Solve your natural language processing problems with smart deep neural networks

This notebook contains entire text preprocessing pipeline for NLP problems. The ready-to-use functions require NLTK and SKlearn package installations. It also contains some prominent text classification models.

NLP starter kit

A basic machine learning model built in python jupyter notebook to classify whether a set of tweets into two categories: racist/sexist non-racist/sexist.

Predicting Political Ideology of Twitter Users.

Turkcell&Miuul Data Science Bootcamp - Assignments

Progetto Text Mining and Search

Fuzzy Matcher utility provides you robust fuzzy matching based on Levenstein distance enabled with caching and parallization

SpamGuard is an intelligent SMS filtering system designed to detect and filter spam messages using machine learning techniques.

The system is implemented to scrape data from a booking website, perform Emotion Analysis on the reviews of the selected hotel and visualized the result over a time axis. R is used to implement the system and Shiny library is used to develop the Front-end.

This notebook contains entire text preprocessing pipeline for NLP problems. The ready-to-use functions require NLTK and SKlearn package installations. It also contains some prominent text classification models.

Proyek ini bertujuan untuk mengembangkan model yang dapat menentukan apakah produk dalam review direkomendasikan atau tidak berdasarkan teks ulasan yang diberikan oleh para pengulas.

NLP LSTM model to predict python codes (Text prediction) (Tokenized special characters)

For the text Mining course I carried out a project related to the analysis and classification of the reviews of the "UCI ML Drug Review" dataset (link: https://archive.ics.uci.edu/ml/datasets/Drug+Review+Dataset+%28Drugs.com%29). I learned to apply techniques such as bag of words, TF-IDF and build sentiment analysis models through the Bert and V…

NLP using NLTK python library

Code in R to classify the news articles depending on whether their content is about financial fraud or complementary subjects

Performed PySpark based text pre-processing including lemmatization, POS tagging and UDF functions on customer feedback. Computed and visualized sentiment score to identify areas of improvements.

The comparison between different embeddings (TF-IDF, USE, and TF-IDF + USE) and various classifiers provides valuable insights into the performance of different techniques for sentiment classification.