bhuribhat/text-classification

2110571 Neural Network Final Project

Text Classification

Description

This web page provides a user-friendly interface for text detection and classification. Users can enter text, and with the click of a button, the language of the text is detected and displayed. If the language is English, users have the option to classify the text as positive or negative.

• The "Detect" button is used to trigger the language detection process.
• The "Classify" button is enabled if the language is English; otherwise, it remains disabled.

Dataset

• Language Detection Dataset
• Text Classification Dataset

Create Docker Container

• The webserver is available at: http://localhost:8000
• Documentation is available at: http://localhost:8000/docs

$ docker build -t text-classify-image .            # build image
$ docker run -p 8000:8000 text-classify-image      # run container

Use docker-compose:

$ docker-compose up --build    # build image and run container
$ docker-compose up -d         # start and run container
$ docker-compose down          # stop and remove container

Run Application

• The webserver is available at: http://localhost:8000
• Documentation is available at: http://localhost:8000/docs

$ uvicorn app.main:app --reload

Webserver UI

The webserver UI is available at: http://localhost:8000/detect

Documentation

Click Try it out to try custom text value in documentation:

Resource

• Language Detection
• Text Classification
• LSTM text Classification