

구조 :

#domain.yml

intents:
  - affirm
  - deny
  - greet
  
entities:
  - name
  
slots:
  concerts:
    type: list
    influence_conversation: false
    
responses:
  utter_greet:
    - text: "Hey there!"
  utter_goodbye:
    - text: "Goodbye :("
    
 actions:
  - action_search_concerts
  - action_search_venues

정의 :

session_config:

session_expiration_time: 60 # value in minutes
carry_over_slots_to_new_session: true
domain- lists input, output of assistant

: intent- user query. (defined with examples)
: entity- extracts keywords from user message
: slot - bot memory. key-value store.
: responses - bot response to intent (utter)
: actions - bot action(coded action action)
: session_config - config. for the bot

config

: pipeline- lists nlu components that define Rasa's sytem.
: component- process user input into structured output. (THE language odel)
: such as MitieNLP, SpaceyNLP,

nlu - deals with understanding human lang. parsing

policies

components that predict dialogue system's next action.
makes decisions about how conversation flow should proceed

rules

special training data to specify rule.

fallback

handling of low confidence message - either to attempt
to disambiguate user input of provide default message.

story

Training data for the dialogue model. Consis conv. between user and bot.