preprocessing of amharic language for chatbot
# amharic_chatbot_for_aau
web_scrapper folder consit one class crawler.py
it is used to scrape data from internet using beautiful soup
data folder consists 3 files
1. intents.json => sample training data for neural network model
2. dictionary.txt => dictionary of words for spelling check
3. stopwords => list of stopwords
4. documents.txt => sample documents for training Information retrieval
pysymspell folder have symspell.py class which uses damerau levenshtien algorithm to check words distance. It is language independent.
preprocess.py
preprocessing of training data
retrieval.py
preprocessing of documents for Information retrieval