This our final project in BMSTU.
- Smyslov Maxim (Back-end)
- Bogdanova Valeria (Front-end)
This app works as a web service, which parses marketplaces, then classifies comments as negative or positive and finaly sends the results in Telegram bot
Telebot
Flask
Aiohttp
TF-IDF
as text transformerXGBoost
as tonality classifiermongoengine
as ORMMongoDB
on parser sidePostgreSQL
on bot side
For classification was taken a dataset of human labeled comments from marketplaces. This dataset contains comments from cat food products and each comment is marked as positive or negative.
Precicion = 0.9260379462934971
Recall = 0.9390572924932551
Roc auc = 0.9466007732268533
- Fill in the credentials in the
.env
file (there is an example) docker-compose build
docker-compose up
- Enjoy