Hi, in this gig I will build you a complete End
to End Data Pipeline with Model deployment on Cloud related to Sentiment
Analysis. Sentiment Analysis includes Sentiments from Twitter users, trending topics on Twitter, Product Reviews, Text (Spam/Ham), and Fake news Classifiers just to name a few. The procedure, Tools, and Framework that I use for sentiment analysis are as follows:
 
Language: Python
 
Python IDE: jupyter Notebook, Pycharm, Google Colab
 
Python Packages: NLTK, Spacy, Gensim, Pytorch, Tensorflow, Torchtext, TextBlob,
Sklearn, Hugging Face
 
Text Pre Processing: Tokenization, Stop words, Stemming,
Lemmatization, removal of garbage values or http/s links
 
Vectorization: Bag of words, Continuous Bag of words, TFIDF, Word Embeddings,
Word2Vec, Avg Word2vec
 
ML Algorithms: XGBOOST, Random forest, Support Vector Machine, KNN, Naive
Bayes classifier, Logistic Regression, etc.
 
Advance DL Algorithms: RNN, LSTM, Bi-directional RNN,
Encoders and Decoders.
 
Transformers Model: Auto-encoding Models (BERT, RoBERTa, and others),
Auto-regressive Models (GPT, GPT-2 and others).