TextAnalyser

Natural Language Processing (NLP) components for analysing and extracting information from text.

Installation

conda create -n textanalyser python=3.6
conda activate textanalyser
python setup.py install

Supported NLP components

• Topic Modeling (NMF, LDA and LSI)
• Spam Classifier
• Sentiment Classifier

Tutorials

Tutorials can be found in examples

How to train

✉️ Spam Classifier

spam_classifier = SpamClassifier()
pipeline = spam_classifier.train(X_train, y_train)
spam_classifier.evaluate(pipeline, X_test, y_test)
spam_classifier.save_pipeline(pipeline,
                              model_path='./models/model.pkl')

📰 Topic Modeling

nmf_model = TopicModeling()
nmf_model.train_model(train_docs, model_type='nmf')
nmf_model.save_model(model_name='../models/nmf_model.pkl',
                     vectorizer_name='../models/nmf_vectorizer.pkl')

💫 Updates:

• 30/06/2019: Added Sentiment Classifier
• 30/06/2019: Added Email Spam Classifier
• 29/06/2019: Added Topic Modeling (NMF, LDA and LSI)