config.yaml

# Configurations for SciBot

# Configuaryion for the keyword extractor
language: 'en'  # Language setting for the keyword extractor.
max_ngram_size: 3 # Maximum size of n-grams for keyword extraction.
deduplication_threshold: 0.9  # Threshold for deduplication during keyword extraction.
deduplication_algo: 'seqm'  # Algorithm used for deduplication during keyword extraction.
windowSize: 1 # Size of the window for keyword extraction.
numOfKeywords: 10 # Number of keywords to extract.

# Configurations to fetch context
user_agent: 'ChatBot/1.0 (thushara.nair02@gmail.com)' # User agent information for fetching context from Wikipedia.
sentence_transformer: 'paraphrase-MiniLM-L6-v2' # Model for converting questions and articles into embeddings.

# Model for prediction 
model_name: 'deepset/roberta-base-squad2' # Pre-trained model name for prediction (RoBERTa-based for question answering).