semantic analysis using word2vector, doc2vector,lstm and other method. mainly for text similarity analysis.

word2vec++ is a Distributed Representations of Words (word2vec) library and tools implementation, written in C++11 from the scratch

Assessing Source Code Semantic Similarity with Unsupervised Learning

A Python implementation of a binary text classifier using Doc2Vec and SVM

Building, Training and Testing Doc2Vec and Word2Vec (Skip-Gram) Model Using Gensim Library for Recommending Arabic Text.

Classification model for poetic genres and a recommendation system

An NLP project that compares different approaches to document representation and classification. The techniques used include Topic-modeling, Tf-Idf, doc2vec, SVM, and CNN

Classe responsável por simplificar o processo de criação de um modelo Doc2Vec (gensim) com facilitadores para geração de um vocab personalizado e com a geração de arquivos de curadoria. Dicas usando elasticsearch e singlestore.

Document Similarity Measurement Using Doc2Vec and Gensim Library

Representação de letras de músicas em vetores para análises semânticas e sintáticas.

It uses extensive NLP and ML techniques at its backend for answering descriptive type queries ( currently, the model has been trained only for Data science and Machine Learning Domain ).

산업안전보건법 법규시각화, 텍스트마이닝을 통한 법들간의 유사도 네트워크화

Highlight Stephen Chow's famous movies using bullet-screen comments, document vector and neural network.

Text classifier for document embeddings

Homework and projects done in the course CSC591 Algorithms for Data-Guided Business Intelligence (ADBI)

A rudimentary Implementation of a simple neural ranking model. Based on word embeddings (Glove, Fasttext). The pre-trained model's inherent vector cosine similarity is the main metric for consideration.

Python projects related to the course- Algorithms for Data Guided Business Intelligence

Argument retrieval model for [Touché @ CLEF 2020] - 1st Shared Task on Argument Retrieval

POStURE - A Framework for Analyzing Political Speeches Evolution and Similarities with Learned Linguistic Representations

This project aims to solve the recruitment process of Unite UW (on campus cultural organization at the University of Washington) by developing a custom ranking function using sentiment analysis and mathematical modelling.