This app can predict sentiment from kaggle Dataset Sentiment140 tweets or for any tweets.

This was a kaggle competition held several years ago. See Sentiment140_dataset for much more detailed documentation.

It is in fact a python Unidirectional LSTM Word2Vec trained on raw text model.h5 with a tokenizer.pkl.

You need to install first some python librairies:

import re, os, time
from flask import Flask, request, jsonify, send_file
import tensorflow as tf
import numpy as np
import nltk
nltk.download('punkt')
nltk.download('stopwords')
nltk.download('wordnet')
nltk.download('snowball_data')
from nltk.corpus import stopwords
from nltk.stem import SnowballStemmer
from nltk.stem import WordNetLemmatizer

from tensorflow.keras.preprocessing.text import Tokenizer
from tensorflow.keras.preprocessing.sequence import pad_sequences
from tensorflow.keras.models import load_model
import pickle

As you know now, our main objective is to predict tweets sentiment although our model was only trained with the Sentiment140 dataset.

• 0 for a negative sentiment
• 1 for a positive sentiment

Our function predict_sentiment gives us a probability score.

if probability_score < 0.5:
        sentiment = "negative"
    else:
        sentiment = "positive"

I strongly recommand you to install packages from the requirements.txt.

This section was created to test our app (unit tests) with pytest.